Anda di halaman 1dari 281

ADULT ACQUIRED FLATFOOT DEFORMITY

10837515/01 $15.00  .00

BIOMECHANICS OF
PROCEDURES USED IN ADULT
FLATFOOT DEFORMITY
Anne P. McCormack, MD, Randal P. Ching, PhD,
and Bruce J. Sangeorzan, MD

Flatfoot, or pes planus, is a generalized term used to describe a


fallen medial longitudinal arch whether it is symptomatic or not. In
reality, atfoot deformity is a complex condition involving forefoot,
midfoot, and hindfoot changes.40 Flatfoot is a description of a common
end point for any process that relaxes the stability of the bone arch of
the foot. Flatfoot may be developmentalexpressed in childhoodor
acquiredexpressed during adulthood.28 Most often, atfoot is a normal
nding. There are many processes that cause symptomatic atfoot. Following are some of the more common causes within these categories:
Causes of developmental atfoot: symptomatic or asymptomatic exible atfoot, rigid atfoot associated with tarsal coalition, atfoot
associated with an accessory navicular bone, residuals of congenital
vertical talus or clubfoot, and generalized ligamentous or soft tissue
laxity (Ehlers-Danlos, Marfans)
Causes of acquired atfoot: posterior tibial tendon (PTT) dysfunction
(rupture, attenuation, or laceration), midfoot laxity, external rotation
of hindfoot, tight heel cord, abduction of the forefoot, subluxation of
the talus, trauma (Lisfranc, talonavicular joint, or calcaneus injuries),
arthroses, Charcots foot, and neuromuscular imbalance (cerebral
palsy, polio)
A atfoot deformity can occur secondary to fairly obvious causes
From the Department of Orthopaedics, University of Washington, Seattle, Washington

FOOT AND ANKLE CLINICS


VOLUME 6 NUMBER 1 MARCH 2001

15

16

McCORMACK et al

(i.e., Charcots changes or clubfoot) or more subtle and less denable


entities (i.e., midfoot laxity or tight heel cord). Complicating the situation
further, it is likely that the cause of an acquired atfoot deformity in an
adult is multifactorial. This likelihood makes the denition, diagnosis,
and appropriate treatment of this condition a daunting task.8, 40
MECHANISM OF FLATFOOT DEFORMITY
If one examines only a subset of adult acquired atfoot deformity
and deletes causes such as trauma, diabetic Charcots foot, and neuromuscular causes, one is left with a group of patients who present with
a seemingly insidious onset of pain, loss of arch height, and increased
(although the term is used widely, pronation is not actually present)
abduction of the forefoot and valgus of the hindfoot.3, 4, 15 In advanced
cases, lateral impingement and subtalar joint subluxation often are
noted.1 Posterior tibial tendinitis, attenuation, or disruption is common
(Fig. 1).

Figure 1. A, The right foot of a 68-year-old woman with unilateral symptomatic atfoot
deformity. The patient had both midfoot sag and a step off at the rst tarsometatarsal joint.
B, Weightbearing radiograph reconstruction that included a lateral column lengthening at
the calcaneocuboid joint, and a medial column stabilization procedure. Postoperatively, the
patient did very well and had no problems with the opposite foot.

BIOMECHANICS OF PROCEDURES USED IN ADULT FLATFOOT DEFORMITY

17

Early investigations focused on PTT dysfunction as the key to


atfoot deformity.19, 20, 26, 27 Over time, however, it has been observed that
the PTT is only part of the problem and may be more the harbinger of
end-stage changes rather than the initiating event.12, 34 Although the
sequence of events that leads to a atfoot deformity is not well understood, it is likely that many forces combine to exceed the physiologic
limit of the static and dynamic restraints of the midfoot. One suggestion
is that a tight heel cord increases the levering upward on the calcaneus,
increasing the tension on the plantar fascia and plantar ligaments. If
there is preexisting midfoot laxity, this might cause an attenuation and
possible rupture of the plantar and medial structures about the medial
longitudinal arch16 (ST Hansen, Jr, personal communication, 1999).
Some studies have sought to identify morphologic predispositions
that might lead to increased risk of atfoot, such as changes in talar
morphology or a preexisting exible atfoot.2, 12 One of the difculties
in trying to sort out this problem in the laboratory is creating a reliable
and reproducible atfoot model that simulates adequately the changes
in an adult atfoot. Several studies have sought to design corrections
for atfoot deformity, but none have been able to reproduce the distinct
conditions seen in a progressive adult atfoot.35, 41, 42 This fact is a
signicant limitation in designing an appropriate corrective procedure
and one that merits further investigation.
BIOMECHANICS OF FLATFOOT DEFORMITY
Many in vitro studies have investigated the biomechanics of atfoot
deformity. Most of these studies were directed toward evaluating the
effectiveness of various surgical interventions.9, 10, 14, 17, 21, 31, 35, 41, 42 Additional studies have been performed that examined more fundamental
biomechanical issues related to atfoot. Huang et al18 examined the
contribution of various structures, including the plantar fascia, plantar
ligaments, and spring ligament, in stabilizing the arch using a sequential
sectioning protocol. Angular relationships between rst metatarsal, navicular bone, and talus were recorded by Thordarson et al43 to study the
effect of individually activated leg muscles on the dynamic support of
the arch during stance phase. Reeck et al37 investigated the magnitude
of force transmission (through joint pressure measurements) to the talus
through its inferior articulations during simulated heel-strike, stance,
and heel rise.
The creation of a clinically relevant atfoot model is an important
aspect of any biomechanical study investigating the mechanism or treatment of atfoot. Many different models have been presented in the
literature.9, 18, 35, 42 In attempts to produce an acquired atfoot model, the
authors attenuated the spring ligament and applied cyclic axial loading
to the specimen.30, 34, 37 Thordarson et al42 created their atfoot model by
releasing the spring ligament, a portion of the talonavicular joint capsule,
the anteromedial aspect of the subtalar joint capsule, and plantar fascia.
Deland et al9 performed a much more extensive soft tissue release,

18

McCORMACK et al

resulting in severe deformities ranging from 10 to 105 in the sagittal


plane. Although simply dividing the spring ligament would not create
a signicant atfoot deformity, the subsequent cyclic axial loading of the
foot would result in a more physiologic atfoot surrogate for in vitro
experiments. A validated atfoot deformity model that could be used to
investigate various intervention strategies would be a welcome addition
to the foot research community (Fig. 2).
Previous studies have investigated hindfoot kinematics.7, 10, 2325, 29, 38,
39, 43, 45
Only Deland et al10 and Thordarson et al43 examined the effect
of the PTT muscletendon unit on three-dimensional changes in foot

Figure 2. A, An intact cadaver foot showing normal midfoot alignment under a 100 Newton
load. B, A atfoot deformity was created in the laboratory after resection of the spring
ligament and cyclic loading.

BIOMECHANICS OF PROCEDURES USED IN ADULT FLATFOOT DEFORMITY

19

kinematics, however. In the study by Deland et al,10 talus, calcaneus,


cuboid, and navicular motion produced by isolated PTT loading was
recorded as well as talonavicular and subtalar joint range of motion.
Thordarson et al43 measured the angular relationships between the rst
metatarsal, navicular, and talus while applying an isolated load to the
PTT.
The authors investigated the functional role of the PTT in acquired
atfoot mechanics.34 The kinematic orientation of the 4-bone hindfoot
complex was compared with and without PTT function using a cadaver
foot model. It was shown that directional changes in the hindfoot bones
associated with releasing the PTT load were largely consistent with
changes observed in acquired atfoot deformity, although the magnitudes of the changes were much smaller. This nding supports the
theory of pathogenesis by gradual attrition of the ligamentous structures
caused by ambulation on an unlocked valgus hindfoot. The authors
examined the effect of restoring PTT function in a atfoot deformity
model. The ndings indicate that restoring PTT function had little effect
in overcoming the soft tissue laxity of the atfoot condition. These
ndings suggest that early realignment may prevent atfoot deformity
development. Restoration of PTT function alone would not correct fully
a radiographically documented atfoot deformity. These results suggest
that the PTT has the greatest inuence on hindfoot kinematics during
heel rise, the interval of stance between foot at and toe-off.

TREATMENT OF FLATFOOT DEFORMITY


Appropriate treatment for patients with atfoot deformity can become as complex as understanding the cause itself. Depending on the
age and activity level of the patient as well as the severity of the
deformity, many options can be presented. A study5 of nonoperative
management in elderly patients with a sedentary lifestyle found that the
use of a well-molded custom orthosis or a UCBL (University of California Biomechanics Laboratory) shoe insert was a satisfactory treatment in
67% of the participants. Although the patients continued to have symptoms related to their atfoot condition, they preferred this treatment to
surgical intervention.
Other investigators22, 32 have documented the efcacy of orthoses in
decreasing ground reaction forces, medial shear, and strain on plantar
structures. In selected cases, these devices may be useful in allowing a
patient with mild to moderate symptoms and a exible deformity to
function well.
Surgery is indicated when there is progressive deformity or recalcitrant pain. Over the years, many surgical interventions have been undertaken. Debridement and augmentation of the PTT with the exor hallucis
longus or exor digitorum27 was tried early on with only limited success

20

McCORMACK et al

because it did not address the concurrent bone and ligamentous deformities.
The modied Hoke-Miller procedure11 applied the concept of combining soft tissue and bone correction to address the deformity about
the navicular bone. This procedure did not consider the attenuation
changes in the plantar ligaments or potential subluxation in the subtalar
joint region. The authors thought that this surgery was suited best for
atfoot with naviculocuneiform sag without abduction of the forefoot.
The use of selective hindfoot arthrodeses has been evaluated35, 38
using an experimental atfoot model that sectioned the spring ligament,
medial capsule, plantar fascia, and interosseous talocalcaneal ligament
to create a severe deformity. The investigators compared a talonavicular,
double (talonavicular and calcaneocuboid), triple, calcaneocuboid, and
subtalar arthrodeses. They found that a subtalar or calcaneocuboid
arthrodesis alone did not correct adequately the atfoot deformity based
on their radiographic criteria. Good correction of a atfoot deformity
was noted with the talonavicular and double arthrodesis. Although
adequate correction of the deformity was noted with the triple
arthrodesis, the investigators could not recommend this surgery for
other than a last resort.
A study looking at reconstructing the spring ligament using the
supercial portion of the deltoid proved to be clinically disappointing
secondary to probable attenuation of the graft.9 A technique using a free
Achilles graft as a stronger substitute was described subsequently, but
long-term follow-up has not been reported.
Tenodesis procedures have been explored using an experimental
atfoot model similar to the one described previously.42 Of the 4 methods
employed to correct the deformity, the investigators thought that a
reconstruction using the peroneus longus tendon, rerouting it through a
bone tunnel in the calcaneus (from medial to lateral) and anchoring it
on the lateral side of the calcaneus, provided the best correction in
sagittal and transverse planes. This laboratory study did not include
cyclic loading. It is not known whether the transferred tendon would
hold up under physiologic loads.
Tendon transfer using the exor digitorum longus combined with a
medial displacement calcaneal osteotomy was performed in 25 patients
with PTT insufciency and showed radiographic improvement in talonavicular coverage angle, talarrst metatarsal angle, and restoration of
the height of the medial arch at a mean of 20 months after surgery.33 It
was thought that the osteotomy provided a mechanical advantage by
reducing the deforming force created by the existing heel valgus, protecting the soft tissue augmentation.
In similar cadaver studies,36, 41 a calcaneal osteotomy alone was
found to correct an experimentally created atfoot deformity. The osteotomy relieved the elongation of the spring ligament, protecting it from
increased forces seen in the intact and lateral columnlengthened conditions.
Lateral column lengthening procedures in conjunction with medial

BIOMECHANICS OF PROCEDURES USED IN ADULT FLATFOOT DEFORMITY

21

Figure 3. Restoration of medial longitudinal arch in a loaded condition after a lateral column
lengthening at the calcaneocuboid joint.

column soft tissue augmentation44 has been shown clinically to provide


good correction of a atfoot deformity. A cadaver study30 examined
the differences in joint contact pressures between the Dillwyn-Evans
procedure and a calcaneocuboid distraction arthrodesis. These results
showed a signicant increase in the joint contact pressures at the anterior
and middle facets of the subtalar joint with the Dillwyn-Evans procedure. This nding might suggest consideration of maintaining the integrity of these joints when lengthening the lateral column to prevent
potential subtalar arthrosis (Fig. 3).
Combination procedures are likely to provide the most tailored
approach to correcting symptomatic atfoot deformities. Two clinical
studies report on combination procedures employing medial and lateral
bone procedures as well as soft tissue balancing. In one, children with
symptomatic atfoot were treated by a combination of Dillwyn-Evans
lateral column lengthening with a naviculocuneiform wedge resection
(medial and plantar), augmented with reconstruction and plication of
the plantar ligaments, rerouting of the anterior tendon, and a Z-plasty
of the Achilles.13 In another, adults were treated with a modied lateral
column lengthening using a distraction arthrodesis of the calcaneocuboid
joint combined with medial column arthrodesis and soft tissue balancing.6 Although the growing foot of a child cannot be compared directly
with the degenerative changes seen in adult atfoot deformity, they may
share common deformity and common deforming forces that must be
considered carefully and compensated for if reliable procedures are to
be developed to treat this condition successfully. More research is needed
to dene further the biomechanics of the foot and to understand the
signicance of the forces that combine to create atfoot deformity.
References
1. Ananthakrisnan D, Ching R, Tencer A, et al: Subluxation of the talocalcaneal joint in
adults who have symptomatic atfoot. J Bone Joint Surg Am 81:11471153, 1999

22

McCORMACK et al

2. Anderson J, Harrington R, Ching R, et al: Alterations in talar morphology associated


with adult atfoot. Foot Ankle Int 18:705709, 1997
3. Barry R, Scranton P: Flat feet in children. Clin Orthop 181:6875, 1983
4. Bordelon R: Hypermobile atfoot in children. Clin Orthop 181:714, 1983
5. Chao W, Wapner KL, Lee TH, et al: Nonoperative management of posterior tibial
tendon dysfunction. Foot Ankle Int 17:736741, 1996
6. Chi T, Toolan B, Sangeorzan B, et al: The lateral column lengthening and medial
column stabilization procedures. Clin Orthop 365:8190, 1999
7. Close JR, Inman VT, Poor PM, et al: The function of the subtalar joint. Clin Orthop
50:15979, 1967
8. Cowan D, Robinson J, Jones B, et al: Consistency of visual assessments of arch height
among clinicians. Foot Ankle Int 15:213217, 1994
9. Deland JT, Arnoczky SP, Thompson FM: Adult acquired atfoot deformity at the
talonavicular joint: Reconstruction of the spring ligament in an in vitro model. Foot
Ankle Int 13:327332, 1992
10. Deland JT, Otis JC, Lee KT, et al: Lateral column lengthening with calcaneocuboid
fusion: Range of motion in the triple joint complex. Foot Ankle Int 16:729733, 1995
11. Duncan J, Lovell W: Modied Hoke-Miller atfoot procedure. Clin Orthop 181:2427,
1983
12. Dyal CM, Feder J, Deland JT, et al: Pes planus in patients with posterior tibial tendon
insufciency: Asymptomatic versus symptomatic foot. Foot Ankle Int 18:8588, 1997
13. El-Tayeby H, Orth C: The severe exible atfoot: A combined reconstructive procedure
with rerouting of the tibialis anterior tendon. J Foot Ankle Surg 38:4149, 1999
14. Goh JC, Lee PY, Lee EH, et al: Biomechanical study on tibialis posterior tendon
transfers. Clin Orthop 319:297302, 1995
15. Gould N: Evaluation of hyperpronation and pes planus in adults. Clin Orthop 181:37
45, 1983
16. Harris R, Beath T: Hypermobile at-foot with short tendoachillis. J Bone Joint Surg
Am 30:116, 1948
17. Horton GA, Myerson MS, Parks BG, et al: Effect of calcaneal osteotomy and lateral
column lengthening on the plantar fascia: A biomechanical investigation. Foot Ankle
Int 19:370373, 1998
18. Huang CK, Kitaoka HB, An KN, et al: Biomechanical evaluation of longitudinal arch
stability. Foot Ankle Int 14:353357, 1993
19. Jahss MH: Spontaneous rupture of the tibialis posterior tendon: Clinical ndings,
tenographic studies, and a new technique of repair. Foot Ankle Int 3:158166, 1982
20. Kettelkamp DB, Alexander HH: Spontaneous rupture of the posterior tibial tendon. J
Bone Joint Surg Am 51:759764, 1969
21. Kitaoka HB, Luo ZP, An KN: Reconstruction operations for acquired atfoot: Biomechanical evaluation. Foot Ankle Int 19:203207, 1998
22. Kogler G, Veer F, Solomonodis S, et al: The inuence of medial and lateral placement
of orthotic wedges on loading of plantar aponeurosis: An in vitro study. J Bone Joint
Surg Am 81:14031413, 1999
23. Lundberg A, Goldie I, Kalin B, et al: Kinematics of the ankle/foot complex: Plantarexion and dorsiexion. Foot Ankle Int 9:194200, 1989
24. Lundberg A, Svensson OK, Bylund C, et al: Kinematics of the ankle/foot
complexPart 2: Pronation and supination. Foot Ankle 9:248253, 1989
25. Lundberg A, Svensson OK, Bylund C, et al: Kinematics of the ankle/foot
complexPart 3: Inuence of leg rotation. Foot Ankle 9:304309, 1989
26. Mann R: Rupture of the tibialis posterior tendon. Instr Course Lect 33:302309, 1984
27. Mann RA: Acquired atfoot in adults. Clin Orthop 181:4651, 1983
28. Mann RA: Flatfoot in adults. In Mann RA, Coughlin MJ (eds): Surgery of the Foot and
Ankle, ed 6. St. Louis, CV Mosby, 1993, pp 757801
29. Manter JT: Movements of the subtalar and transverse tarsal joints. Anat Rec 80:4, 1941
30. McCormack AP, Niki H, Kiser P, et al: Two reconstructive techniques for atfoot
deformity comparing contact characteristics of the hindfoot joints. Foot Ankle Int
19:452461, 1998

BIOMECHANICS OF PROCEDURES USED IN ADULT FLATFOOT DEFORMITY

23

31. Michelson JD, Mizel M, Jay P, et al: Effect of medial displacement calcaneal osteotomy
on ankle kinematics in a cadaver model. Foot Ankle Int 19:132136, 1998
32. Miller C, Laskowski E, Suman V: Effect of corrective rearfoot orthotic devices on
ground reaction forces during ambulation. Mayo Clin Proc 71:757762, 1996
33. Myerson MS, Corrigan J, Thompson F, et al: Tendon transfer combined with calcaneal
osteotomy for treatment of posterior tibial tendon insufciency: A radiological investigation. Foot Ankle Int 16:712718, 1995
34. Niki H, Ching RP, Kiser P, et al: The effect of posterior tibial tendon dysfunction on
hindfoot kinematics. Foot Ankle Int 2001
35. OMalley MJ, Deland JT, Lee KT: Selective hindfoot arthrodesis for the treatment of
adult acquired atfoot deformity: An in vitro study. Foot Ankle Int 16:411417, 1995
36. Otis J, Deland J, Kenneally S, et al: Medial arch strain after medial displacement
calcaneal osteotomy: An in vitro study. Foot and Ankle Int 20:222231, 1999
37. Reeck J, Felten N, McCormack AP, et al: Support of the talus: A biomechanical
investigation of the contributions of the talonavicular and talocalcaneal joints, and the
superomedial calcaneonavicular ligament. Foot Ankle Int 19:674682, 1998
38. Sands A, Early J, Harrington RM, et al: Effect of variations in calcaneocuboid fusion
technique on kinematics of the normal hindfoot. Foot Ankle Int 19:1925, 1998
39. Siegler S, Chen J, Schneck CD: The three-dimensional kinematics and exibility characteristics of the human ankle and subtalar jointsPart I: Kinematics. J Biomech Eng
110:364373, 1988
40. Tareco J, Miller N, MacWilliams B, et al: Dening atfoot. Foot Ankle Int 20:456460,
1999
41. Thordarson DB, Hedman T, Lundquist D, et al: Effect of calcaneal osteotomy and
plantar fasciotomy on arch conguration in a atfoot model. Foot Ankle Int 19:374
378, 1998
42. Thordarson DB, Schmotzer H, Chon J: Reconstruction with tenodesis in an adult
atfoot model: A biomechanical evaluation of four methods. J Bone Joint Surg Am
77:15571564, 1995
43. Thordarson DB, Schmotzer H, Chon J, et al: Dynamic support of the human longitudinal arch: A biomechanical evaluation. Clin Orthop 316:165172, 1995
44. Toolan B, Sangeorzan B, Hansen S: Complex reconstruction for the treatment of
dorsolateral peritalar subluxation of the foot. J Bone Joint Surg Am 81:15451560, 1999
45. Wright DG, Desai SM, Henderson WH: Action of the subtalar and ankle-joint complex
during the stance phase of walking. J Bone Joint Surg Am 46:361364, 1964
Address reprint requests to
Bruce J. Sangeorzan, MD
Department of Orthopaedics
Harborview Medical Center
University of Washington
325 Ninth Avenue, Box 359798
Seattle, WA 981042499

Calcaneal Osteotomy in the


Treatment of Adult Acquired
Flatfoot Deformity
Abhijit R. Guha, MBBS(Hons), MS(Orth), FRCS(Ed), FRCS(Tr & Orth)*,
Anthony M. Perera, MBChB, MRCS, MFSEM, PGDip (Med Law), FRCS (Orth)

KEYWORDS
Adult acquired flatfoot Arthrodesis Lateral column lengthening
Medial displacement calcaneal osteotomy Tibialis posterior
KEY POINTS
Adult acquired flatfoot deformity is a common problem after middle age.
Failure of the tibialis posterior tendon, followed by the plantar calcaneonavicular ligament
and the anterior fibers of the deltoid ligament, leads to deformity.
Medial displacement calcaneal osteotomy (MDCO) is used in stage 2 deformity to correct
the ground reaction force and prevents the heel from going into valgus. It converts the
gastrosoleus from an evertor to an invertor of the heel.
Lateral column lengthening procedures are used to correct forefoot abduction and
uncovering of the talar head.
Medial displacement calcaneal osteotomy may be used in combination with lateral column
lengthening to improve correction.

Adult acquired flatfoot deformity (AAFD) is a chronic, debilitating condition common


beyond middle age. It is characterized by a painful flattening of the medial longitudinal
arch and has been referred to as posterior tibial tendon deficiency. However, in
recognition of the fact that other structures such as the plantar calcaneonavicular
(spring) ligament and the deltoid ligament were also involved, the nomenclature has
been changed to AAFD.
Weakness of the tibialis posterior leads to the inability to invert the heel taking it
from valgus in the midstance phase of gait into inversion. This is key to initiating heel
rise by changing the line of pull of the Achilles; it also locks the midfoot, providing a
rigid lever for push-off. Without this, the transverse tarsal joint axes remain in
alignment and there is no longer a rigid lever arm for the gastrocnemius muscle to
Department of Trauma and Orthopaedics, University Hospital of Wales, Heath Park, Cardiff,
Wales, CF14 4XW, UK
* Corresponding author.
E-mail address: footandanklesurgery@gmail.com
Foot Ankle Clin N Am 17 (2012) 247258
doi:10.1016/j.fcl.2012.02.003
foot.theclinics.com
1083-7515/12/$ see front matter Crown Copyright 2012 Published by Elsevier Inc. All rights reserved.

248

Guha & Perera

contract against in the heel rise phase of gait. This subsequently puts a repetitive load
on the medial supporting structures, leading to the eventual degeneration and
weakening of the tibialis posterior, spring ligament and the anterior fibers of the
deltoid ligament.
Failure of these structures leads to the classic pes planovalgus deformity. The
consequent shortening of the gastrocnemius further exacerbates the flattening of the
medial longitudinal arch. In the presence of a short gastrocnemius muscle, terminal
dorsiflexion of the ankle is achieved by rotation around the talonavicular joint; over
time, this joint becomes subluxed. This peritalar instability results in the navicular
moving superiorly, laterally, and supinating further, contributing to instability of the
medial column and the collapse of the arch.
CLASSIFICATION

The condition was originally classified by Johnson and Strom1 into 3 stages with a
fourth stage added later.2 The use of a calcaneal osteotomy is most pertinent to stage
2, although there are rare instances in stages 3 and 4 when this may be required in
addition to fusion surgery to achieve full correction. For instance, when there is a
severe deformity but insufficient bone stock to perform the correction purely through
a fusion.
Stage 2 disease represents tendon failure in a planovalgus foot that has not yet
become rigid or degenerate and with normal ankle alignment. The aim of surgery is
functional reconstruction and rebalancing of the foot with a tendon transfer. There is
clear evidence that a bony heel correction is a key part of the operative management
of stage 2 disease. However, the issue with this classification system is that stage 2
includes a significant variation in the type and severity of deformity of the midfoot and
hindfoot. Therefore, this stage has been further subdivided by various authors,
depending on the degree of coverage of the talonavicular joint and forefoot varus
(Anderson, Instructional Course Lectures, 2003), or the degree of resting supination of
the forefoot and its correctability in the Truro classification.3 Myerson and colleagues4
have proposed a detailed classification of AAFD that would help to guide the surgeon
to decide on the appropriate surgical options available at the various stages of the
disease. This is a comprehensive classification, incorporating the wide variety of
presentations in each stage and helps to individualize treatment according to the
patients particular pathology. This morphologic approach aids with preoperative
planning in selecting the most appropriate type of bony correction: Medial Displacement Calcaneal Osteotomy (MDCO), lateral column lengthening, a combination, or
arthroereisis (not strictly an osteotomy but included here for completeness; Table 1).
OPERATIVE MANAGEMENT

Failure of nonoperative management is the commonest indication to proceed to


operative management. Most patients with AAFD requiring operative intervention
have stage 2 deformity. The operative management of stage 2 deformity has
undergone a change in direction over the last 30 years.5 Joint-preserving procedures,
which aim to realign the hindfoot and augment the dysfunctional tibialis posterior
tendon, have now taken precedence over hindfoot fusion procedures, which are now
used if the hindfoot joints are already arthritic or in the revision/salvage setting. Before
proceeding to surgery, one must consider the flexibility of the foot, the apex of the
deformity, the condition of the tibialis posterior tendon, the spring ligament, the
deltoid ligament, and presence of arthritis in the hindfoot joints. Augmentation of
the dysfunctional tibialis posterior tendon by transfer of the flexor digitorum longus

Calcaneal Osteotomy

Table 1
Comprehensive staging with pathology and treatment options in AFFD
Stage

Subdivision Pathology

Recommendation

Flexible hindfoot valgus with


Flexible forefoot varus

Medial posting/brace tendoachilles


lengthening if forefoot varus
corrects only in equinus

Fixed forefoot varus

MDCO (arthroereisis) FDL transfer


Cotton osteotomy if fixed forefoot
varus

Forefoot abduction (at transverse


tarsal, first TMT joint or both)
Talar head uncovering 40%

MDCO FDL transfer

Talar head uncovering 40%

FDL transfer Lateral column


lengthening MDCO/arthroereisis
(if residual heel valgus)

Medial ray instability


Persistent forefoot varus after
correction of heel
Talonavicular/naviculocuneiform/
first TMT joint level

Fusion of appropriate joint if


arthritic/Cotton osteotomy

Abbreviations: Cotton osteotomy, opening wedge medial cuneiform osteotomy; FDL, flexor digitorum longus; TMT, tarsometatarsal.
Adapted from Bluman EM, Title CI, Myerson MS. Posterior tibial tendon rupture: a refined
classification system. Foot Ankle Clin 2007;12:233 49.

(FDL) tendon does not seem to work when performed alone. This has been
established by various authors in both clinical and laboratory studies.6-8 Hence,
corrective surgery most often involves a calcaneal osteotomy, repair of the spring
ligament and talonavicular joint capsule (principal static restraints) and augmentation
of the tibialis posterior with the FDL, flexor hallucis longus, or sometimes even the
peroneus brevis. If the gastrocnemius muscle is contracted, a recession is performed
to improve dorsiflexion at the ankle and reduce strain on the medial structures.
Subtalar arthroereisis has also been advocated in the recent literature, to improve foot
alignment by blocking rotation of the anterolateral talar body into the sinus tarsi and
restricting subtalar joint movement preventing the peritalar instability.9 The addition of
the MDCO was the major step forward in the success of this surgery. Since then,
other calcaneal osteotomies have been described.
MDCO
History

Gleich first attempted a calcaneal osteotomy to treat pes planus in 1893.10 He


attempted to restore a normal calcaneal pitch angle by performing a medial closing
wedge osteotomy of the calcaneus and displacing the posterior fragment anteriorly,
medially, and downward. Dwyer in 1960 proposed a lateral opening wedge osteotomy
using tibial bone graft.
Biomechanics

MDCO allows greater correction of the deformity; most important, it changes the
calcaneal axis protecting the soft tissue reconstruction using the ground reaction

249

250

Guha & Perera

Fig. 1. Medial displacement calcaneal osteotomy.

force to prevent the heel from going into valgus. It also exerts a positive influence on
walking dynamics. The MDCO (Figs. 1 and 2) restores the line of pull of the
tendoachilles to the medial side of the subtalar joint axis and this helps to correct the
valgus deformity of the hindfoot. The moment arm of the gastrosoleus in improved

Fig. 2. Change in hindfoot axis after medial displacement calcaneal osteotomy.

Calcaneal Osteotomy

and it is converted from an evertor in the deformed foot to an invertor of the heel in
the corrected one.
MDCO has been reported to reduce the contact stresses at the tibiotalar joint and
it is hoped that this would protect the ankle joint from degeneration in the future.11
Other cadaveric studies have reported an increase in the peak pressure over the
lateral forefoot and heel after MDCO.12 Translation of 10 mm substantially decreases
the load on the first metatarsal, medial arch, and the moment at the talonavicular joint,
and increases the load on the lateral foot,12 reversing the change in distribution of
force that occurs in a failed tibialis posterior tendon.
Indications and Outcomes

Koutsogiannis13 reintroduced the MDCO for treatment of AAFD in 1971. Since then,
many authors including Myerson and colleagues,4 Pomeroy and Manoli,14 Fayazi and
associates,15 Wacker and co-workers,16 Guyton and colleagues,17 and Sammarco
and Hockenbury18 have published good results after MDCO, tendon transfer (either
FDL or flexor hallucis longus), and spring ligament repair for the treatment of AAFD.
MDCO is now recognized as an established procedure for surgical treatment of AAFD.
MDCO can be used in any situation where hindfoot valgus is present and protection
of the medial soft tissues is required. Classically, it is used in the surgical treatment
of stage 2 disease, in combination with other procedures. A cadaveric study showed
that MDCO with FDL transfer was adequate in moderate degrees of deformity, but in
cases of severe deformity, additional procedures were required.19 It can be used in
combination with tenosynovectomy of the tibialis posterior tendon in stage 1 disease
and also with triple fusion, spring ligament reconstruction, and deltoid ligament
reconstruction.
MDCO is a straightforward, reproducible, reliable procedure with predictable
results. Myerson presented 120 patients treated in this way with marked improvement
in pain and function in 90%.4 However, radiologic arch improvement is not always
achieved with this protocol; other series have shown that, despite good clinical
outcomes, radiologic improvement in the arch itself may not be achieved in 0% to
75%. It has been noted that even this improvement in the radiologic parameters can
be lost with the passage of time.20 Furthermore, although there may be improvement
in the radiographic height and alignment of the medial longitudinal arch in some
patients, the outer appearance is significantly changed in only 4% of patients.17
Technique

The calcaneus is exposed through an oblique incision in line with the osteotomy or an
extended lateral approach (Fig. 3A) with development of a full thickness flap
anteriorly. The oblique incision allows a direct approach to the lateral calcaneal wall
and is made parallel to the posterior facet of the subtalar joint and 2 cm posterior to
it. A subperiosteal elevation is performed and soft tissue retractors are placed in the
soft areas at either end of the incision.
The extended lateral approach is a radical anatomic dissection of a flap involving
the skin, fascia, and local muscle with its blood supply. It exploits the watershed
between the cutaneous supply of the posterior peroneal artery superiorly and the
lateral plantar artery inferiorly. In contrast with the direct lateral approach, damage to
the sural nerve is reduced and wound healing is not a concern.21-24 The location of the
skin incision is critical if damage to the sural nerve is to be avoided. The proximal arm
of the incision runs almost to the midline posteriorly, staying away from the sural
nerve. Care must be taken while dissecting around the peroneal tubercle to protect
the peroneal tendons and the sural nerve is again potentially in danger at the distal

251

252

Guha & Perera

Fig. 3. (A) Extended lateral approach (B) Medial displacement calcaneal osteotomy held with
step locking plate.

limb of the incision.22 The extended lateral approach affords excellent exposure of the
entire lateral wall of the calcaneus.25
The lateral wall of the tuberosity is exposed and Hohman retractors are placed on
the dorsal and plantar aspects of the tuberosity. The dorsal retractor protects the
Achilles tendon attachment, and the inferior one protects the plantar fascia. The
osteotomy is started with an oscillating saw blade at a 90 angle to lateral wall if
translation without plantarflexion is planned and at a right angle to the calcaneal axis

Calcaneal Osteotomy

(parallel to the posterior facet). A chevron-shaped osteotomy has been proposed as


a more stable construct.
Completion of the medial cortex is done in a carefully controlled manner using a
broad osteotome. The medial periosteum is then released on either side with an
elevator. Displacement of at least 10 to 12 mm can be achieved after gentle sustained
stretching of the medial soft tissues using a laminar spreader. The osteotomy can be
fixed with either lag screws or step plates with locking screws (see Fig. 3B). It is
important to hold the foot in dorsiflexion while fixing the osteotomy to compress the
bony surfaces together.
Crushplasty of the prominent lateral edge has been recommended to avoid
peroneal tendon pathology and sural nerve irritation. Lateral step plates are easy to
apply, but can be prominent and preclude the use of a crushplasty. One to two
6.5-mm screws allow compression, easier soft tissue closure, and crushplasty, but
can be prominent on the back of the heel.
The osteotomy is most often done in addition to other procedures that require
immobilization of the limb in a plaster cast. However, after an isolated osteotomy,
weight bearing may be allowed from 2 weeks; however, when a tendon transfer or
spring ligament repair has also been undertaken, weight bearing is allowed after 4 to
6 weeks.
Complications

Nonunion is extremely rare and malunion can be avoided by good fixation and
controlling the amount of bony contact that is maintained. A correction of 10 to 15 mm
allows more than 50% contact and, thus, a stable construct.
Complications mainly involve the soft tissue structures on the medial side. Injury to
the medial neurovascular bundle can be avoided by support under the distal tibia so
that the medial ankle is unsupported, avoiding pressure on the medial structures. A
pulsing action is used on the saw to cut to the far cortex and then broaching the
medial cortex with an osteotome in a controlled fashion.
Sporadic case reports of injury to medial structures after calcaneal osteotomy
appear in the literature. Tibial nerve palsy,26 pseudoaneurysm of the posterior tibial
artery and arteriovenous fistula,27 and lateral plantar artery pseudoaneurysm28 have
been reported. A cadaveric study reported that a minimum of 2 medial structures
crossed the osteotomy site at various positions. In most cases, branches of the lateral
plantar nerve and the posterior tibial artery were involved.29 The calcaneal sensory
branch and the second branch of the lateral plantar nerve (Baxters nerve) crossed the
osteotomy site in 86% and 95% of specimens, respectively.30 Injury to these
branches may result in postoperative medial hindfoot pain. The authors recommended completion of the medial cortex of the osteotomy in a carefully controlled
manner.
Limitations

MDCO does not correct forefoot abduction and uncovering of the talonavicular joint.
To improve this, a lengthening procedure of the lateral column needs to be
undertaken. This involves either a lengthening osteotomy through the anterior aspect
of the calcaneus or a distraction arthrodesis of the calcaneocuboid joint if this joint
has evidence of arthritic changes.
Another potential criticism is that it inconsistently restores the height of the medial
longitudinal arch.31 Wacker and co-workers16 showed that, despite good early
success, in 75% there was a recurrence there was a 10% recurrence rate at 3 to 5
years. It is not clear whether this is related to the degree of correction of the foot.

253

254

Guha & Perera

LATERAL COLUMN LENGHTENING OSTEOTOMY


History

Lateral column lengthening with tibialis posterior augmentation has also been used to
correct stage 2 AAFD. The concept of lateral column was introduced by Evans.32 He
observed that, in patients with clubfeet, adduction of the forefoot could be corrected
by shortening of the lateral column. Conversely, he proposed lengthening of the
lateral column to correct forefoot abduction in congenital spastic flatfoot. He thus
introduced the classic Evans osteotomy to lengthen the lateral column. This concept
was later applied to correction of the adult flatfoot.33 This aspect will be dealt with in
detail in another article by Calder and colleagues elsewhere in this issue.
DOUBLE CALCANEAL OSTEOTOMY

This includes a MDCO along with a lateral column lengthening osteotomy of the
calcaneus.
Biomechanics

These osteotomies evolved to correct 2 major aspects of AAFD. They incorporate a


MDCO to correct hindfoot valgus with a lateral column lengthening to correct forefoot
abduction and increase arch height. Moreover, the aim is to share the amount of
correction between the 2 sites, reducing the pressures in each area. This can result
in a more anatomic correction of the deformity and may further decrease the load on
the dysfunctional posteromedial structures.
Indications and Outcomes

When used in combination with FDL transfer, there is symptomatic pain relief and the
correction obtained has been noted to be lasting at intermediate follow-up.32,33 There
is little published on this technique.
MALERBA OSTEOTOMY
History

Malerba and De Marchi34 advocated the use of calcaneal osteotomies in both varus
and valgus deformities of the hindfoot. They noted that preservation of the subtalar
joint and correction of its biomechanics afforded a positive influence on walking
dynamics. Malerba and De Marchi described a Z osteotomy of the retrothalamic
portion of the calcaneus (Fig. 4). This combines a medial displacement of the
calcaneal tuberosity and lateral column lengthening through an extensile approach.
Technique

The calcaneus is approached directly through an oblique incision parallel and distal to
the peroneal tendons, thereby reducing risk of injury to the peroneal tendons or the

Fig. 4. Malerba Z osteotomy.

Calcaneal Osteotomy

sural nerve. The osteotomy is performed using an oscillating saw and completed with
an osteotome to protect the medial neurovascular structures. There are 3 osteotomic
sections, similar in length and at right angles to each other. The central horizontal limb
is at least 1.5 cm long, parallel to the weight bearing surface and distal to the peroneal
tubercle, avoiding the need to expose the peroneal tendons. The proximal vertical
limb exits into the dorsal concavity of the calcaneus between the attachment of the
Achilles tendon and the posterior facet of the subtalar joint, whereas the distal vertical
limb reaches the plantar cortex immediately anterior to the tuberosity, keeping well
away from the calcaneocuboid joint.
This osteotomy allows the calcaneal tuberosity to be shifted medially or laterally;
removal of a laterally based wedge adds to the correction of heel varus in pes cavus
surgery. The osteotomy incorporates a large bony surface and its inherent stability
encourages early consolidation. It is held with staples or screws and plaster
immobilization is not necessary unless required by other associated procedures. Full
weight bearing is usually possible within 6 weeks. Malerba osteotomy is a simple
technique with limited risk of injury to the peroneal tendons or sural nerve and affords
excellent potential to correct deformity on the frontal plane.
ARTHROEREISIS SCREW

Although this is not strictly a calcaneal osteotomy, it is included because it is used for
a similar effect. It is perhaps the most controversial of the techniques described and
is primarily indicated in the skeletally immature patient with dorsolateral peritalar
instability. In this group, it is theorized that the implant blocks heel valgus and thus
allows the foot to remodel with growth; nonetheless, it has been used in adults.
Biomechanics

In addition to correcting heel alignment, a 6-mm arthroereisis screw is successful at


reversing the transfer in load to the medial foot that occurs with a tibialis posterior
tendon dysfunction in a biomechanical model. The load on the first metatarsal, medial
arch and talonavicular moment was reduced significantly.35
Indications and Outcomes

An arthroereisis screw can correct only mild deformity; therefore, it may have a role
in the management of patients with little or no deformity and those in whom an
adjunct to osteotomy is required. It may not correct the arch; the intention is really to
block pronation protecting the medial tissues.
Complications

Despite the ease of insertion and ability to correct mild heel valgus complications are
common. Needleman36 reported that at 44 months 46% had sinus tarsi pain.
Instability of the implant, pain, stiffness, and failure to improve deformity are also
common.
SUMMARY

Calcaneal osteotomies are an essential part of our current armamentarium in the


treatment of AAFD. Soft tissue correction or bony realignment alone have failed to
adequately correct the deformity37; therefore, both procedures are used simultaneously to achieve long-term correction. Medial displacement and lateral column
lengthening osteotomies in isolation or in combination and the Malerba osteotomy
have been employed along with soft tissue balancing to good effect by various

255

256

Guha & Perera

authors. The goal is to create a stable bony configuration with adequate soft tissue
balance to maintain dynamic equilibrium in the hindfoot.
In pronatory syndromes, the relation of the osteotomy to the posterior subtalar
facet modifies the biomechanics of the hindfoot in different ways. Anterior calcaneal
osteotomies correct deformities in the transverse plane (forefoot abduction), whereas
posterior tuberosity osteotomies result in varization of the calcaneus and correct the
frontal plane deformity.
The choice of osteotomy depends on the plane of the dominant deformity. If the
subtalar axis is more horizontal than normal, transverse plane movement is cancelled
out and the frontal plane eversioninversion is predominant. The patient presents with
marked hindfoot valgus without significant forefoot abduction. Conversely, if the
subtalar axis is more vertical than normal, transverse plane movement is predominant
and the patient presents with forefoot abduction and instability of the medial midtarsal
joints, although without significant hindfoot valgus. In this situation, a lateral column
lengthening procedure is recommended to decrease the uncovering of the talar head
and improve the height of the arch while correcting the forefoot abduction. With a
predominant frontal plane deformity, medialization of the calcaneal tuberosity is used
to displace the calcaneal weight bearing axis medially, aligning it with the tibial axis
and restoring the function of the gastrosoleus as a heel invertor. An essential
prerequisite for this is the absence of arthritis affecting the subtalar joint. The Achilles
tendon may need to be lengthened at the same time.
REFERENCES

1. Johnson KA, Strom DE. Tibialis posterior tendon dysfunction. Clin Orthop Relat Res
1989;239:196 206.
2. Bluman EM, Title CI, Myerson MS. Posterior tibial tendon rupture: a refined classification system. Foot Ankle Clin 2007;12:233 49.
3. Parsons S, Naim S, Richards PJ, et al. Correction and prevention of deformity in type
II tibialis posterior dysfunction. Clin Orthop Relat Res 2010;468:102532.
4. Myerson MS, Badekas A, Schon LC. Treatment of stage II posterior tibial tendon
deficiency with flexor digitorum longus tendon transfer and calcaneal osteotomy. Foot
Ankle Int 2004;25:44550.
5. Pinney SJ, Lin SS. Current concept review: acquired adult flatfoot deformity. Foot
Ankle Int 2006;27:66 75.
6. Mann RA, Thompson FM. Rupture of the posterior tibial tendon causing flat foot.
Surgical treatment. J Bone Joint Surg Am 1985;67:556 61.
7. Kitaoka HB, Luo ZP, An KN. Reconstruction operations for acquired flatfoot: biomechanical evaluation. Foot Ankle Int 1998;19:2037.
8. Hill K, Saar WE, Lee TH, et al. Stage II flatfoot: what fails and why. Foot Ankle Clin
2003;8:91104.
9. Needleman RL. Current topic review: subtalar arthroereisis for the correction of flexible
flatfoot. Foot Ankle Int 2005;26:336 46.
10. Haddad SL, Myerson MS, Younger A, et al. Symposium: adult acquired flatfoot
deformity. Foot Ankle Int 2011;32:95111.
11. Hadfield M, Snyder J, Liacouras P, et al. The effects of a medializing calcaneal
osteotomy with and without superior translation on Achilles tendon elongation and
plantar foot pressures. Foot Ankle Int 2005;26:36570.
12. Arangio GA, Salathe EP. A biomechanical analysis of posterior tibial tendon dysfunction, medical displacement calcaneal osteotomy and flexor digitorum longus transfer
in adult acquired flatfoot. Clin Biomech 2009;24:38590.

Calcaneal Osteotomy

13. Koutsogiannis E. Treatment of mobile flat foot by displacement osteotomy of the


calcaneus. J Bone Joint Surg Br 1971;53:96 100.
14. Pomeroy GC, Manoli A. A new operative approach for flatfoot secondary to posterior
tibial tendon insufficiency: a preliminary report. Foot Ankle Int 1997;18:206 12.
15. Fayazi AH, Nguyen HV, Juliano PJ. Intermediate term follow-up of calcaneal osteotomy and flexor digitorum longus transfer for treatment of posterior tibial tendon
dysfunction. Foot Ankle Int 2002;23:110711.
16. Wacker JT, Hennessy MS, Saxby TS. Calcaneal osteotomy and transfer of the tendon
of flexor digitorum longus for stage-II dysfunction of tibialis posterior. Three- to
five-year results. J Bone Joint Surg Br 2002;84:54 8.
17. Guyton GP, Jeng C, Krieger LE, et al. Flexor digitorum longus transfer and medial
displacement calcaneal osteotomy for posterior tibial tendon dysfunction: a middleterm clinical follow-up. Foot Ankle Int 2001;22:62732.
18. Sammarco GJ, Hockenbury RT. Treatment of stage II posterior tibial tendon dysfunction with flexor hallucis longus transfer and medial displacement calcaneal osteotomy.
Foot Ankle Int 2001;22:30512.
19. Vora AM, Tien TR, Parks BG, et al. Correction of moderate and severe acquired
flexible flatfoot with medializing calcaneal osteotomy and flexor digitorum longus
transfer. J Bone Joint Surg Am 2006;88:1726 34.
20. Bolt PM, Coy S, Toolan BC. A comparison of lateral column lengthening and medial
translational osteotomy of the calcaneus for the reconstruction of adult acquired
flatfoot. Foot Ankle Int 2007;28:111523.
21. Freeman BJ, Duff S, Allen PE, et al. The extended lateral approach to the hindfoot.
Anatomical basis and surgical implications. J Bone Joint Surg Br 1998;80:139 42.
22. Eastwood DM, Irgau I, Atkins RM. The distal course of the sural nerve and its
significance for incisions around the lateral hindfoot. Foot Ankle 1992;13:199 202.
23. Harvey EJ, Grujic L, Early JS, et al. Morbidity associated with ORIF of intra-articular
calcaneus fractures using a lateral approach. Foot Ankle Int 2001;22:868 73.
24. Gould N. Lateral approach to the os calcis. Foot Ankle 1984;4:218 20.
25. Benirschke SK, Sangeorzan BJ. Extensive intraarticular fractures of the foot. Surgical
management of calcaneal fractures. Clin Orthop Relat Res 1993;292:128 34.
26. Krause FG, Pohl MJ, Penner MJ, et al. Tibial nerve palsy associated with lateralizing
calcaneal osteotomy: case reviews and technical tip. Foot Ankle Int 2009;30:258 61.
27. Doty JF, Alvarez RG, Asbury BS, et al. Arteriovenous fistula and pseudoaneurysm of
the posterior tibial artery after calcaneal slide osteotomy: a case report. Foot Ankle Int
2010;31:329 32.
28. Ptaszek AJ, Aminian A, Schneider JR, et al. Lateral plantar artery pseudoaneurysm
after calcaneal osteotomy: a case report. Foot Ankle Int 2006;27:1413.
29. Vermeulen K, Neven E, Vandeputte G, et al. Relationship of the scarf valgus-inducing
osteotomy of the calcaneus to the medial neurovascular structures. Foot Ankle Int
2011;32:S540 4.
30. Greene DL, Thompson MC, Gesink DS, et al. Anatomic study of the medial
neurovascular structures in relation to calcaneal osteotomy. Foot Ankle Int 2001;
22:569 71.
31. Pomeroy GC, Pike RH, Beals TC, et al. Acquired flatfoot in adults due to dysfunction
of the posterior tibial tendon. J Bone Joint Surg Am 1999;81:1173 82.
32. Evans D. Calcaneo-valgus deformity. J Bone Joint Surg Br 1975;57:270 8.
33. Sangeorzan BJ, Mosca V, Hansen ST. Effect of calcaneal lengthening on relationships
among the hindfoot, midfoot, and forefoot. Foot Ankle 1993;14:136 41.
34. Malerba F, De Marchi F. Calcaneal osteotomies. Foot Ankle Clin 2005;10:523 40.

257

258

Guha & Perera

35. Arangio GA, Reinert KL, Salathe EP. A biomechanical model of the effect of subtalar
arthrodesis on the adult flexible flat foot. Clin Biomech 2004;19:84752.
36. Needleman RL. A surgical approach for flexible flatfeet in adults including a subtalar
arthrodesis with the MBA sinus tarsi implant. Foot Ankle Int 2006;27:9 18.
37. Trnka HJ, Easley ME, Myerson MS. The role of calcaneal osteotomies for correction
of adult flatfoot. Clin Orthop Relat Res 1999;365:50 64.

Is There a Role for Subtalar


Arthroereisis in the Management
of Adult Acquired Flatfoot?
Pablo Fernndez de Retana, MDa,, Fernando lvarez, MDb,
Gustavo Bacca, MDb

KEYWORDS
Adult acquired Arthroereisis Pes planovalgus Subtalar implant
Subtalar joint
KEY POINTS
Subtalar arthroereisis is 1 treatment option for stage IIA posterior tibial tendon
dysfunction, as an alternative to calcaneal medializing osteotomy.
Subtalar arthroereisis, often combined with Achilles tendon lengthening, is a simple
and effective way to treat flexible flatfoot in adults.
The most common complication is pain in sinus tarsi that usually disappears after
removal of implant. Midterm results are good and it does not hinder other treatments
in the future.

Adult acquired flatfoot is a common problem for foot and ankle surgeons. The
incidence is increasing and it is becoming more widely known among orthopedic
surgeons in the last years. The main cause is rupture of posterior tibial tendon (PTT).
The morphologic characteristics of this condition are heel valgus and flattening of the
medial longitudinal arch. Other characteristics are usually observed, such as supination and abduction of the forefoot and tightening of the Achilles tendon. The deformity
is progressive and patients describe sinking of their foot. There is no spontaneous
correction with time and many patients become symptomatic. Treatment is required
when there is progression of deformity and/or pain. Conservative treatment includes
supportive footwear and ankle orthosis.1 The natural evolution of adult flatfoot and
failure to treat this condition could lead to persistent pain in the hindfoot, hallux
valgus, degenerative arthritis, metatarsalgia, and knee or low back pain.

a
Orthopaedic Surgery Department, Hospital San Rafael, Passeig Vall d=Hebron 107-117, 08035
Barcelona, Spain; b Foot and Ankle Unit, Orthopaedic Surgery Department, Hospital San Rafael,
Passeig Vall d=Hebron 107-117, 08035 Barcelona, Spain
* Corresponding author.
E-mail address: pfernan@hsrafael.es

Foot Ankle Clin N Am 17 (2012) 271281


http://dx.doi.org/10.1016/j.fcl.2012.03.006
1083-7515/12/$ see front matter 2012 Elsevier Inc. All rights reserved.

foot.theclinics.com

272

Fernndez de Retana et al

Surgery is commonly indicated in adult flatfeet. Flexible flatfoot should be treated


preferably with extra-articular procedures. There are different surgical options:
Tendon transfers, osteotomies, and subtalar joint motion blocking procedures.
Shortening of the gastrocnemius or the Achilles tendon is very common in flatfoot.
Often, it is necessary to lengthen these structures. The final decision is usually based
on the clinical and radiographic assessment of the patient on an individual basis and
the surgeon=s preferences. Chambers2 in 1946 was the first to introduce the idea of
restricting subtalar joint motion with an autologous bone block filling the sinus tarsi.
In 1952, Grice3 observed that extra-articular subtalar arthrodesis with cortical graft for
flatfoot deformity in children had several problems, such as development of degenerative arthritis in adjacent joints and inability of the hindfoot to adapt to uneven
surfaces. Haraldsson4 and Lelievre5 pointed out that the most important was to block
the sinus tarsi restricting subtalar motion while avoiding arthrodesis. Since then, the
term arthroereisis (Latin: artro joint; -ereisis support or prop up) is used to
describe the limitation of subtalar motion.5 Viladot and co-workers6 presented their
experience treating PTT dysfunction stage II with arthroereisis.
SUBTALAR ARTHROEREISIS WITH SINUS TARSI IMPLANT

Arthroereisis has been used predominantly to treat flatfeet in the pediatric population.7,8 Many arthroereisis procedures have been described to limit subtalar joint
motion and to improve position. Lelievre5 in 1970 introduced the term lateral
arthroereisis using a temporary staple across the subtalar joint. In 1977, Subotnick9
was the first to describe a sinus tarsi implant. He used a block of silicone elastamer.
Smith and Millar10 used a polyethylene screw in sinus tarsi (STA peg, Dow Corning
Wright, Midland, MI, USA) and reported a 96% success rate in children. Viladot11
reported the use of a silicon implant with a cup shape, obtaining 99% good results in
234 children. Presently, Silicon implants are not usually used and have been replaced
with troncoconical implants like Kalix (New Deal, Lyon, France) and MBA (Kinitekos
Medical, Carlsbad, CA, USA). Zaret and Myerson7 reported 23 children with flexible
flatfoot treated with MBA implants.7 They reported postoperative sinus tarsi pain in
18% of patients that improved with rest and corticosteroid injection. Two children
required implant removal at a mean of 9 months after surgery. Subtalar joint range of
motion increased after implant removal. They observed an interesting phenomenon:
Removal of the implant was not associated with a reversal of the foot structure.
Vogler12 classified sinus tarsi implants into 3 types based on their biomechanical
properties: Self-locking wedge, axis-altering implant, and impact-blocking device.12
The majority of sinus tarsi implants are self-locking wedges. All sinus tarsi devices
restrict hindfoot valgus and the calcaneus is vertically orientated beneath the ankle
joint.8 The talus is dorsiflexed and externally deviated. Talonavicular subluxation and
forefoot abduction are corrected.
Indications

According to Johnson and Strom,13 PTT dysfunction is classified in 3 stages. Stage


I PTT has degenerative changes, but no deformity. Stage II is a flexible flatfoot. In the
beginning, the hindfoot is in valgus and progresses to abduction and supination in the
forefoot. Stage III is a rigid deformity. Myerson14 described stage IV when the ankle
joint is affected. Recently, a refined classification system has been proposed that
takes into account forefoot supination, forefoot abduction, and medial column
instability.15 Stage II has been subclassified in A, B, and C substages. There is a
valgus deformity in substage A, abduction in substage B, and medial column stability
in substage C.

Subtalar Arthroereisis in Adult Acquired Flatfoot

Fig. 1. (A) Preoperative lateral x-ray. (B) Preoperative anteroposterior x-ray. (C) Postoperative lateral x-ray with Kalix endorthesis. The forefoot is in slight supination. (D) Postoperative
anteroposterior x-ray with Kalix in supple adduction.

Surgery is indicated for symptomatic flatfoot stage II, which persists despite
nonsurgical treatment or when progression of deformity is observed. Tenosynovectomy is not recommended for stage II disease.16 Surgical options are osteotomies, medial column arthrodesis, tendon transfers, and sinus tarsi implants.
Sinus tarsi implants are useful for correcting valgus (substage A). Forefoot
abduction (substage B and C) will not be corrected with sinus tarsi implant, even
though temporarily some forefoot adduction and supination is observed after
arthroereisis (Fig. 1).
Flexor digitorum longus (FDL) transfer has become a mainstay for soft-tissue
procedures in patients with stage II PTT dysfunction.14,17 Biomechanical analysis
shows no change in arch alignment or height with this procedure alone.18 When
performed alone, it fails to restore normal biomechanics of the foot.14 It has long been
recognized that the spring ligament is essential in the foot structure.19 It should be
sutured or replaced when it is torn. An osteotomy should be used to correct deformity
in the long term. Medial displacement calcaneal osteotomy (MDCO) and FDL tendon
transfer is commonly accepted for treatment in stage IIA (valgus deformity). Sinus
tarsi implant and FDL transfer has the same indication as MDCO and FDL transfer.
The advantages of arthroereisis compared with MDCO are that is easy and quick to
perform, less immobilization is required, there is no risk of nonunion, and it avoids

273

274

Fernndez de Retana et al

Fig. 2. (A, B) Loss of position of the implant.

medial neurovascular structures injuries, troughing, and malunion. It has been


demonstrated that the use of a subtalar arthroereisis returns substantially deforming
forces and moments of the flatfoot back toward the values experienced in a normal
foot.20 The disadvantages of arthroereisis are a high percentage of implant intolerance (around 30% in our experience) and limitation of subtalar motion. There are no
controlled trial studies comparing these techniques and the level of scientific
evidence is low.
Complications

Complications may be divided into general and implant specific.8 General complications include persistent sinus tarsi pain, malposition, overcorrection, undercorrection,
wrong implant size, and loss of position (Fig. 2).8 The most common complication is
intolerance of the implant owing to pain. Saxena and Nguyen21 indicated that the
implants must be designed and shaped to fit adequately the configuration of the tarsal
sinus tarsi. Overcorrection is a cause of pain in sinus tarsi and implant removal has
been recommended with good results.22 Debris from wear, foreign body reaction,
implant degradation, and implant fracture are implant-specific complications.8 Arthritis, synovitis, implant fracture, and implant loosening have been reported with the
STA-peg implant.10 There are case reports of intraosseous cysts within the talus,23
osteonecrosis of the talus,24 peroneal muscle spasms, and small fracture of the
calcaneus.25 The authors have observed a case with complex regional pain syndrome
type I in a patient with a sinus implant (Fig. 3).

Subtalar Arthroereisis in Adult Acquired Flatfoot

Fig. 3. Complex regional pain syndrome type I after Sinus implant.

Wound complications can occur with surgical incisions. Predisposing factors for
development of these complications include tissue quality, lack of subcutaneous
tissue around approaches, and postoperative swelling.26
Operative Technique

The authors usually perform foot surgery under block anesthesia in adults. The patient
is placed in a supine position with a pillow under the ipsilateral buttock to avoid
excessive external rotation of the foot and facilitate approach to the sinus tarsi. A
pneumatic tourniquet is applied at the thigh and the surgical field is prepared leaving
the leg exposed.
In most cases, the first step of the operation is the lengthening of the Achilles
tendon. To evaluate whether this tendon is retracted, the authors hold the foot with
the hindfoot in neutral position and the knee completely extended. After that, the
authors passively dorsiflex the foot. If the authors cannot achieve 10 of dorsiflexion,
the authors consider that the Achilles tendon is retracted and it needs to be
lengthened. This helps in correcting valgus of the calcaneus and implanting the
endorthesis. The authors recommend percutaneous lengthening. Two 5-mm-long
incisions are performed on the lateral border of the tendon (at 2 and 6 cm from its
insertion in the calcaneus) and one 5-mm-long incision on the medial border (4 cm
from the insertion). Through each incision, the tendon is divided transversely
approximately half of its width. The foot is dorsiflexed so that the tendon is lengthened
until about 15 of dorsiflexion is obtained while the knee is extended.
FDL transfer is made through a medial incision along the entire length of the
PTT. The sheath of the PTT is opened and the tendon is inspected. A tenodesis is
recommended between the PTT and FDL when the PTT is healthy. FDL transfer is
performed if the PTT is in poor condition. The spring ligament should be repaired
if there is a disruption. The FDL is passed through the hole in the navicular from
plantar to dorsal.14 The tendon is sutured after insertion of sinus tarsi implant. It
is important that the implant (final or trial implant) is inserted before completing the
tendon transfer. If it is not done this way, the tension of the tendon will probably
be wrong and the transfer will fail.
For implantation of the arthroereisis plug, a slightly curved 2-cm skin incision is
made on the lateral side of the hindfoot centered over the sinus tarsi, just anterior to

275

276

Fernndez de Retana et al

the tip of the lateral malleolus. Care must be taken not to damage the peroneal
tendons, the sural nerve, and the intermediate dorsal cutaneous nerve (a branch of the
superficial peroneal nerve). A direct approach to the sinus tarsi is obtained. The sinus
tarsi is debrided removing its contents: fatty tissue with abundant nerve endings. This
is important to eliminate the irritating stimuli of the endorthesis in the sinus, which
could originate the contracture of the peroneal tendons. The powerful interosseous
talocalcaneal ligament must not be damaged, because it is an important hindfoot
stabilizer.
The next step consists of restoration of the foot arch and correction of calcaneous
valgus deviation. This is achieved using a blunt lever, which is introduced through the
sinus tarsi and under the neck of the talus. The correcting procedure is performed:
The lever is pushed distally, so that the hindfoot is supinated. At the same time, the
assistant pronates the forefoot. The aim of this procedure is to move the head of the
talus upward, backward, and outward, so that it is repositioned in its physiologic
position and hindfoot pronation is corrected. It is important to bear in mind that, in
flatfoot, the talus has slipped forward, downward, and inward.
To conserve the correction obtained, an endorthesis is inserted in the sinus tarsi.
First, the trial implants (with increasing diameters) are inserted until the appropriate
implant size is determined. The authors choose the smallest implant that corrects the
deformity and remains stable in the sinus tarsi while moving the subtalar joint. It is
advisable to check the correct position of the trial implant with fluoroscopy. The
authors consider that the implant is in the correct position when the lateral edge of the
implant is aligned with the lateral border of the talus. Then, the definitive endorthesis
is implanted, the same size as the trial implant. The FDL is sutured in a slight inversion
permitting plantigrade position simulating weight bearing. Hemostasia is done after
taking off Esmarch. Closure of the wound is performed in a routine fashion and a
below-the-knee compression cast is applied. The whole procedure does not take
more than 40 to 50 minutes. If necessary, additional procedures can be associated,
including removal of accessory navicular bone, removal of tarsal coalition, and
peroneal tendon lengthening.
Postoperative management

A below-the-knee walking cast is applied and sutures are removed 2 to 3 weeks


postoperatively. A walker is placed when the skin has healed and weight-bearing is
allowed. If there is pain with weight-bearing, the patient should avoid weight-bearing.
The walker is removed 6 weeks after surgery if there is no pain with weight-bearing.
The patient is advised to wear rigid orthopedic insoles for 6 months to support the
correction.
Endorthesis

Since 1998, the authors have been using the Kalix endorthesis (Newdeal SA, Vienne,
France) for child and adult flatfoot surgical treatment (Fig. 4). This implant consists of
a metal cone trunk, which is introduced into another polyethylene cone trunk that
expands. It is manufactured with biocompatible materials: Titanium and high-density
polyethylene. Its conical shape fits well into the sinus tarsi and the lateral fins prevent
the implant from moving out of the sinus tarsi. In 2009, the authors started using the
Sinus implant (Fig. 5). It is a troncoconical device that can be screwed into the sinus
tarsi. Both implants can be visualized radiographically owing to their metallic
component.

Subtalar Arthroereisis in Adult Acquired Flatfoot

Fig. 4. Kalix endorthesis.

RESULTS

Subtalar arthroereisis has proved to yield good mid to long-term results in children.9-11,27
However, there is little scientific evidence that proves its usefulness in adults with
acquired flatfoot deformity.7 Lelievre5 treated flatfeet in children and adults inserting
a bone graft into the sinus tarsi reporting a 90% success rate. Maxwell and
associates28 described the use of the Maxwell-Brancheau arthroereisis for the
correction of PTT dysfunction having successful outcomes. Zaret and Myerson7
reported on 12 adults treated with the MBA prosthesis. Good clinical and radiographic results were obtained, although 2 patients required removal of the prosthesis
owing to persistent sinus tarsi pain; no recurrence of the deformity was observed in
these patients after prosthesis removal. Viladot and colleagues6 studied 19 patients
with stage II PTT dysfunction treated with Kalix implant and tendon repair. They
excluded overweight patients, those who had previous surgery, and neuropathic and
osteoarthritic foot. American Orthopedic Foot and Ankle Society (AOFAS) score
improved from 47.2 to 81.6 at follow-up. Only 2 patients were dissatisfied. Two
patients required implant removal. Needleman29 reported his results on 28 feet with
flatfoot deformities in adult patients (18 years old) using an MBA implant. He

Fig. 5. Sinus implant.

277

278

Fernndez de Retana et al

included neurologic flatfoot. The exclusion criteria included infection, rigid flatfoot,
evidence of osteoarthritis, and previous hindfoot surgery. Excessive weight did not
affect inclusion in this study. AOFAS score improved from 52 to 87 points, and 78%
of patients were satisfied with the result. Four patients (5 feet; 18%) said that they
would not have the surgery again. The MBA implant was removed from 11 of 28 feet
(39%) because of postoperative sinus tarsi pain.
More recently, Adelman and co-workers30 reported their results on 10 patients with
symptomatic stage IIB PTT dysfunction treated by endoscopic gastrocnemius
recession, subtalar arthroereisis with the MBA implant and FDL tendon transfer to
tarsal navicular. All patients showed significant improvement of radiologic measurements. Four patients (40%) referred pain at sinus tarsi but only 1 patient required
implant removal. The authors conclude that this surgical combination provides
satisfactory correction of stage II PTT dysfunction.
Arthroereisis has been used for treatment in children with talocalcaneal coalition
owing to bad results after interposition. The results were encouraging.31 The use in
adults needs to be determined with more studies.
DISCUSSION

Acquired flexible flatfoot in adults is a common condition that requires surgery in most
cases. If they fulfill the inclusion criteria mentioned, surgical treatment is indicated.
Numerous treatment options have been proposed and, probably, there is not a single
treatment that is appropriate for every patient. The operative procedure most
frequently used to treat adult flatfoot secondary to stage II PTT dysfunction consists
of MDCO and FDL transfer. This technique has proven to be effective for this
condition, although calcaneal osteotomies are not free from complications. Myerson
and colleagues32 published good results in 129 patients, but with these complications
directly or indirectly related to the calcaneal osteotomy: One progressive hindfoot
valgus deformity, 2 varus malunion of the osteotomy, 3 sural neuritis, and 5 numbness
in the distribution of the medial plantar nerve. Greene and associates33 studied the
relationship between the MDCO and the medial neurovascular structures in 22
cadaver specimens. They found that the osteotomy site was crossed by an average
of 4 neurovascular structures, most of them branches of the lateral plantar nerve and
the posterior tibial artery.
Subtalar arthroereisis is an alternative to calcaneal osteotomy in the treatment of
stage II PTT dysfunction, particularly in stage IIA, where valgus hindfoot is the main
deformity. These are the advantages of subtalar arthroereisis compared with MDCO:
It is easy and quick to perform; it is a less invasive procedure; there is no risk of
nonunion or malunion; there is no risk of damaging medial neurovascular structures
and the sural nerve; and it requires less immobilization time and less nonweightbearing time after surgery (shorter recovery time). Subtalar arthroereisis has a greater
potential of hindfoot valgus correction than calcaneal osteotomy: The limit for
calcaneal displacement is one third of its width approximately, whereas greater
valgus correction can be achieved with arthroereisis using increasing implant sizes.
Finally, subtalar arthroereisis provides a 3-dimensional correction of the flatfoot
deformity by repositioning the talus in its physiologic position, preventing it from
slipping forward, inward, and downward. It is not so clear that MDCO provides this
type of correction.
There are also disadvantages of subtalar arthroereisis: It limits subtalar motion,
although it is not completely blocked. The rate of pain at sinus tarsi is high (10% 40%
of patients); this pain may be temporary and may require implant removal. Another

Subtalar Arthroereisis in Adult Acquired Flatfoot

Fig. 6. (A) Preoperative lateral x-ray. (B) Postoperative lateral x-ray with Sinus subtalar
arthroereisis. (C) Lateral x-ray after Sinus implant removal. The correction continues after
removal.

disadvantage of arthroereisis is that most implants in the market are more expensive
than the 1 or 2 screws needed for calcaneal osteotomy.
Calcaneal osteotomy and subtalar arthroereisis can be performed combined on the
same patient. Schon34 performs the MDCO first and, if he cannot achieve proper heel
position with the osteotomy, he performs a subtalar arthroereisis through a different
approach to gain more heel valgus correction.34 Finally, he repairs the PTT. The
authors prefer to start with the arthroereisis because, if more correction is needed, the
authors use a bigger size implant and, in most cases, the calcaneal osteotomy is not
needed. In any case, the fact is that both techniques can be associated if needed.
Bruyn and colleagues35 had good results using a combination of Evans calcaneal
osteotomy and STA-arthroereisis. They reported 25 symptomatic severe flexible pes
valgo planus treated with a combined operative technique and patients were 100%
satisfied or very satisfied. The authors do not recommend combining subtalar
arthroereisis with a calcaneal lengthening osteotomy. In this situation, the osteotomy
is performed very close to the implant site and implant stability may be at risk.
Only minor complications have been reported with subtalar arthroereisis in adults.
Sinus tarsi pain is the most frequent complication.6-8,30,33 Different causes could
explain sinus tarsi pain: Impingement of the implant against the posterior subtalar
facet and irritation of the nerve ends at the sinus tarsi caused by the prosthesis or
implantation of an oversized prosthesis. When this complication develops, it can be

279

280

Fernndez de Retana et al

managed in different ways: Immobilization in a brace for several weeks, physiotherapy, or corticosteroid injections into the sinus tarsi. These treatments solve the
problem in most, but not all, patients. When pain persists, removal of the implant is
indicated. The question is: Can implant removal cause a relapse of the deformity?
Almost all the literature available on subtalar arthroereisis in children and adults
agrees that the possibility of recurrence is very low after implant removal. It is also our
experience that implant removal usually eliminates sinus tarsi pain without clinical or
radiologic recurrence of the flatfoot (Fig. 6). The authors think that this phenomenon
can be explained because arthroereisis acts as an internal orthotic device that keeps
the foot in the correct position while the medial soft tissues are healing. The authors
think also that some degree of arthrofibrosis generates at the subtalar joint that
prevents it from subluxating after implant removal. In our experience, the incidence of
implant intolerance with sinus tarsi pain is very frequent. When the authors recommend a subtalar arthroereisis sinus tarsi implant to treat a stage IIA PTT dysfunction,
the patient is informed about the possibility of postoperative sinus tarsi pain that may
require a second surgery for implant removal.
SUMMARY

Subtalar arthroereisis, often combined with Achilles tendon lengthening, is a simple


and effective way to treat flexible flatfoot in adults. The most common complication
is pain in sinus tarsi, which usually disappears after removal of the implant. Midterm
results are good and it does not hinder other treatments in the future.
REFERENCES

1. Noll KH. The use of orthotic devices in adult acquired flatfoot deformity. Foot Ankle
Clin North Am 2001;6:2536.
2. Chambers EF. An operation for the correction of flexible flat feet of adolescents. West
J Surg Obstet Gynecol 1946;54:603 4.
3. Grice DS. An extra-articular arthrodesis of the subastragalar joint for correction of
paralytic flat feet in Children. J Bone Joint Surg Am 1952;34:927 40.
4. Haraldsson S. Operative treatment of pes planovalgus staticus juvenilis. Acta Orthop
Scand 1962;32:492 8.
5. Lelievre J. The valgus foot: current concepts and correction. Clin Orthop 1970;70:
4355.
6. Viladot R, Pons M, lvarez F, et al. Subtalar arthroereisis for posterior tibial tendon
dysfunction: a preliminary report. Foot Ankle Int 2003;24:600 6.
7. Zaret DI, Myerson MS. Arthroereisis of the subtalar joint. Foot Ankle Clin North Am
2003;8:60517.
8. Needleman RL. Current topic review: subtalar arthroereisis for correction of flexible
flatfoot. Foot Ankle Int 2005;26:336 46.
9. Subotnick S. The subtalar joint lateral extra-articular arthroereisis: a follow-up report.
J Am Podiatry Assoc 1977;32:2733.
10. Smith SD, Millar EA. Arthroereisis by means of the subtalar polyethylene peg implant
for correction of hindfoot pronation in children. Clin Orthop 1983;181:1525.
11. Viladot A. Surgical treatment of the child=s flatfoot. Clin Orthop 1992;283:34 8.
12. Vogler H. Subtalar joint blocking operations for pathological pronation syndromes. In:
McGlamery ED, editor. Comprehensive textbook of foot surgery. Baltimore: William &
Wilkins; 1987. p. 466 82.
13. Johnson KA, Strom DE. Tibialis posterior tendon dysfunction. Clin Orthop 1989;239:
196 206.

Subtalar Arthroereisis in Adult Acquired Flatfoot

14. Myerson MS. Adult acquired flatfoot deformity. J Bone Joint Surg Am 1996;78:780 92.
15. Bluman E, Title CI, Myerson MS. Posterior tibial tendon rupture: a refined classification
system. Foot Ankle Clin North Am 2007;12:233 49.
16. Hill K, Saar WE, Lee TH, et al. Stage II flatfoot: what fails and why. Foot Ankle Clin
North Am 2003;8:91104.
17. Mann R, Thompson FM. Rupture of the posterior tibial tendon causing flat foot:
surgical treatment. J Bone Joint Surg 1985;67:556 61.
18. Kitaoka HB, Luo ZP, An KN. Reconstructive operations for acquired flatfoot: biomechanical evaluation. Foot Ankle Int 1998;19:303 4.
19. Gazdag AR, Cracchiolo A. Rupture of the posterior tibial tendon. Evaluation of injury of
the spring ligament and clinical assessment of tendon transfer and ligament repair. J
Bone Joint Surg 1997;79A:675 81.
20. Arangio GA, Reinert KL, Salathe EP. A biomechanical model of the effect of subtalar
arthroereisis on the adult flexible flat foot. Clin Biomech 2004;19:84752.
21. Saxena A, Nguyen A. Preliminary radiographic findings and sizing implications on patients
undergoing bioabsorbable subtalar arthroereisis. J Foot Ankle Surg 2007;46:17580.
22. Vedantam R, Capelli AM, Schoenecker PL. Subtalar arthroereisis for the correction of
planovalgus foot in children with neuromuscular disorders. J Pediatr Orthop 1998;18:2948.
23. Rocket AK, Mangum G, Mendicino SS. Bilateral intraosseous cystic formation in the
talus: a complication of subtalar arthroereisis. J Foot Ankle Surg 1996;37:4215.
24. Siff TE, Granberry WM. Avascular necrosis of the talus following subtalar arthroereisis
with a polyethylene endoprosthesis: a case report. Foot Ankle Int 2000;21:2479.
25. Kuwada GT, Dockery GL. Complications following traumatic incidents with STA-peg
procedures. J Foot Surg 1988;27:236 9.
26. Dalton GP, Wapner KL, Hecht PJ. Complications of Achilles and posterior tibial
tendon surgeries. Clin Orthop 2001;391:1339.
27. Giannini S. Operative treatment of the flatfoot: why and how. Foot Ankle Int 1998;19:
52 8.
28. Maxwell JR, Carro A, Sun C. Use of the Maxwell-Brancheau arthroereisis implant for
the correction of posterior tibial tendon dysfunction. Clin Podiatr Med Surg 1999;16:
479 89.
29. Needleman RL. A surgical approach for flexible flatfeet in adults including a subtalar
arthroereisis with the MBA sinus tarsi implant. Foot Int 2006;27:9 18.
30. Adelman VR, Szczepanski JA, Adelman RP. Radiographic evaluation of endoscopic
gastrocnemius recession, subtalar joint arthroereisis, and flexor tendon transfer for
surgical correction of stage II posterior tibial tendon dysfunction: a pilot study. J Foot
Ankle Surg 2008;47:400 8.
31. Giannini S, Ceccarelli F, Vannini F, et al. Operative treatment of flatfoot with talocalcaneal coalition. Clin Orthop 2003;411:178 87.
32. Myerson MS, Badekas A, Schon LC. Treatment of stage II posterior tibial tendon
deficiency with flexor digitorum tendon transfer and calcaneal osteotomy. Foot Ankle
Int 2004;25:44550.
33. Greene DL, Thompson MC, Gesink DS. Anatomic study of the medial neurovascular
structures in relation to calcaneal osteotomy. Foot Ankle Int 2001;22:569 71.
34. Schon LC. Subtalar arthroereisis: a new exploration of an old concept. Foot Ankle Clin
North Am 2007;12:329 39, vii.
35. Bruyn JM, Cerniglia MW, Chaney DM. Combination of Evans calcaneal osteotomy
and STA-peg arthroereisis for correction of severe pes valgo planus deformity. J Foot
Ankle Surg 1999;38:339 46.

281

Foot Ankle Clin N Am


12 (2007) 317327

Management of the Rigid Adult


Acquired Flatfoot Deformity
Rommel Francisco, DO, Christopher P. Chiodo, MD*,
Michael G. Wilson, MD
Brigham Foot and Ankle Center, 1153 Centre Street, Jamaica Plain, Boston, MA 02130, USA

The optimal management of the adult acquired atfoot requires careful


assessment of the deformity. Although it is essential to recognize the location and degree of malalignment, it is of equal importance to appreciate
whether the deformity is exible or rigid. For patients undergoing surgery
for a exible atfoot, various soft tissue procedures and bony osteotomies
may be used to restore alignment while preserving joint motion. However,
for patients undergoing surgery for a rigid deformity, such procedures are
indicated less often, and arthrodesis is usually necessary.

Nonoperative management
Nonoperative treatment always should be considered when treating
a rigid atfoot deformity. Many of these patients are low demand and
elderly, and they either may not require surgical intervention or may be
poor surgical candidates. Furthermore, they may not be able to safely comply with postoperative protocols that involve a period of nonweightbearing.
Accommodative bracing and orthotics are central to nonoperative management. With a rigid deformity, the prescribed orthosis generally should be
accommodative, rather than corrective, in nature [1]. Whereas a corrective
orthosis can result in skin breakdown because of the rigid nature of the deformity, an accommodative device will distribute ground forces more evenly
to the deformed foot. An Arizona brace or custom anklefoot orthosis also
may be considered [2]. However, caution must be employed with the rigid
deformity to avoid skin problems. It is prudent to alert the orthotist that
the goal of the device is to accommodate, not correct, the deformity.
E-mail address: cchiodo@partners.org (C.P. Chiodo).
1083-7515/07/$ - see front matter 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.fcl.2007.03.013
foot.theclinics.com

318

FRANCISCO

et al

Nonsteroidal anti-inammatory drugs (NSAIDs) may be a useful adjunct


for pain relief, because rigid deformities often occur in combination with
some degree of arthritis. Rather than daily dosing, the authors recommend
strategic intermittent NSAID use combined with anticipated periods of
increased activity. With prolonged NSAID use, serial screening laboratory
studies are necessary.

Surgical management
In general, correction of a rigid atfoot deformity is accomplished best
with fusion of those joints through which there is deformity and/or arthrosis. Triple arthrodesis has been the gold standard, but limited hindfoot fusions (double fusion, isolated subtalar fusion) may be possible in certain
situations. The disadvantage of any fusion procedure involves the potential
for progressive degenerative change in adjacent joints. Furthermore, with severe deformity, hindfoot fusion may not provide sucient correction. In
these instances an adjunctive procedure may be necessary.
Triple arthrodesis
Historically, triple arthrodesis of the subtalar, calcaneocuboid, and talonavicular joints has been the gold standard for the surgical correction of
a rigid atfoot deformity (Fig. 1). This procedure is strongly indicated
when there are degenerative changes in two or more of these joints and
when previous hindfoot reconstruction has failed.
In clinical literature, the results of triple arthrodesis have been studied extensively. Fortin and Walling reported the results of 32 triple arthrodeses
performed for stage III or stage IV posterior tibial tendon dysfunction
(PTTD) [3]. The average follow-up was 4.3 years. In this study, the postoperative American Orthopaedic Foot and Ankle Society (AOFAS) hindfoot
score improved by an average of 36 points. There was one nonunion and
two malunions. All patients except one were satised and reported they
would undergo the procedure again. Similarly, Sangeorzan and colleagues
[4] reviewed the results of 44 feet (nine with PTTD) with an average 4.9year follow-up. Thirty-four patients had good results. Four failures were attributed to two nonunions and two unsatisfactory corrections. Graves and
colleagues reported on 18 feet (10 with PTTD). Fourteen patients were satised with their results with all reporting decreased pain.
Recently, Pell and colleagues [5] reported a large series examining 111 patients (132 feet) who underwent triple arthrodesis. Seventy cases had a preoperative diagnosis of posterior tibial tendon (PTT) rupture or atfoot. The
average age at surgery was 54.9 years with a mean follow-up of 5.7 years.
Using a 10-point subjective satisfaction scale (0 not satised, 10 completely satised), the average patient satisfaction was 8.3. Although 79

RIGID ADULT ACQUIRED FLATFOOT DEFORMITY

319

Fig. 1. Triple arthrodesis. Preoperative (A, B) and postoperative (C, D) radiographs.

patients (60%) showed clear radiographic progression of ankle arthritis, this


nding was not signicantly correlated with patient satisfaction.
The data of Pell and colleagues continue to raise the question of whether
or not adjacent joint arthrosis will eventually compromise the long-term results of triple arthrodesis. To this end, Smith and colleagues [6] recently reported a series of 27 patients (31 feet) with a minimum follow-up of 10 years
(average follow-up of 14 years). In this series, 93% of patients were satised
with their surgical outcome. However, the mean Short-Form (SF-36) physical component outcome score was only 35.2 points (the mean score for the
United States population is 50). Twenty feet showed radiographic evidence
of degenerative changes in at least one adjacent joint, which was poorly correlated with patient reports of pain.
Surgical technique
An examination under anesthesia is performed rst to check for the presence of a gastrocnemius or Achilles tendon contracture. If either condition is
present, an appropriate surgical release is performed expeditiously before inating the tourniquet.

320

FRANCISCO

et al

For the arthrodesis, the authors recommend a two-incision approach and


believe that this allows optimal exposure of the hindfoot. Both incisions are
extensile. A lateral incision extends from the tip of the bula and extends
toward the base of the fourth metatarsal. Deep dissection is accomplished
by elevating the extensor hallucis brevis muscle o the calcaneus and debriding the contents of the sinus tarsi. On the medial side, an incision is made
from the medial malleolus to the naviuclar. Through this incision, the talonavicular joint is exposed. Sometimes it is preferrable to place this incision
immediately adjacent to the tibialis anterior tendon or even lateral to this
tendon, thus facilitating exposure of the central aspect of the talonavicular
joint.
Once the joints have been exposed, the articular cartilage is debrided with
curettes or exible chisels. Then the subchondral bone is prepared with
a low-speed bur under irrigation or, alternatively, exible chisels. With the
former, care must be taken to avoid thermal necrosis. With the latter,
a at resection of a curved articular surface can leave an undesirable gap.
All three joints are prepared before attempting mobilization/reduction.
Also, it is important to expose, debride, and decorticate the articulation between the four bones of the hindfoot. Obtaining a fusion in this space (sometimes referred to as the quadruple space of Jahss) is believed to minimize the
clinical impact of a nonunion in one of the joints being fused.
The subtalar joint is reduced and xed rst. The reduction maneuver involves internal rotation of the calcaneus on the talus. The joint is positioned
and xed in approximately 5 of valgus; then it is xed with one or two cannulated screws at least 6.5 mm in diameter. Subsequently, the talonavicular
joint is reduced and xed, followed by the calcaneocuboid joint. When reducing these joints, care must be taken to obtain a plantigrade forefoot
by derotating the forefoot out of supination. Fixation is completed according to surgeon preference and may be accomplished with screws, staples, or
even plates.
Allograft or autograft from the iliac crest or lateral proximal tibia is used
frequently and believed to enhance fusion rates. Postoperatively, patients
are immobilized and kept nonweight-bearing in a short leg cast for 69
weeks, followed by weightbearing in a walking cast or fracture boot for another 36 weeks or until there is evidence of radiographic fusion. It is important to counsel patients that, even with a successful fusion, it can take up to
1 year for postoperative pain to resolve. This is believed to correspond with
the maturation of woven bone to lamellar bone.
When correcting an advanced deformity, the surgeon may be faced with
a gap at the calcaneocuboid joint following xation of the talonavicular and
subtalar joints. This may be addressed by either shortening the medial column through aggressive debridement/joint preparation or by lengthening
the lateral column [7]. The latter is accomplished either with interpositional
bone graft or sliding graft technique, as developed and described by Schon
and colleagues [8].

RIGID ADULT ACQUIRED FLATFOOT DEFORMITY

321

In select cases with advanced deformity, a standard triple arthrodesis still


may result in suboptimal correction. One surgical option for obtaining further correction is the insertion of allograft or autograft wedges, which act as
shims, into the posterior facet of the subtalar joint. Alternatively, the subtalar component of the fusion may be combined with a medial displacement
osteotomy of the calcaneal tuberosity either at the same setting or in a staged
fashion. Finally, Horton and Olney [9] have reported the results of simultaneous triple arthrodesis with lateral column lengthening for severe planovalgus deformity. In this study of 14 patients, the lateral column was
lengthened through the calcaneocuboid joint. The average correction of
the talar-rst metatarsal angle was 25 in both anterolateral and lateral
planes. All patients fused by 12 weeks.
Isolated subtalar fusion
Isolated subtalar fusion is indicated when the majority of a rigid atfoot
deformity is comprised of hindfoot valgus and located in the subtalar joint.
The transverse tarsal joint must be nonarthritic and either be exible or have
minimal contraction/deformity. Compared with triple arthrodesis, the theoretical advantage of isolated subtalar fusion is to protect surrounding joints
from developing arthrosis when compensatory motion occurs.
When performing an isolated subtalar fusion for either a exible or rigid
atfoot deformity, it is important to consider adjunctive procedures for the
forefoot. Usually there is some degree of forefoot abduction, and a exor
digitorum longus (FDL) transfer is indicated to correct this component of
the deformity and to augment/replace PTT function (Fig. 2). Meanwhile,
forefoot varus may be corrected by a medial column bony procedure: either
an opening wedge Cotton osteotomy of the medial cuneiform [10] or a rst
tarsometatarsal arthrodesis [11].
Beyond a rigid deformity, other potential indications for a subtalar fusion in conjunction with an FDL transfer and possible medial column procedure include obesity and well-controlled inammatory arthropathy.
Kitaoka and Patzer [12] studied 21 patients treated with isolated subtalar
fusion for PTTD. The average follow-up was 3 years, with all patients
achieving successful union. Clinically, 16 patients were rated as having
good or excellent results, and 16 were satised without reservations. However, 11 patients still reported some level of pain. In this study, the investigators did not perform a concomitant soft tissue augmentation of the PTT
(eg, FDL tendon transfer.)
Stephens and colleagues [13] reported on 25 feet with mid to late stage II
disease. The average AOFAS hindfoot score increased from 38 preoperatively to 77 postoperatively. All patients healed at an average of 13.5 weeks
postoperatively. Johnson and colleagues [14] reviewed the results of 16 patients who had stage II disease. The investigators combined FDL tendon
transfer and spring ligament reeng with subtalar fusion. The average

322

FRANCISCO

et al

Fig. 2. Subtalar arthrodesis with FDL transfer. Preoperative (A, B) and postoperative (C, D)
radiographs.

postoperative AOFAS score was 82. All patients fused at an average of 10


weeks. They concluded that that this combination of procedures was an effective and reliable procedure that was able to simultaneously correct hindfoot valgus and forefoot abduction.
Surgical technique
The fusion procedure is performed through a lateral extensile incision as described earlier. A Strayer procedure or Achilles tendon lengthening is performed with a separate incision as necessary. Care is taken to protect the
peroneal tendons and sural nerve. The extensor hallucis brevis muscle is reected and the sinus tarsi debrided, which exposes the subtalar joint. Further
visualization of the joint is facilitated by use of a lamina spreader that is positioned under the talar neck. Correction is accomplished by internally rotating
the calcaneus on the talus. Further correction in advanced deformity has been
gained by inserting allograft or autograft shims into the posterior facet of
the joint. Then the joint is xed and compressed with partially threaded cannulated screws, ideally with the hindfoot in 5 of physiologic valgus. Usually,

RIGID ADULT ACQUIRED FLATFOOT DEFORMITY

323

two screws are used: one extending from the calcaneal tuberosity into the talar
body, and the other extending from the tuberosity into the talar neck.
Once the fusion is complete, attention is turned distally. In most cases, an
FDL transfer is performed with a possible Cotton osteotomy or rst tarsometatarsal arthrodesis.
Double fusion (talonavicular and calcaneocuboid joints)
Clain and Baxter [15] popularized a double arthrodesis of the talonavicular and calcaneocuboid joints (Fig. 3), citing the procedure as a simple
and eective alternative to triple arthrodesis. These investigators claimed
that while an isolated fusion of the talonavicular joint eectively blocks
hindfoot motion, the addition of the calcaneocuboid joint to the fusion
adds stability by securing the lateral column of the foot. They were careful
to point out that this procedure is not indicated for supple deformity in the
atfoot. Rather, they felt that it was indicated where foot deformity is principally at the transverse tarsal joint (as is often seen from chronic PTT rupture), .proceeding to triple arthrodesis only if the deformity proves not to
be completely correctable intraoperatively.
Clain and Baxter [15] studied 16 feet treated with double fusion for a variety of disorders. Average follow-up was 83 months with results rated as
good or excellent in 12 patients. There were no poor results. Progressive degeneration in surrounding joints was common. There was one asymptomatic
nonunion of the talonavicular joint.
Subsequently, Mann and Beaman [16] reviewed 24 double fusions (16 for
PTTD) with an average follow-up of 56 months. In the atfoot, these investigators indications for a double arthrodesis was xed forefoot varus
greater than 1015 degrees, with a subtalar joint that passively corrects to
at least the neutral position, and the patient with transverse tarsal hypermobility. Mann and Beaman stressed that subtalar degenerative change is
a contraindication to the procedure. An 83% overall satisfaction rate,

Fig. 3. Anteriorposterior plane (A) and lateral radiographs (B) demonstrating a double fusion
through the talonavicular and calcaneocuboid joints.

324

FRANCISCO

et al

with 76% good or excellent results was reported. Complications were more
frequent in patients who had PTTD. Although most patients were asymptomatic, progression of arthrosis in surrounding joints was common. The
most frequent complication was talonavicular nonunion in four patients.
In the authors experience, it is uncommon to see a rigid atfoot deformity without substantial rigid heel valgus. To this end, one of the senior authors (MGW) has successfully performed double arthrodesis through the
talonavicular and subtalar joints (Fig. 4), rather than the talonavicular and
calcaneocuboid joints. This allows for simultaneous correction of the hindfoot valgus in addition to the forefoot abduction.
Stage IV adult acquired atfoot
Patients with a long-standing rigid atfoot deformity may develop ankle
arthritis with valgus talar til and incompetence of the deltoid ligament

Fig. 4. Double arthrodesis through the talonavicular and subtalar joints. Preoperative (A, B)
and postoperative (C, D) radiographs.

RIGID ADULT ACQUIRED FLATFOOT DEFORMITY

325

complex (Fig. 5). Nonoperative treatment often is unsuccessful in this small


subset of patients.
Surgical treatment options in this patient population have involved
mainly tibiotalocalcaneal or pantalar fusion. However, in some cases,
such extensive fusions may be avoided. In patients whose ankle arthritis is
asymptomatic or minimally symptomatic, triple arthrodesis, while preserving the ankle joint, may be considered. It is important to realign the hindfoot out of valgus, thus unloading the lateral aspect of the ankle joint.
Alternatively, if the majority of symptoms are believed to be coming from
the ankle, the authors have had success with ankle arthrodesis performed
in conjunction with bracing or orthotic stabilization of the hindfoot. In
these cases, a portion of the hindfoot valgus can be corrected through the
ankle fusion. Theoretically, residual deformity also might be addressed
through extra-articular osteotomies (eg, medial calcaneal slide and/or Cotton osteotomy).
Some patients may develop a rigid hindfoot deformity below ankle valgus
that is secondary to deltoid insuciency. This can be detected by using stress
radiographs of the ankle to dierentiate between tibiotalar arthritis and instability. Surgical treatment in this instance may entail a triple arthrodesis

Fig. 5. Stage IV atfoot deformity. Anterolateral plane radiograph (A) of the ankle, and anterolateral plane (B) and lateral (C) radiographs of the foot.

326

FRANCISCO

et al

performed in conjunction with a deltoid reconstruction as described by Deland and colleagues [17] or by Bluman [18].
Finally, in select patients who have stage IV disease, one may consider
a triple arthrodesis performed in conjunction with ankle arthroplasty, although further research is necessary to delineate and determine the indications for such an approach.

Summary
The surgical management of the rigid atfoot deformity usually involves
a triple arthrodesis. However, in select patients, it may be possible to preserve some of the joints of the hindfoot. Careful preoperative assessment
of the deformity (including appropriate imaging studies to detect underlying
arthritis) is critical in this decision-making process. Whether performing
a fusion or joint sparing procedure, it also is important to consider adjuvant
procedures to achieve a plantigrade foot.

References
[1] Wapner KL, Chao W. Nonoperative treatment of posterior tibial tendon dysfunction. Clin
Orthop 1999;365:3945.
[2] Augustin JF, Lin SS, Berberian WS, et al. Nonoperative treatment of adult acquired at foot
with the Arizona brace. Foot Ankle Clin 2003;8(3):491502.
[3] Fortin PT, Walling AK. Triple arthrodesis. Clin Orthop Relat Res 1999;365:919.
[4] Sangeorzan BJ, Smith D, Veith R, et al. Triple arthrodesis using internal xation in treatment of adult foot disorders. Clin Orthop Relat Res 1993;294:299307.
[5] Pell RF, Myerson MS, Schon LC. Clinical outcome after primary triple arthrodesis. J Bone
Joint Surg Am 2000;82:4757.
[6] Smith CL, Shen W, DeWitt S, et al. Triple arthrodesis in adults with non-paralytic disease. A
minimum ten-year follow-up study. J Bone Joint Surg Am 2004;86:270713.
[7] Hintermann B, Valderrabano V, Kundert HP. Lengthening of the lateral column and reconstruction of the medial soft tissue for treatment of acquired atfoot deformity associated
with insuciency of the posterior tibial tendon. Foot Ankle Int 1999;20(10):6229.
[8] Schon LC, Acevedo JI, Mann MR. Sliding wedge local bone graft for midfoot arthrodesis.
Foot Ankle Int 1999;20(5):3401.
[9] Horton GA, Olney BW. Triple arthrodesis with lateral column lengthening for treatment of
severe planovalgus deformity. Foot Ankle Int 1995;16(7):395400.
[10] Hirose CB, Johnson JE. Plantarexion opening wedge medial cuneiform osteotomy for correction of xed forefoot varus associated with atfoot deformity. Foot Ankle Int 2004;25(8):
56874.
[11] Chi TD, Toolan BC, Sangeorzan BJ, et al. The lateral column lengthening and medial column stabilization procedures. Clin Orthop Relat Res 1999;365:8190.
[12] Kitaoka HB, Patzer GL. Subtalar arthrodesis for posterior tibial tendon dysfunction and pes
planus. Clin Orthop Relat Res 1997;345:18794.
[13] Stephens HM, Walling AK, Solmen JD, et al. Subtalar repositional arthrodesis for adult acquired atfoot. Clin Orthop Relat Res 1999;365:6973.
[14] Johnson JE, Cohen BE, DiGiovanni BF, et al. Subtalar arthrodesis with exor digitorum
longus transfer and spring ligament repair for treatment of posterior tibial tendon insuciency. Foot Ankle Int 2000;21(9):7229.

RIGID ADULT ACQUIRED FLATFOOT DEFORMITY

327

[15] Clain MR, Baxter DE. Simultaneous calcaneocuboid and talonavicular fusion. Long-term
follow-up study. J Bone Joint Surg Br 1994;76(1):1336.
[16] Mann RA, Beaman DN. Double arthrodesis in the adult. Clin Orthop Relat Res 1999;365:
7480.
[17] Deland JT, de Asla RJ, Segal A. Reconstruction of the chronically failed deltoid ligament:
a new technique. Foot Ankle Int 2004;25(11):7959.
[18] Bluman E. Minimally invasive deltoid ligament reconstruction: a comparison of three techniques. American Orthopaedic Foot and Ankle Specialty Day. Chicago (IL); 2006.

Management of the Rigid Arthritic


Flatfoot in Adults: Triple Arthrodesis
Jamal Ahmad, MD*, David Pedowitz, MS, MD

KEYWORDS
Arthritis Calcaneocuboid Double Rigid flatfoot Subtalar Talonavicular
Triple arthrodesis
KEY POINTS
The traditional surgical treatment for the rigid arthritic flatfoot (AAFD) is a triple
arthrodesis of the subtalar (ST), talonavicular (TN), and calcaneocuboid (CC) joints
through a combined lateral and medial approach.
The triple arthrodesis is highly successful in correcting deformity, alleviating pain, and
improving function.
After triple arthrodesis, patients can develop either early or late postoperative
difficulties.
Alternatives to the dual-incision triple arthrodesis include a single medial approach to
the triple arthrodesis, a double arthrodesis of the TN and CC joints, and a modified
double arthrodesis of the ST and TN joints.
Further study is needed to define the role of alternate approaches and arthrodeses in
managing the rigid, AAFD.

Appropriate management of the adult flatfoot deformity (AAFD) is highly dependent


upon its severity. Upon failure of nonsurgical treatment, patients with a flexible flatfoot
and no arthritic changes are amenable to soft-tissue reconstruction and bony osteotomies that correct deformity and preserve joints. Once adults develop a rigid, arthritic
flatfoot, however, such joint-sparing procedures are neither appropriate nor sufficient.
The traditional surgical treatment for the rigid arthritic flatfoot that has failed nonoperative
management is a triple arthrodesis of the subtalar (ST), talonavicular (TN), and calcaneocuboid (CC) joints through a combined lateral and medial approach.
Ryerson1 first described this dual-incision triple arthrodesis in 1923 as treatment of
a rigid deformity secondary to paralytic conditions. As these types of diseases

Rothman Institute Orthopaedics, Thomas Jefferson University Hospital, 925 Chestnut Street,
Philadelphia, PA 19107, USA
* Corresponding author.
E-mail address: jamal.ahmad@rothmaninstitute.com
Foot Ankle Clin N Am 17 (2012) 309 322
http://dx.doi.org/10.1016/j.fcl.2012.03.008
1083-7515/12/$ see front matter 2012 Elsevier Inc. All rights reserved.

foot.theclinics.com

310

Ahmad & Pedowitz

became less common and more of a historical significance, the triple arthrodesis was
adapted to address the rigid arthritic adult flatfoot.2,3 To date, this procedure is highly
successful in correcting deformity and relieving both mechanical and arthritic pain.4,5
However, the triple arthrodesis is not without shortcomings. Early and late postoperative
problems include lateral wound complications, malunion, nonunion, and adjacent joint
arthritis.6-9 Several modifications in the traditional technique of triple arthrodesis have
recently been proposed to overcome these difficulties. Such alterations include the use
of a single medial incision, exclusion of the ST or CC joint, and greater inclusion of the
medial column of the midfoot.10-12
TRIPLE ARTHRODESIS
Surgical Indications

The triple arthrodesis is indicated as the traditional surgical treatment for the rigid,
arthritic AAFD.13 Whether owing to posttraumatic arthritis or advanced posterior tibial
tendon dysfunction, it is ideally suited for addressing both deformity and arthritis of
the ST, TN, and CC joints. This procedure is contraindicated in patients with a poor
soft-tissue envelope or vascular status of the hindfoot that would endanger wound
healing.
Preoperative Planning

Preoperative planning for the triple arthrodesis begins with a thorough history taking
and physical examination of the patient. Special attention should be paid to patient
factors that can affect bone and wound healing. These include medical conditions
such as diabetes and inflammatory arthropathy, medications such as steroids and
non-steroidal anti-inflammatory drugs, and social factors like tobacco usage. Upon
patient examination, the rigidity of the flatfoot must be confirmed. Particular attention
should be paid to the alignment of the hindfoot, whether there is an equinus
contracture of the ankle, and if rigid forefoot varus is a contributing factor in the
deformity. In addition, the patients soft-tissue and vascular condition should be
assessed and deemed appropriate for surgical treatment.
Weight-bearing radiographs of the foot in the anteroposterior, lateral, and
oblique planes are most important in planning a triple arthrodesis. Weight-bearing
films allow for more accurate evaluation of malalignment and loss of joint space.14
Close attention should be paid to the apex of the flatfoot deformity on the lateral
radiograph. Arch collapse can develop at the TN, naviculocuneiform (NC),
tarsometatarsal joints, or any combination thereof. To provide the most appropriate surgical treatment, the surgeon should be aware of the precise location of the
patients deformity. It is also important to assess the ankle for its possible
involvement in the rigid flatfoot with weight-bearing radiographs in the anteroposterior, lateral, and mortise planes. If the ankle also displays rigid deformity and/or
arthritis, then a triple arthrodesis is insufficient for complete treatment.15 Rather,
the patient may require an arthrodesis that includes the ankle joint, which is
outside the scope of this chapter (Fig. 1).16,17
The alignment of the whole leg needs to be assessed because this may affect the
final positioning of the proposed triple arthrodesis. If there is a history of prior limb
trauma, full-length limb radiographs may be required. At the very least, the tibial
alignment should be ascertained in all cases.
Traditional Operative Technique

Currently, the traditional surgical exposure for a triple arthrodesis is a combined


lateral and medial approach.18 The lateral incision begins at the distal end of the

Triple Arthrodesis

Fig. 1. (A, B) Preoperative radiographs of an adult patient with a rigid flatfoot deformity.
Note the degenerative changes at the ST, TN, and CC joints that accompany the planovalgus
deformity.

lateral malleolus and extends to the base of the fourth metatarsal. The muscle belly of
the extensor digitorum brevis is isolated superior to the peroneal tendons and
elevated to visualize the CC joint. The sinus tarsi is thoroughly debrided at the
proximal end of this incision to expose the ST joint. The medial incision begins at the
medial malleolus and extends to the NC joint. A plane is created between the distal
PTT and anterior tibial tendons to visualize the TN joint. There are no major vulnerable
neurovascular structures that cross this plane and it is a relatively safe exposure.
Once all 3 joints have been exposed, their osteophytes and diseased articular
cartilage are removed to expose subchondral bone. Various instruments that can be
used for this purpose include a micro-oscillating sagittal saw, chisels, and/or curettes.
If curettes are to be used, drilling of the cartilage with a guide pin may be necessary
to fenestrate and debride particularly sclerotic areas. Care should be taken to avoid
excess bony resection and shortening of the foot.
Once the joints are properly prepared, they can be mobilized sequentially to
correct the patients rigid flatfoot deformity. This begins with the ST joint as it is
reduced and fixed in 0 to 5 of valgus. ST fixation is achieved with 1 or 2 partially
threaded, cannulated cancellous screws of at least 6.5 mm in diameter.19 These
can be placed either starting from plantar calcaneus through the talar body or vice
versa.20 It is important to note that there is no current consensus among surgeons
with regard to the specific number, size, and direction of screws to use.21,22 The
first and second authors preferences are to respectively use a single 6.5- or
7.0-mm partially threaded cannulated screw from the posterior, inferior calcaneus
toward the central talar body.
The TN joint is reduced next with combined inversion of the transverse tarsal joints
and plantarflexion of the first ray. This specific realignment maneuver restores the
windlass mechanism and arch shape of the foot.23,24 A wide variety of materials have

311

312

Ahmad & Pedowitz

been utilized to achieve TN arthrodesis. Such devices include cannulated compression screws, staples, plate and screw constructs, and a combination thereof.25-27 The
difficulty with TN arthrodesis fixation in particular is 2-fold. The apposing surfaces are
curved in both the coronal and sagittal plane, making preparation arduous. In
addition, the starting point for most screws is the tuberosity, which is far medial from
the central axis of the joint and makes compression of the more central and lateral
joint surfaces difficult to achieve. There are a few biomechanical studies that compare
different types of TN implants, but no single device has a proven advantage over
another.28 The first authors preference is to use two, 4.5-mm, partially threaded,
cannulated cancellous screws in a distal-to-proximal direction. The second author
routinely uses a compression staple plate across the TN joint dorsally and additional
fixation with a single cannulated screw starting in the tuberosity passed into the talar
body.
Finally, any flatfoot deformity that remains is corrected through arthrodesis of the
CC joint. Upon ST and TN realignment and fusion, the space at the CC joint is
carefully assessed. If there is minimal persistent flatfoot deformity and/or gapping at
the CC joint, it can be fused in situ. However, this cannot be done if there is significant
residual deformity and/or space at the joint. Rather, the CC joint must be addressed
through an arthrodesis with interpositional bone graft to provide simultaneous bony
apposition, lateral column lengthening, and complete deformity correction.29 In
situations where the CC joint requires this type of distraction arthrodesis, consideration should be given to performing this before addressing the TN joint. Akin to the TN
joint, CC arthrodesis can be achieved with a wide variety of devices. Such implants
include screws, staples, or a plate and screw construct.30,31 To date, different
implants have only been compared with regard to providing a distraction CC
arthrodesis. Kimball and associates32 found a plate and screw construct to have a
significantly higher stiffness and load to failure than cortical screws in a biomechanical study. The first authors preference is to use compression staples or a 4-hole
lateral column lengthening plate to perform an in situ or distraction CC arthrodesis
respectively. The second author routinely uses a compression staple plate for the CC
joint (Fig. 2).
After the triple arthrodesis is complete, the foot should be assessed for any
persistent deformity. If patients have contracture of the gastrocnemius muscle or
Achilles tendon, an additional Strayer recession or Achilles lengthening is performed,
respectively.33 Residual forefoot varus can be managed either with an openingwedge Cotton osteotomy of the medial cuneiform or a tarsometatarsal arthrodesis.34,35 The choice of midfoot procedures is dependent upon midfoot rigidity and/or
arthritis and is outside the scope of this article.
The use of adjuvant bone graft during a triple arthrodesis remains controversial. For
patients with minimal deformity, the subchondral bone requires minimal resection and
graft may not be necessary to achieve bony apposition at any of the 3 joints.
Rosenfeld and co-workers36 performed 100 triple arthrodeses with local bone graft
from resected subchondral bone. With a reported nonunion rate as low as 4%, they
conclude that bone grafting is not needed in most situations. However, it is important
to recognize that the lateral column of the foot can shorten with a severe enough
deformity.37 Treating this situation requires bone grafting to lengthen the lateral
column, completely correct the deformity, and achieve a plantigrade foot. Graft
options vary between morcellized and structural wedge allografts and autografts.38
The most commonly used sources of autograft are the iliac crest, proximal tibia, and
distal tibia.39-41 To date, there are limited studies that compare the use of allograft,
autograft, and no bone graft when a triple arthrodesis is performed. McGarvey and

Triple Arthrodesis

Fig. 2. (A, B) Postoperative radiographs of a patient who underwent a triple arthrodesis.


Prior deformity has been corrected through careful joint preparation and rigid internal
fixation.

Braly42 performed a nonrandomized, retrospective study on 24 and 17 patients who


received iliac crest autografts and allografts with their triple arthrodesis, respectively.42 More nonunions were seen in the allograft (3 of 24 [12.5%]) than the autograft
(1 of 17 [5.9%]) population. However, this difference was not found to be significant.
For primary arthrodesis, the first and second authors use morcellized and structural
allograft for minimal and severe deformities, respectively. Both authors reserve iliac
crest autograft for revision arthrodesis. In the event that a patient may require
treatment of a future ankle condition, some consideration should be given to avoiding
harvesting bone graft from the distal tibia.
Postoperatively, patients are immobilized and rendered nonweight-bearing until
bony healing becomes evident on radiographs. This typically occurs at 6 to 8 weeks
from surgery.43 At this point, patients are allowed progressive weight-bearing in a
fracture boot or walking cast until bony fusion is complete. The time to achieve full
healing is typically between 3 and 6 months.44
Although there is much literature with long-term follow-up regarding the
traditional 2-incision triple arthrodesis, it is important to distinguish early from
recent studies. Early research described outcomes in which the procedure was
performed with absent or minimal internal fixation. Saltzman and associates45
examined long-term outcomes in 57 patients with 67 triple arthrodeses. Whereas
most patients had significant pain improvement, 78% had some recurrence of
deformity and 19% developed a nonunion of the fusion, with the TN joint affected
the most.
Over the past decade, more literature has been published that examines operative
outcomes with the use of modern, rigid internal fixation. These more recent studies
predictably describe better results. In one of the larger and more current studies, Pell

313

314

Ahmad & Pedowitz

Fig. 3. Patient who developed a deep lateral wound dehiscence after a triple arthrodesis
with a distraction CC fusion. The wound healed with surgical debridement and vacuumassisted closure.

and colleagues4 conducted medium and long-term follow-up on 111 patients with a
rigid, arthritic flatfoot. They reported rates of maintaining deformity correction, patient
satisfaction, and bony union to be higher than 90%. More recently, Czurda and
associates46 performed medium-term follow-up on a smaller population of 20
patients. Although 25% of their patients reported some residual pain, the authors
observed that all patients had improvement in pain, function, and deformity with no
occurrence of nonunion. Child and colleagues47 also performed a medium-term study
in 24 patients, but focused on correlations between correction of radiographic
deformity and change in symptoms. They found the triple arthrodesis to be highly
reliable and reproducible in correcting a rigid flatfoot deformity. They also found
improvement in symptoms to be greatest when the flatfoot was realigned closest to
a normal shape.
With regard to adverse events after a triple arthrodesis, it is important to distinguish
between those that occur in the short term versus the long term. The most common
acute or subacute problems that have been reported are wound complications,
malunion, and nonunion. Wound healing problems have been noted in 2% to 25% of
patients, with the higher range of this incidence occurring in patients with rheumatoid
arthritis. The majority of reported wound complications is superficial and affects the
lateral incision. The lateral wound is at greater risk for problems, because its skin and
soft tissue may be placed under tension with flatfoot deformity correction as the
midfoot is adducted to correct the common midfoot abduction deformity. Malunion
has been reported to occur in as many as 4% of patients. Maenpaa and associates6
found that half of malunited patients had undercorrection, whereas the other half
displayed overcorrection of deformity. Although all of these patients had rheumatoid
arthritis, the authors attributed most of these malunions to be owing to surgeon errors
and the inherent complexity of the procedure. Among all short-term complications,
nonunion after arthrodesis is the most reported. Observed rates of nonunion range
between 2 and 13%, with the TN joint being affected the most (Fig. 3).48
Even after the triple arthrodesis is fully healed, patients may develop dysfunction in
the long term. Several authors make note of significant limitations with walking

Triple Arthrodesis

inclines, accommodating uneven ground, and stair climbing.49 Formal gait studies
after triple arthrodesis show decreased ankle dorsiflexion, cadence, and stride with
decreased midfoot motion. Gait and function are further affected if patients develop
adjacent joint arthritis. Smith and co-workers5 followed patients for 15 years after
surgery and found that 27% of patients with a triple arthrodesis had developed
symptomatic adjacent joint arthritis in the ankle, NC, and/or tarsometatarsal joints. De
Groot and associates50 examined only radiographs after triple arthrodesis and found
adjacent joint arthritis in 47% of patients at 10-year follow-up. This occurs likely
because of the increased demands on the foot and ankle with the ST, TN, and CC
joints fused. All these problems can worsen if a triple arthrodesis is performed on
patients with bilateral rigid flatfeet. To date, there is scant literature regarding how
patients function with a bilateral triple arthrodesis. However, it is logical to reason that
a bilateral triple arthrodesis can be as limiting to a patient, if not more so, than a
unilateral fusion. To date, many authors have acknowledged surgical outcomes that
improve within the intermediate term and then deteriorate by long-term follow-up.51
Raikin52 described the triple arthrodesis as a salvage procedure in which gait
dysfunction and adjacent joint arthritis should be considered an expected sequelae
rather than a failure of the procedure. Although the triple arthrodesis is a valid
treatment option for the rigid and/or arthritic flatfoot, its risks and consequences
cannot be ignored.
Single Incision Triple Arthrodesis

In an effort to minimize wound complications from the traditional dual-incision


surgical approach, some surgeons have attempted the triple arthrodesis through a
single incision. Bono and co-workers53 utilized a single lateral incision in a cadaver
model to determine the amount of cartilage that could be removed from the ST, TN,
and CC joints. They report that, although they were able to debride cartilage and
subchondral bone from 80% of the ST and 90% of the CC joint, only 38% of cartilage
could be removed from the TN joint. The authors attributed the difficulty with
preparing the TN joint to inadequate visualization of the talar head. They recommended performing the TN arthrodesis through a medial incision and concluded that
an appropriate triple arthrodesis could not be conducted through a single lateral
approach.
After such cadaveric studies, the single incision approach to the triple arthrodesis
has since been attempted only from the medial side. Akin to Bono and associates,53
Jeng and colleagues54 evaluated the quantity of articular cartilage that could be
debrided from the ST, TN, and CC joints through a single medial incision in cadavers.
To obtain both medial and lateral joint exposure from a single medial incision, the authors
admit to making some modifications in technique from the medial approach of a
traditional triple arthrodesis. Before starting the medial incision, they percutaneously
released the peroneal tendons to relieve longstanding contracture from the hindfoot
valgus component of deformity. Akin to the traditional triple arthrodesis, the medial
incision that Jeng and co-workers make is still from the medial malleolus to the NC joint to
expose the TN joint. However, the authors perform increased soft-tissue dissection to
visualize the ST joint. Then, they increasingly distracted the TN joint to expose the CC
joint. With this technique, they report that they were able to expose and remove greater
than or equal to 90% of articular cartilage from all 3 joints. Unlike a single lateral surgical
exposure, they conclude that the medial approach allows adequate and comparable
access to all 3 joints for a triple arthrodesis.
Although these results are promising, there are limited clinical studies regarding a
single medial incision to perform a triple arthrodesis. After their cadaveric work, Jeng

315

316

Ahmad & Pedowitz

and co-workers55 performed 17 triple arthrodeses through a single medial incision


with final follow-up at an average of 3.5 years postoperatively. No patients developed
wound healing problems. Although all patients achieved ST and TN arthrodesis, CC
fusion was incomplete in 2 of the 17 patients (both with an asymptomatic pseudoarthrosis). It is important to recognize that using only a medial incision may have
limited the authors ability to fuse the CC joint. Although they were able to place ST
and TN implants through the medial incision directly, CC implants had to be limited to
screws that were placed percutaneously. Despite this shortcoming, Jeng and
associates concluded that the medial approach to triple arthrodesis was a reliable
technique with outcomes comparable to the standard dual-incision triple arthrodesis.
Although this single incision for a triple arthrodesis does avoid lateral wound
complications and may hold promise as an alternative technique, additional data are
required to determine its future role in treatment.
DOUBLE ARTHRODESIS (TN AND CC)

As stated, the triple arthrodesis has been the traditional surgical treatment for the
rigid, arthritic AAFD despite observed long-term complications. Certainly, if the ST,
TN, and CC joints are all arthritic, a triple arthrodesis cannot be avoided. Over the past
decade, there has been interest in fusing fewer joints to attain the same deformity
correction while avoiding problems like adjacent joint arthritis. If the AAFD is rigid but
1 joint can be spared owing to minimal arthritic changes, a double arthrodesis may be
a viable treatment option. To date, there has been less certainty in the historical
literature and common jargon as to which joints are involved when referring to a
double arthrodesis, or double as it is often shortened. In 1959, DuVries defined a
double arthrodesis as a fusion of the TN and CC joints.56 In more recent situations,
a double has been used to describe a ST and TN arthrodesis.57 With these
circumstances in mind, the term modified double is used in this chapter for
clarification of an ST and TN arthrodesis.
Compelling biomechanical data suggest that a double arthrodesis can effectively
treat the rigid, arthritic AAFD when there is minimal ST arthritis. In a cadaver model,
OMalley and co-workers58 assessed the ability to correct deformity with selective
hindfoot arthrodeses. They found that although the ST, TN, and CC joints are
interconnected, they differ in their contribution to deformity. In their study, an ST
fusion did not achieve consistent deformity correction. However, they found that
either a TN or double arthrodesis restored alignment in a flatfoot with considerable
laxity through the transverse tarsal joints. Astion and colleagues59 also examined the
ST, TN, and CC joints in cadaveric specimens, but used a 3-dimensional magnetic
space tracking system. They found that any combination of arthrodeses that included
the TN joint limited motion at the remaining joint or joints to approximately 2 of
motion.59 In contrast, they found that exclusion of the TN joint from the arthrodesis
resulted in considerable motion through the hindfoot. They concluded that the TN
joint was the key to the triple joint complex and its fusion essentially eliminated
motion at the other 2 joints. Using an ultrasonic motion system, similar results were
obtained by Wlker and associates,60 who found that ST motion was decreased by
approximately 75% with TN fusion.
Clinical studies further support the role of a double arthrodesis in certain patients
with rigid AAFD. Mann and Beaman61 performed a 4-year follow-up on patients with
double arthrodeses and found high rates of satisfaction (83%) and symptom relief
(76%). The most common complication seen in this study was nonunion of the TN
joint. Beischer and colleagues62 performed a comparative study on patients who
underwent either a triple or double at longer than 5 years postoperatively. They found

Triple Arthrodesis

Fig. 4. (A, B) Patient with a double TN and CC arthrodesis. Note that there are no degenerative changes at the ST joint.

that both the triple or double arthrodesis were similar in providing decreased pain and
increased function. Although this literature holds promise, it is important to realize that
there are currently no studies with long-term follow-up of the double arthrodesis. It
remains to be seen if this procedure can avoid some of the long-term dysfunction
seen after a triple arthrodesis.
Both authors experience with the double arthrodesis is similar to that of Mann and
Beischer. Surgical indications are for patients with advanced flatfoot deformity
without arthritic change at the ST joint. Both authors perform this through a
dual-incision technique similar to the traditional triple arthrodesis. The fusion itself is
achieved using compression staple type plates and additional interfragmentary
compression screws across the arthrodesis site. If needed, this is combined with an
Achilles lengthening or Strayer recession. Neither author routinely performs isolated
TN arthrodeses because of the high biomechanical demands placed on that joint
when correcting deformity.63 Instead, the CC joint is added to the arthrodesis for
further stability (Fig. 4).
MODIFIED DOUBLE ARTHRODESIS (ST AND TN)

When the rigid, arthritic AAFD includes ST arthritis, there is no debate that the ST joint
should be included in the arthrodesis. However, if the CC joint has minimal
degenerative changes, a modified double arthrodesis may offer a viable treatment
option. The modified double can be performed either through combined medial and
lateral incisions or a single medial approach. The dual-incision technique is the same
as that used in a standard triple arthrodesis, with the exception that the lateral incision
does not need to be extended distally to expose the CC joint. When performed

317

318

Ahmad & Pedowitz

through a single medial incision, the technique is similar to that described above for
the single medial incision triple arthrodesis, with the exception that the CC joint need
not be exposed.
Although literature is limited, the modified double arthrodesis has found success
with both pain relief and deformity correction for certain patients with rigid AAFD.
Knupp and associates64 evaluated 32 modified double arthrodeses through a single
medial approach at an average of 21 months and found a complete fusion in all
patients at 13 weeks. Brilhault65 reviewed 11 patients who underwent this procedure,
performed primarily because they had a deficient or concerning lateral skin flap. At an
average of 21.5 months postoperatively, he found that all patients had improvement
in hindfoot alignment, pain, ambulation, and footwear requirements.65 Sammarco and
associates66 achieved similar results in 16 patients at 18 months after a modified
double arthrodesis. All patients were highly satisfied with improvements in pain,
function, deformity, and return to desired shoe wear. All of these authors concluded
that the modified double arthrodesis is a reasonable option for treating the rigid,
arthritic AAFD in which the CC joint is relatively spared.
Although recent studies offer promise, it is important to realize that there are
currently no studies with long-term follow-up of the modified double arthrodesis.
Many surgeons who perform this procedure claim that it has several advantages
over a triple arthrodesis. Because the CC joint is excluded from the fusion, the
modified double arthrodesis has shorter operative times and eliminates any
complications that could be attributed to a CC fusion such as nonunion. In
preserving CC joint motion at the lateral column, it has been argued that patients
should develop less long-term adjacent joint disease with a modified double than
a triple arthrodesis. However, this particular advantage remains theoretical,
because its supporting evidence is currently anecdotal. To date, there are no

Fig. 5. (A, B) Patient with a double ST and TN arthrodesis. Note that there are no degenerative changes at the CC joint.

Triple Arthrodesis

studies that compare outcomes of the modified double and triple arthrodesis. It
remains to be seen if the modified double arthrodesis can avoid some of the
long-term dysfunction seen after a triple arthrodesis (Fig. 5).
SUMMARY

The traditional surgical treatment for adults with a rigid, arthritic flatfoot is a
dual-incision triple arthrodesis. Over time, this procedure has proved to be reliable
and reproducible in obtaining successful deformity correction through fusion and
good clinical results. However, the traditional dual-incision triple arthrodesis is not
without shortcomings. Early complications include lateral wound problems, malunion,
and nonunion. Long-term follow-up of patients after a triple arthrodesis has shown
that many develop adjacent joint arthritis at the ankle or midfoot. This particular
problem should be considered an expected consequence, rather than a failure of the
procedure. Although the indications for and surgical techniques used in triple
arthrodesis have evolved and improved with time (predictably improving results in the
intermediate term), the triple arthrodesis should be regarded as a salvage procedure.
Certain measures can be taken by the surgeon to avoid some problems. If patients
are at risk for lateral wound complications, the arthrodesis could be performed
through a single medial incision. However, this can make some aspects of the CC
fusion more difficult. Implants would have to be inserted percutaneously, which
prevents the surgeon from using either staples or plates. If a patient were to need a
lateral column lengthening through a CC distraction fusion, this would not be possible
medially. If either the ST or CC joints have minimal degenerative changes, they could
be spared through a double or modified double arthrodesis, respectively. Although
these procedures that deviate from the traditional triple arthrodesis offer promise,
further study is required to better define their role in treatment of the rigid, arthritic
AAFD.
Triple arthrodesis is, by no means, a simple surgery. It requires preoperative
planning, meticulous preparation of bony surfaces, cognizance of hindfoot positioning, and rigidity of fixation. The procedure also requires enough experience on the
part of the operating surgeon to anticipate postoperative problems and provide
modifications in traditional technique for certain patients.
REFERENCES

1. Ryerson E. Arthrodesing operations on the feet. J Bone Joint Surg Am 1923;5:


45371.
2. Wilson FC Jr, Fay GF, Lamotte P, et al. Triple arthrodesis: a study of the factors
affecting fusion after three hundred and one procedures. J Bone Joint Surg Am
1965;47:340 8.
3. Duncan JW, Lovell WW. Hoke triple arthrodesis. J Bone Joint Surg Am 1978;60:
795 8.
4. Pell RF, Myerson MS, Schon LC. Clinical outcome after primary triple arthrodesis. J
Bone Joint Surg Am 2000;82:4757.
5. Smith RW, Shen W, DeWitt S, et al. Triple arthrodesis in adults with non-paralytic
disease. J Bone Joint Surg Am 2004;86:270713.
6. Maenpaa H, Lehto MU, Belt EA. What went wrong in triple arthrodesis? An analysis of
failures in 21 patients. Clin Orthop Relat Res 2001;391:218 23.
7. Figgie MP, OMalley MJ, Ranawat C, et al. Triple arthrodesis in rheumatoid arthritis.
Clin Orthop Relat Res 1993:292;250 4.
8. Knupp M, Skoog A, Tornkvist H, et al. Triple arthrodesis in rheumatoid arthritis. Foot
Ankle Int 2008;29:2937.

319

320

Ahmad & Pedowitz

9. Doayan A, Albayrak M, Uayur F, et al. Triple arthrodesis in rigid foot deformities and the
effect of internal fixation on clinical and radiographic results. Acta Orthop Traumatol
Turc 2006;40:200 27.
10. Jeng CL, Vora AM, Myerson MS. The medial approach to triple arthrodesis. Indications and Technique for management of rigid valgus deformities in high-risk patients.
Foot Ankle Clin North Am 2005;10:51521.
11. Weinraub GM, Schuberth JM, Lee M, et al. Isolated medial incisional approach to
subtalar and talonavicular arthrodesis. J Foot Ankle Surg 2010;49:326 30.
12. Barg A, Brunner S, Zwicky L, et al. Subtalar and naviculocuneiform fusion for
extended breakdown of the medial arch. Foot Ankle Clin North Am 2011;16:69 81.
13. Bennett GL, Graham CE, Mauldin DM. Triple arthrodesis in adults. Foot Ankle Int
1991;12:138 43.
14. Wapner KL. Triple arthrodesis in adults. J Am Acad Orthop Surg 1998;6:188 96.
15. Bohay DR, Anderson JG. Stage IV posterior tibial tendon insufficiency: the tilted ankle.
Foot Ankle Clin North Am 2006;12:619 36.
16. Acosta R, Ushiba J, Cracchiolo A. The results of a primary and staged pantalar
arthrodesis and tibiotalocalcaneal arthrodesis in adult patients. Foot Ankle Int 2000;
21:18294.
17. Kelly I, Nunley J. Treatment of stage 4 adult acquired flatfoot. Foot Ankle Clin North
Am 2001;6:16778.
18. Johnson JE, Yu JR. Arthrodesis techniques in the management of stage II and III
acquired adult flatfoot deformity. J Bone Joint Surg Am 2005;87:1866 76.
19. Easley ME, Trnka HJ, Schon LC, et al. Isolated subtalar arthrodesis. J Bone Joint Surg
Am 2000;82:61324.
20. McGlamry MC, Robitaille MF. Analysis of screw pullout strength: a function of screw
orientation in subtalar joint arthrodesis. J Foot Ankle Surg 2004;43:277 84.
21. Gosch C, Verrette R, Lindsey DP, et al. Comparison of initial compression force
across the subtalar joint by two different screw fixation techniques. J Foot Ankle Surg
2006;45:168 73.
22. Lee JY, Lee JS. Optimal double screw configuration for subtalar arthrodesis: a finite
element analysis. Knee Surg Sports Traumatol Arthrosc 2011;19:8429.
23. Richie Jr DH. Biomechanics and clinical analysis of the adult acquired flatfoot. Clin
Podiatr Med Surg 2007;24:617 44.
24. Giza E, Cush G, Schon LC. The flexible flatfoot in the adult. Foot Ankle Clin North Am
2007;12:25171.
25. Harper MC. Talonavicular arthrodesis for the acquired flatfoot in the adult. Clin Orthop
Relat Res 1999;365:65 8.
26. Chen CH, Huang PJ, Chen TB, et al. Isolated talonavicular arthrodesis for talonavicular arthrodesis. Foot Ankle Int 2001;22:633 8.
27. Weinraub GM, Heilala MA. Isolated talonavicular arthrodesis for adult onset flatfoot
deformity/posterior tibial tendon dysfunction. Clin Podiatr Med Surg 2007;24:74552.
28. Jarrell III SE, Owen JR, Wayne JS, et al. Biomechanical comparison of screw versus
plate/screw construct for talonavicular fusion. Foot Ankle Int 2009;30:150 6.
29. Horton GA, Olney BW. Triple arthrodesis with lateral column lengthening for treatment
of severe planovalgus deformity. Foot Ankle Int 1995;16:395 400.
30. Kann JN, Parks BG, Schon LC. Biomechanical evaluation of two different screw
positions for fusion of the calcaneocuboid joint. Foot Ankle Int 1999;20:33 6.
31. Mereau TM, Ford TC. Nitinol compression staples for bone fixation in foot surgery. J
Am Podiatr Med Assoc 2006;96:102 6.

Triple Arthrodesis

32. Kimball HL, Aronow MS, Sullivan RJ, et al. Biomechanical evaluation of calcaneocuboid distraction arthrodesis: a cadaver study of two different fixation methods. Foot
Ankle Int 2000;21:845 8.
33. DiGiovanni CW, Langer P. The role of isolated gastrocnemius and combined Achilles
contractures in the flatfoot. Foot Ankle Clin North Am 2007;12:36379.
34. Hirose CB, Johnson JE. Plantarflexion opening wedge medial cuneiform osteotomy
for correction of fixed forefoot varus associated with flatfoot deformity. Foot Ankle Int
2004;28:568 74.
35. Thompson IM, Bohay DR, Anderson JG. Fusion rate of first tarsometatarsal arthrodesis in the modified Lapidus procedure and flatfoot reconstruction. Foot Ankle Int
2005;26:698 703.
36. Rosenfeld PF, Budgen SA, Saxby TS. Triple arthrodesis: is bone grafting necessary?
The results in 100 consecutive cases. J Bone Joint Surg Br 2005;87:175 8.
37. Sands AK, Tansey JP. Lateral column lengthening. Foot Ankle Clin North Am
2007;12:301 8.
38. Neufeld SK, Uribe J, Myerson MS. Use of structural allograft to compensate for bone
loss in arthrodesis of the foot and ankle. Foot Ankle Clin North Am 2002;7:117.
39. Fitzgibbons TC, Hawks MA, McMullen ST, et al. Bone grafting in surgery about the
foot and ankle: indications and techniques. J Am Acad Orthop Surg 2011;19:11220.
40. Whitehouse MR, Lankester BJ, Winson IG, et al. Bone graft harvest from the proximal
tibia in foot and ankle arthrodesis surgery. Foot Ankle Int 2006;27:913 6.
41. Danzinger MB, Abdo RV, Decker JE. Distal tibia bone graft for arthrodesis of the foot
and ankle. Foot Ankle Int 1995;16:18790.
42. McGarvey WC, Braly WG. Bone graft in hindfoot arthrodesis: allograft vs autograft.
Orthopedics 1996;19:389 94.
43. Francisco R, Chiodo CP, Wilson MG. Management of the rigid adult acquired flatfoot
deformity. Foot Ankle Clin North Am 2007;12:31727.
44. Knupp M, Stufkens SAS, Hintermann B. Triple arthrodesis. Foot Ankle Clin North Am
2011;16:617.
45. Saltzman CL, Fehrle MJ, Cooper RR, et al. Triple arthrodesis: twenty-five and
forty-four year average follow-up of the same patients. J Bone Joint Surg Am
1999;81:1391 402.
46. Czurda T, Seidl M, Seiser AS, et al. Triple arthrodesis in treatment of degenerative
hindfoot deformities: clinical, radiological and pedobarographic results. Z Orthop
Unfall 2009;147:356 61.
47. Child BJ, Hix J, Catanzariti AR, et al. The effect of hindfoot realignment in triple
arthrodesis. J Foot Ankle Surg 2009;48:28593.
48. Jarde O, Abiraad G, Gabrion A, et al. Triple arthrodesis in the management of acquired
flatfoot deformity in the adult secondary to posterior tibial tendon dysfunction: a
retrospective study of 20 cases. Acta Orthop Belg 2002;68:56 62.
49. Wu WL, Huang PJ, Lin CJ, et al. Lower extremity kinematics and kinetics during level
walking and stair climbing in subjects with triple arthrodesis of subtalar fusion. Gait
Posture 2005;21:26370.
50. De Groot IB, Reijman M, Lunig HA, et al. Long-term results after a triple arthrodesis of
the hindfoot: function and satisfaction in 36 patients. Int Orthop 2008;32:237 41.
51. Daglar B, Deveci A, Delialioglu OM, et al. Results of triple arthrodesis: effect of primary
etiology. J Orthop Sci 2008;13:3417.
52. Raikin SM. Failure of triple arthrodesis. Foot Ankle Clin 2002;7:12133.
53. Bono JV, Jacobs RL. Triple arthrodesis through a single lateral approach: a cadaveric
experiment. Foot Ankle 1992;13:408 12.

321

322

Ahmad & Pedowitz

54. Jeng CL, Tankson CJ, Myerson MS. The single medial approach to triple arthrodesis:
a cadaver study. Foot Ankle Int 2006;27:11225.
55. Jeng CL, Vora AM, Myerson MS. The medial approach to triple arthrodesis. Indications and technique for management of rigid valgus deformities in high-risk patients.
Foot Ankle Clin North Am 2005;10:51521.
56. Inman VT, editor. DuVries surgery of the foot. 3rd edition. St. Louis: CV Mosby Co.;
1973:491-4.
57. Philippot R, Wegrzyn J, Besse JL. Arthrodesis of the subtalar and talonavicular joints
through a medial surgical approach: a series of 15 cases. Arch Orthop Trauma Surg
2010;130:599 603.
58. OMalley MJ, Deland JT, Lee KT. Selective hindfoot arthrodesis for the treatment of
adult acquired flatfoot deformity: an in vitro study. Foot Ankle Int 1995;16:4117.
59. Astion DJ, Deland JT, Otis JC, et al. Motion of the hindfoot after simulated arthrodesis.
J Bone Joint Surg Am 1997;79:241 6.
60. Wlker N, Stukenborg C, Savory KM, et al. Hindfoot motion after isolated and
combined arthrodeses: measurements in anatomic specimens. Foot Ankle Int 2000;
21:9217.
61. Mann RA, Beaman DN. Double arthrodesis in the adult: Clin Orthop Relat Res
1999;365:74 80.
62. Beischer AD, Brodsky JW, Pollo FE, et al. Functional outcome and gait analysis after
triple or double arthrodesis. Foot Ankle Int 1999;20:54553.
63. Thelen S, Rtt J, Wild M, et al. The influence of talonavicular versus double arthrodesis
on load dependent motion of the midtarsal joint. Arch Orthop Trauma Surg 2010;130:
4753.
64. Knupp M, Schuh R, Stufkens SA, et al. Subtalar and talonavicular arthrodesis through
a single medial approach for the correction of severe planovalgus deformity. J Bone
Joint Surg Br 2009;91:6125.
65. Brilhault J. Single medial approach to modified double arthrodesis in rigid flatfoot with
lateral deficient skin. Foot Ankle Int 2009;30:21 6.
66. Sammarco VJ, Magur EG, Sammarco GJ, et al. Arthrodesis of the subtalar and
talonavicular joints for correction of symptomatic hindfoot malalignment. Foot Ankle
Int 2006;27:661 6.

Management of the Rigid Arthritic


Flatfoot in the Adults
Alternatives to Triple Arthrodesis
Christopher E. Gentchos, MDa, John G. Anderson, MDb,*,
Donald R. Bohay, MDb

KEYWORDS
Triple arthrodesis Flatfoot Arthritic flatfoot
KEY POINTS
The goals of the surgical management of the acquired flatfoot deformity are to restore a
plantigrade foot that is stable in the stance phase and functional for propulsion.
Every alternative to triple arthrodesis in the rigid acquired flatfoot deformity is predicated
on limiting the patient exposure to the complication associated with triple arthrodesis.
Successful treatment is dependent on thoughtful patient evaluation and examination,
meticulous joint preparation, careful positioning with rigid fixation, and judicious use of
adjunctive procedures to achieve the goal of a plantigrade foot that functions well and is
minimally painful.

The human foot collapses in a finite number of patterns. This collapse can be driven
by injury, time, and weight in a susceptible person. Symptoms occur at various points
along the spectrum of the different patterns. But all patterns are united by the
common denominator of medial column incompetence and gradual attenuation of
the static and dynamic supports of the medial longitudinal and transverse arch in the
presence of a progressive triceps surae contracture. Regardless of the cause, it is
unclear why the supple deformity becomes rigid. After this threshold has been
passed, treatment options become limited to arthrodesis, although there are alternatives to the triple arthrodesis.
Nonarthrodesis surgical solutions to the rigid acquired flatfoot deformity are rarely
indicated and rarely performed. To the best of our knowledge, there are no authors or
published studies that have recommended joint-sparing procedures for this condition.
The author has nothing to disclose.
a
Foot & Ankle Orthopaedic Specialist, Concord Orthopaedics PA, 264 Pleasant Street, Concord,
NH 03301, USA; b Orthopaedic Associates of Michigan, PC, 1111 Leffingwell Avenue, NE, Grand
Rapids, MI 49525, USA
* Corresponding author.
E-mail address: John.Anderson@oamichigan.com
Foot Ankle Clin N Am 17 (2012) 323335
doi:10.1016/j.fcl.2012.03.009
1083-7515/12/$ see front matter 2012 Elsevier Inc. All rights reserved.

foot.theclinics.com

324

Gentchos et al

However, to contemplate the full breadth of treatment options, particularly adjunctive


procedures, surgeons must possess an understanding of the pathogenesis that leads to
the deformity and recognize the multiple individual components of that deformity that can
be addressed and potentially corrected. Additionally, one must recognize all patterns of
collapse to understand the rigid deformity.
The adult acquired flatfoot deformity, in all its various stages of presentation, has
become synonymous with dysfunction of the posterior tibial tendon. But this narrow
definition of collapse potentially misses other patterns and obscures the common
findings in all patterns. The epidemic of obesity that currently plagues our nation has
created a unique opportunity. Sufficient numbers of patients are crushing their feet,
which allows us to observe an accelerated version of a process that used to take a
lifetime to occur.1 Regardless of the driving force gravity, time, or a tendency to
collapse, the breakdown of the arch in the transverse and sagittal plane is the
cornerstone to understanding idiopathic acquired conditions of the foot and ankle
and is critical to understanding the treatment of both the supple and the rigid acquired
flatfoot deformity correction.
ANATOMY AND PATHOPHYSIOLOGY

Adult flatfoot deformity can result from a developmental cause, including tarsal
coalition, congenital vertical talus or neuromuscular conditions,2 or as an acquired
condition in an adult.1 The principal mechanical contributors to the adult acquired
flatfoot are contracture of the triceps surae complex,3 attenuation of the ligamentous
supports of the foot,1 and posterior tibial tendon (PTT) dysfunction.4
The soleus muscle originates on the posterior tibia and the gastrocnemius
originates on the posterior distal femur. Depending on the rotation of the hip and the
position of the knee, the triceps surae complex has a dynamic effect on the foot,
depending on the phase of gait.1 The Achilles inserts on the posterior tuberosity of the
calcaneus, acts though the tibiotalar joint, which sits in the posterior third of the
medial longitudinal arch. At mid to late stance, the triceps surae contracts, heel rise
and toe off proceed because of a competent medial column and a functional PTT.4 An
incompetent medial column and a dysfunctional posterior tibial tendon prevent full
inversion during midstance. This laxity then allows the triceps surae to act through the
midfoot instead of the metatarsal heads, leading to further attenuation and collapse.
The lack of full excursion of the Achilles then leads to the inevitable tightness, either
gastrocnemius or combined Achilles. Not only does gastrocnemius tightness cause
symptoms in isolation,5 but it is ubiquitous in the acquired flatfoot deformity.6,7
With the force of the triceps surae acting through the midfoot, ultimately the
attenuation progresses across the entire medial column: talonavicular, naviculocuneiform, and tarsometatarsal articulations. The static supports of the arch, plantar
fascia, spring ligament, and the talocalcaneal interosseous ligament, also stretch.
Finally, compensatory heel valgus ensues due to the progressive forefoot supination.
This constellation of progression leads to the radiographic appearance of lateral or
dorsilateral peritalar subluxation.8,9
The posterior tibial tendon has a normal excursion of only 1.5 to 2.0 cm.10 As the
5-point attachment of the posterior tibial tendon moves more laterally, the demands
on the tendon increase. This accounts for the chronic reparative changes, mucinous
degeneration, fibroblast hypercellularity, chondroid metaplasia, and neovascularization11 that occur in the critical watershed zone: the 14-mm region 1 to 1.5 cm distal
to the medial malleolus.12 A triceps surae contracture is conjectured to contribute to,
if not cause, this failure.7,13 At some point during this process, some threshold is
passed and symptoms occur.

Rigid Arthritic Flatfoot in the Adults

An acquired flatfoot deformity is considered rigid when the deformity cannot be


passively corrected. There is not universal agreement on this definition since the
global deformity involves the hindfoot, transverse tarsal, and midfoot articulations, all
of which can have some degree of rigidity. In 1989 Johnson and Strom14 suggested
the simplest definition. A flatfoot is rigid if the hindfoot is stiff and deformed.
There is no logical explanation why some acquired flatfoot deformities become
rigid. Ellington and Myerson suggested that the transition from flexible to rigid
deformity is a gradual mechanical change during flattening of the arch due to a triceps
surae contracture.15 They further cited 2 other opinions on the cause: habitual
overstrain in patients who are developmentally weak16 and injury to the interosseous
ligament of the subtalar joint.17 The authors opine that progressive capsular contracture is due to chronic joint subluxation in the presence of a triceps surae contracture,
medial column incompetence, and progressive muscle imbalance.
CLASSIFICATION

In 1989, Johnson and Strom proposed a classification of posterior tibial tendon


dysfunction.14 This classification was based on the structure and function of the posterior
tibial tendon and the position of the foot. The third stage in their classification was defined
as an elongated posterior tibial tendon with a fixed hindfoot deformity. Pain is present
medially and sometimes laterally and a treatment with arthrodesis was suggested.
Myerson18 expanded this classification to include the acquired flatfoot deformity with the
tilted ankle. Several further classifications have also been proposed focusing on the
posterior tibial tendon,19,20 and an algorithmic approach to treatment has also been
suggested.21 Posterior tibial tendon dysfunction stage 3 has also been subdivided
into 3A, a deformity that can be corrected by triple arthrodesis, and 3B, with forefoot
abduction corrected with additional bone block arthrodesis.22
To better guide our understanding and treatment, in 2007, the authors introduced a
simple, comprehensive biomechanical model of the distinct types or patterns of foot
collapse, including the acquired flatfoot deformity. This model aimed to globally unite
idiopathic acquired conditions of the foot and ankle in their relation to medial column
incompetence.23
Data were collected on 1142 cases, all of which involved at least 1 of 13 different
procedure types used for acquired idiopathic foot and ankle conditions. An analysis
of these cases showed that there are characteristics common to each of these
conditions. These common characteristics, present to varying degrees in idiopathic
acquired foot and ankle conditions, were analyzed. Five distinct types or patterns of
collapse were identified using data derived from the empirical keying method.24
These patterns are based on a biomechanical model of progressive medial column
incompetence in the presence of a triceps surae contracture and are summarized.
Pattern 1: Precollapse

This pattern involves isolated triceps surae contracture with associated foot pain in
the absence of acquired deformity. Foot pain is due to acquired strain in the absence
of attenuation of the ligamentous supports of the longitudinal and transverse arch
without incompetence of the medial column and without deformity of the talonavicular, naviculocuneiform, intercuneiform, or cuneiform-metatarsal articulations.
Pattern 2: Forefoot

In this pattern, isolated acquired forefoot deformity is the principal component. The
ligamentous supports of the foot are attenuated. Forefoot symptoms result from

325

326

Gentchos et al

multiplanar instability of the medial column in the presence of a triceps surae contracture,
without failure of the transverse arch. Incompetence of the medial column causes greater
weight-bearing demands on the more stable second or third ray.
Pattern 3: Midfoot

A clinical flatfoot appearance was present with arch collapse through the midfoot and
radiographic evidence of midfoot arthrosis was the principal component. This
represents the failure of the transverse arch. Multiple distinct patterns of midtarsal
arthrosis exist. In the most common pattern, medial column incompetence in the
presence of a triceps surae contracture leads to overload of the more stable second
and third rays at the midfoot. The initial point of failure in this type is the more stable
second metatarsalmiddle cuneiform joint. Medial column incompetence here is
suggested as the more stable transverse arch fails by dorsal collapses (dorsal 2nd
tarsometatarsal spur). Failure progresses to the third metatarsallateral cuneiform
joint, followed by the medial navicular cuneiform articulation, depending on the
degree of intercuneiform incompetence.
Pattern 4: Hindfoot

The arch has collapsed. Hindfoot valgus is the principal component. Advanced medial
column incompetence in the presence of a triceps surae contracture results in
forefoot supination leading to more proximal ligamentous attenuation and compensatory heel valgus, which includes the rigid deformity, and is not limited to cases of
symptomatic posterior tibial tendon dysfunction. Radiographic patterns include
dorsolateral peritalar subluxation, lateral peritalar subluxation, or overt collapse
through the midfoot with hindfoot valgus.
Pattern 5: Ankle

Acquired tilted ankle deformity was the principal component. These patients presented with a clinical flatfoot, hindfoot valgus, and ankle pain or overt arthritis. These
ankle deformities were rigid. The triceps surae contracture was advanced, globally
involving the gastrocnemius and soleus. In these cases, the acquired flatfoot
deformity lead to failure of the final medial restraint, the deltoid ligament, and
subsequent ankle valgus.
ALTERNATIVES TO TRIPLE ARTHRODESIS

The goals of the surgical management of the acquired flatfoot deformity are to restore
a plantigrade foot that is stable in the stance phase and functional for propulsion.
Additionally, since the principal symptom that leads to surgery is pain, a minimally
symptomatic foot and ankle should be a principal consideration.
The triple arthrodesis remains a time-tested treatment for the rigid flatfoot deformity, but alternatives exist. Which and how many joints to involve in the arthrodesis
remain questions.
Subtalar Arthrodesis

Johnson and Strom favored an isolated subtalar joint arthrodesis over triple arthrodesis to treat the posterior tibial tendon deficient acquired flatfoot with a deformed
and stiff hindfoot.14 This has been shown to be an effective procedure for stage 2 and
3 deformities25 as well as when combined with a flexor digitorum longus transfer.25,26
Indications include less than 30% talonavicular uncovering, isolated symptomatic
subtalar joint arthrosis, and isolated lateral foot or sinus pain.15 Greater than 10 to

Rigid Arthritic Flatfoot in the Adults

Fig. 1. Preoperative sagittal radiograph of a painful rigid acquired flatfoot deformity. Medial
soft tissue was tenuous and isolated subtalar joint arthrodesis was recommended. The
forefoot had a fixed 15 varus.

15 of forefoot varus or joint hypermobility is a contraindication to isolated subtalar


joint arthrodesis (Figs. 15).
Mann and colleagues suggested that the fixed forefoot varus could cause excessive load on the lateral border of the foot in this circumstance.27 In a review of isolated
subtalar joint arthrodesis for a host of conditions including congenital and acquired
flatfoot deformity (8 of 44), Mann and colleagues27 reported transverse tarsal motion
diminished by 40%, dorsiflexion by 30%, and plantarflexion by 9%. Additionally, they
reported a 36% and 41% incidence of mild progression of ankle and transverse tarsal
arthritis, respectively. Despite this, the residual motion of the transverse tarsal joints,
also reported as 26% and 56% residual mobility, respectively,28 may diminish the rate
of progression of ankle arthrosis when compared to a triple arthrodesis.29
The outcomes of isolated subtalar arthrodesis can be good. Kitoaka and Patzer29
achieved 100% union rate in 21 patients who underwent isolated subtalar joint
arthrodesis for treatment of posterior tibial tendon deficient acquired flatfoot, without
fixed forefoot varus. The only poor outcome was in one patient who had residual

Fig. 2. Immediate postoperative sagittal radiograph shows excellent hindfoot restoration;


however, the forefoot remains in a varus position.

327

328

Gentchos et al

Fig. 3. Immediate postoperative radiographs of an isolated subtalar arthrodesis; an adjunctive tendo-Achilles lengthening was also performed. The axial projection suggested acceptable hindfoot alignment.

hindfoot varus. Deformity was corrected with a 20 increase in talo first metatarsal
angle and a 5-mm increase in arch height, both statistically significant.
As expected, overall complication rates are lower for isolated joint arthrodesis,
particularly since nonunions more commonly occur at the talonavicular and

Fig. 4. Lateral standing radiographs show overt failure of the hindfoot alignment and
recurrence of the flatfoot deformity 12 months after surgery.

Rigid Arthritic Flatfoot in the Adults

Fig. 5. Anteroposterior radiographs show overt failure of the hindfoot alignment and
recurrence of the flatfoot deformity 12 months after surgery.

calcanoecuboid joints with triple arthrodesis.30 Malunion can also cause a poor
result. Mann and colleagues27 considered the residual motion of the transverse
tarsal joint to be protective with an isolated subtalar joint arthrodesis; however, with
varus malalignment, the transverse tarsal joint is locked and the compensatory
motion is diminished. Easley and colleagues31 agreed when they reported the results
of 184 subtalar joint fusions. There was a 9% incidence of symptomatic varus
malunion, 4 of which were considered for hindfoot osteotomy. Careful attention needs
to be given to joint position. Hansen32 demonstrated the use of a lamina spreader
carefully placed into the anterior process of the calcaneus and the lateral shoulder of
the talus used to open the sinus tarsi and reduce the uncovered talus. This is a
powerful tool and overcorrection into varus is possible (Figs. 67).
Undercorrection, with residual hindfoot valgus, can also lead to progressive
midfoot collapse and recalcitrant subfibular impingement.31 For selected patients
with forefoot varus less that 15, when combined with adjunctive procedures, an
isolated subtalar joint arthrodesis can correct the acquired flatfoot deformity with
good clinical outcomes.25
Surgical Technique

The patient and surgical site are appropriately marked according to hospital or
institutional protocol. The procedure is performed under a general anesthetic with a
popliteal block. The anticipation is for outpatient surgery, unless compelling reasons
mandate an inpatient hospitalization. The patient is positioned supine with thigh
tourniquet and an ipsilateral hip bump. Typically the foot retains a considerable
external rotational posture despite this and care must be taken to position the leg with
the anticipation of deformity correction in mind. Assuming no significant malalignment
of the leg exists, the leg is carefully positioned so the knee is directly anterior
regardless of the overall resting position of the foot at this stage. Pressure points on
the well leg are padded and the patient is secured in position. The authors use the
same setup protocol for all procedures reviewed.

329

330

Gentchos et al

Fig. 6. (AD) Correction of he rigid acquired flatfoot deformity with isolated subtalar joint
arthrodesis, tendo-Achilles lengthening, flexor digitorum longus transfer, and spring ligament repair (exostectomy of the talonavicular joint). Lateral hindfoot pain developed 14
months after surgery, likely due to overcorrection of deformity and locking of the transverse
tarsal joint.

The authors start with a tendon Achilles lengthening. Typically, a three-incision


Hoke procedure is the most effective. The extensile lateral approach is utilized, in line
with the fourth metatarsal. The extensor digitorum brevis is elevated. In these
deformities, there are typically no remaining soft tissue attachments in the sinus, and
the subtalar joint is readily entered. A key elevator is used to primarily disrupt
adhesion in the posterior facet. A large lamina spreader is then utilized to
progressively distract the joint. Any remaining articular cartilage is then removed

Rigid Arthritic Flatfoot in the Adults

Fig. 7. (AC) Isolated subtalar joint arthrodesis and tendo-Achilles lengthening was
preformed restoring sagittal alignment.

331

332

Gentchos et al

until the subchondral bone has been fully denuded. The lamina spreader is
repositioned until all articular surfaces have been fully prepared. The exposure is
considered adequate and the contracted joint capsule released when the subtalar
joint is fully distracted and the flexor hallucis longus tendon can be easily seen in
the posterior medial aspect. The joint is irrigated of all loose articular fragments
and the subchondral plate is disrupted. The joint is carefully positioned; the lamina
spreader technique32 is the authors preferred method. Once properly positioned
the joint is compressed with solid or cannulated lag screws. The authors then
proceed to transfer the flexor digitorum longus as well as any further medial
column procedures.
Talonavicular Arthrodesis

Isolated talonavicular arthrodesis can be considered in patients with acquired flatfoot


deformity who retain some rear foot mobility without evidence of arthrosis in the
subtalar joint.33 An isolated talonavicular arthrodesis cannot correct midtarsal instability or excessive hindfoot valgus,15 particularly if the hindfoot is stiff. This arthrodesis
can provide a powerful correction at the focal point of instability.34 However, in a
cadaver model, while the correction can be the same as a double or triple arthrodesis,
hindfoot motion is diminished by 80%.35 Harper and Tisdel33,34 reported the results
of isolated talonavicular arthrodesis for treatment of adult acquired flatfoot deformity
in 26 patients. Their ages ranged from 39 to 74 years, with an average age of 57. The
author used a medial approach and fixed the fusion with cannulated screws. He
reported the clinical outcome, including any loss of motion, the radiographic results,
and the physical appearance of the foot, relative to the contralateral side. The majority
of patients retained some hindfoot motion, up to 20, but was variable. Compared to
the contralateral side, ankle loss of motion was 10, mostly in plantarflexion. The foot
was rated as symmetric in appearance in all but 3 patients. Those 3 were assessed
as residual valgus. There was radiographic progression of arthrosis in 5 patients. This
included the tibiotalar joint, calcaneocuboid joint, and naviculocuneiform joint. An
isolated talonavicular arthrodesis can be an effective treatment for the acquired
flatfoot deformity. When symptoms are present that can clearly be related to the
subtalar joint, an isolated talonavicular arthrodesis is contraindicated. This procedure
addresses the apex of the deformity and offers the possibility of a single surgical
approach, but does pose the risk of residual hindfoot malalignment. When present,
this should be corrected with a hindfoot osteotomy.36
Surgical Technique

The authors proceed initially with Hoke tendo-Achilles lengthening. The talonavicular
joint is entered through an extensile medial utility incision. The talar head is easily
exposed and denuded of articular cartilage. Meticulous preparation of the navicular is
typically more challenging and requires frequent repositioning of the lamina spreader
to ensure that the subchondral bone is fully exposed. Position needs to be carefully
performed but is typically less challenging assuming preoperative assessment of
hindfoot mobility was accurate. Multiple joint stapling devises are available, but the
authors prefer lag screw fixation. A final screw placed percutaneous on the more
lateral aspect of the talonavicular joint can increase the number of lag screws
compressing the joint, a possible concern when relying on correction of deformity
through the talonavicular joint in isolation. A medializing calcaneus osteotomy can be
added if concern exists about foot position.

Rigid Arthritic Flatfoot in the Adults

Double Arthrodesis (Talonavicular and Calcaneocuboid)

The double arthrodesis has been advocated for use in the posterior tibial tendon
deficient flatfoot deformity as an alternative to triple arthrodesis.37,38 Clain and
Baxter37 reported 1 asymptomatic nonunion in 16 patients followed over an 83-month
period. Twelve of their patients had a good or excellent result. They reserved this
procedure for the rigid flatfoot where the deformity originated in the transverse tarsal
joint, and a passively correctable hindfoot valgus. The decision to proceed with a
triple arthrodesis was made intraoperative based on the correct-ability of the hindfoot.
Mann and Beaman38 reviewed the results of 24 double fusions. Their indications
were a fixed forefoot varus on 10 to 15 and a subtalar joint that is passively
correctable to neutral, with subtalar joint arthrosis as a contraindication. They
reported an overall patient satisfaction rate of 83%. Four talonavicular nonunions
were reported, and adjacent joint arthrosis was common.
In a review of the rigid adult acquired flatfoot deformity, Francisco and colleagues39 thought that the constellation of physical findings, the rigid acquired
flatfoot with deformity originating at the transverse tarsal joint without significant
heel valgus or passively correctable valgus, is uncommon. They reported an
alternative double arthrodesis involving the talonavicular and subtalar joints,
allowing correction for the hindfoot valgus and forefoot abduction. The authors
concur that indications for the double arthrodesis in the rigid acquired flatfoot
deformity are rare. Additionally, the functional result of a double arthrodesis in
complete loss of hindfoot mobility, we prefer a triple arthrodesis, or a modification
as suggested by Francisco and colleagues.39
Surgical Technique

The authors initially proceed with Hoke tendo-Achilles lengthening. The talonavicular
joint is exposed and prepared as previously described. The lateral hindfoot is then
approached through an extensile lateral incision in line with the fourth metatarsal. The
calcaneocuboid joint typically is neither difficult to expose nor prepare. Rigid fixation
with joint compression is based on surgeon preference.
Adjunctive Procedures

Surgical procedures in addition to arthrodesis should be expected in planning the


operative management of the rigid acquired flatfoot deformity. A triceps surae
contracture is ubiquitous6,7 and should be treated with a gastrocnemius recession or
tendo-Achilles lengthening. The authors believe that in the more advanced rigid
flatfoot deformities, a gastrocnemius lengthening can be inadequate. A tendo-Achilles
lengthening is almost always needed and has proved a considerable adjunct to
positioning. This is consistent with the suggestion of a progressive triceps surae
contracture associated with the more advanced and rigid acquired flatfoot deformities.15 When considering an isolated subtalar arthrodesis, treating a symptomatic
posterior tibial tendon tear with flexor digitorum longus transfer can be critical.25
Medial column incompetence in the coronal and sagittal plane is also ubiquitous in the
acquired flatfoot deformity. Judicious use of medial column stabilization is not readily
required in the rigid deformity, particularly since the talonavicular joint is commonly
involved in the arthrodesis.
SUMMARY

Every alternative to triple arthrodesis in the rigid acquired flatfoot deformity is


predicated on limiting the patient exposure to the complication associated with triple

333

334

Gentchos et al

arthrodesis. When possible, avoiding arthrodesis of either the talonavicular and


calcaneocuboid joints, with their higher nonunion rates, seems a cogent option.
Successful treatment is dependent on thoughtful patient evaluation and examination,
meticulous joint preparation, careful positioning with rigid fixation, and judicious use
of adjunctive procedures to achieve the goal of a plantigrade foot that functions well
and is minimally painful.
REFERENCES

1. Van Boerum DH, Sangeorzan BJ. Biomechanics and pathophysiology of flat foot.
Foot Ankle Clin 2003;8(3):419 29.
2. McCormack AP. Ching RP, Sangeorzan BJ. Biomechanics of procedures used in
adult flat foot deformity. Foot Ankle Clin 2001;6(1):1523.
3. Harris, RI, Beath T. Hypermobile flat foot with short tendo Achilles. J Bone Joint Surg
Am 1948;30A:116 40.
4. Thordarson DB, Schmotzer H, Chon J, et al. Dynamic support of the human longitudinal arch: a biomechanical evaluation. Clin Orthop 1995;316:16572.
5. Digiovanni CW, Kuo R, Tejwani N, et al. Isolated gastrocnemius tightness. J Bone
Joint Surg 2002;84:96270.
6. Coetzee JC, Castro MD. The indications and biomechanical rationale for various
hindfoot procedures in the treatment of posterior tibialis tendon dysfunction. Foot
Ankle Clin 2003;8:4539.
7. DiGiovanni CW, Langer P. The role of isolated gastrocnemius and combined Achilles
contractures in the flatfoot. Foot Ankle Clin 2007;12:36379.
8. Anderson JG, Hansen ST. Surgical treatment of posterior tendon pathology. In
Kelikien AS, editor. Operative treatment of the foot and ankle. Stanford (CT): Appleton
& Lange; 1999. p. 2113.
9. Hansen ST. Functional reconstruction of the foot and ankle. Baltimore (MD): Williams
& Wilkins; 2000: p. 195207.
10. Mann RA. Biomechanics of the foot and ankle. In Mann RA, Coughlin MJ, editors.
Surgery of the foot and ankle. St Louis (MO): Mosby; 1993. p. 3 43.
11. Mosier SM, Lucas DR. Pomeroy GC, et al. Pathology of the posterior tibial tendon in
posterior tibial tendon insufficiency. Foot Ankle Int 1998;19:520 4.
12. Frey CC, Schereff M, Greenidge N. Vascularity of the posterior tibial tendon. J Bone
Joint Surg Am 1990;72: 884 8.
13. Mosier SM, Pomeroy G, Manoli A. Pathoanatomy and etiology of posterior tibial
tendon dysfunction. Clin Orthop 1999;365:1222.
14. Johnson KA, Strom DE. Tibialis posterior tendon dysfunction. Clin Orthop 1989;239:
196 206.
15. Ellington JK, Myerson MS. The use of arthrodesis to correct rigid flatfoot deformity.
Instruc Course Lec 2011;60:31120.
16. Todd AH. The treatment of pes cavus. Proc R Soc Med 1934;28(2):11728.
17. Lapidus PW. The so-called longitudinal arch: some new thoughts. Am J Phys
Anthropol 1948;6(2):241.
18. Myerson MS. Adult acquired flatfoot deformity: treatment of dysfunction of the
posterior tibial tendon. J Bone Joint Surg 1996;78:780 92.
19. Conti S, Michelson J, Jahss M. Clinical significance of magnetic resonance imaging in
preoperative planning for reconstruction of posterior tibial tendon ruptures. Foot Ankle
1992;13:208 14.
20. Cooper AJ, Mizel MS, Patel PD, et al. Comparison of MRI and local anesthetic tendon
sheath injection in the diagnosis of posterior tibial tendon tenosynovitis. Foot Ankle Int
2007;28:1124 7.

Rigid Arthritic Flatfoot in the Adults

21. Weinraub GM, Heilala MA. Adult flat foot/posterior tibial tendon dysfunction; outcomes analysis of surgical treatment utilizing an algorithmic approach. J Foot Ankle
Surg 2000;39:359 64.
22. Bluman EC, Title CI, Myerson MS. Posterior tibial tendon rupture: a refined classification. Foot Ankle Clin 2007;12:233 49.
23. Gentchos CE, Jelenick A, Bohay DR, et al. Patterns of arch collapse associated with
progressive gastrocnemius contracture. Presented at the annual scientific meeting of
the American Orthopaedic Foot and Ankle Society. Toronto (Canada), July 2007.
24. Hathaway SR, McKinley JC. Minnesota Multiphasic Personality Inventory revised.
Minneapolis (MN): University of Minnesota Press; 1943.
25. Johnson JE, Cohen BE, DiGiovanni BF, et al. Subtalar arthrodesis with flexor digitorum longus transfer and spring ligament repair for treatment of posterior tibial tendon
insufficiency. Foot Ankle Int 2000;21(9):7229.
26. Stephens HM, Walling AK, Solmen JD, et al. Subtalar repositional arthrodesis for adult
acquired flatfoot. Clin Orthop Rel Res 1999;365:69 73.
27. Mann RA, Beaman, DM, Horton GA. Isolated subtalar arthrodesis. Foot Ankle Int
1998;19(8):5119.
28. Astion DJ, Deland JT, Otis JC, et al. Motion of the hindfoot after simulated arthrodesis.
J Bone Joint Surg Am 1997;79:241 6.
29. Kitaoka HB, Patzer GL. Subtalar arthrodesis for posterior tibial tendon dysfunction
and pes planus. Clin Orthop Rel Res 1997;345:18794.
30. Kadakia AR, Haddad SL. Hindfoot arthrodesis for the adult acquired flat foot. Foot
Ankle Clin 2003;8(3):569 94.
31. Easley ME, Trnka HJ, Schon LC, et al. Isolated subtalar joint arthrodesis. J Bone Joint
Surg Am 2000;82(A):61324.
32. Hansen ST. Functional reconstruction of the foot and ankle. Baltimore (MD): Williams
& Wilkins; 2000. p. 3023.
33. Harper MC, Tisdel CL. Talonavicular arthrodesis for the painful adult acquired flatfoot
deformity. Foot Ankle Int 1996;17(11):658 61.
34. Harper MC. Talonavicular arthrodesis for the acquired flatfoot in adults. Clin Orthop
Relat Res 1999;365:65 8.
35. OMalleyMJ, Deland JT, Lee KT. Selective hindfoot arthrodesis for the treatment of
acquired flatfoot deformity: an in vitro study. Foot Ankle Int 1995;16(7):4117.
36. Fortin PT. Posterior tibial tendon insufficiency. Isolated fusion of the talonavicular joint.
Foot Ankle Clin 2001;6(1):13751.
37. Clain MR, Baxter DE. Simultaneous calcaeocuboid and talonavicular fusion. Long
term follow up study. J Bone Joint Surg Br 1994;76(1):133 6.
38. Mann RA, Beaman DN. Double arthrodesis in the adult. Clin Orthop Rel Res 1999;
365:74 80.
39. Francisco R, Chiodo CP, Wilson MG. Management of the rigid adult acquired flatfoot
deformity. Foot Ankle Clin 2007;12:31727.

335

Foot Ankle Clin N Am


12 (2007) 287299

Medial Column Procedures


in the Acquired Flatfoot Deformity
Bruce E. Cohen, MD*, Field Ogden, MD
O.L. Miller Foot and Ankle Institute, OrthoCarolina, 1001 Blythe Boulevard,
Suite 200, Charlotte, NC 28203, USA

An acquired atfoot deformity remains a complex problem with many


causes and even more solutions. Typically characterized by a depressed longitudinal arch, this deformity entails varying degrees of hindfoot valgus,
forefoot varus, and forefoot abduction [1]. Its etiology includes posterior
tibial tendon dysfunction, congenital atfoot, trauma, arthritis, neuromuscular disorders, and Charcot arthropathy [24]. Posterior tibial tendon dysfunction accounts for the majority of the cases of acquired atfoot in the
adult population [5,6].
The posterior tibial tendon (PTT) acts as the primary stabilizer of the medial longitudinal arch of the foot. The PTT works in concert with the exor
digitorum longus (FDL) and the exor hallucis longus (FHL) in this function, but it is the most signicant. The PTT also contributes to the integrity
of the static restraints of the arch, including the spring ligament and the
plantar calcaneonavicular ligament [7]. Dysfunction or insuciency of the
PTT may lead to collapse of the arch through any or all of the medial
column joints: the naviculocuneiform (NC), talonavicular (TN), and rst
metatarsocuneiform (MTC) [7]. The persistent hindfoot valgus results in
signicant rst ray elevation or forefoot varus, which can be exible and correctable in the early stages of the disease or rigid and xed in the later stages.
Physical examination is the mainstay of staging and establishing the diagnosis of posterior tibial tendon dysfunction [8]. In stage I, the disease is characterized by minimal or no deformity. The patient is able to perform a single
limbheel rise. Function of the PTT persists but causes pain. Stage II is characterized by tendon elongation and degeneration. On physical examination,
there is a exible atfoot deformity and the patient is unable to perform
a single limbheel rise. A xed, rigid, irreducible deformity is the hallmark

* Corresponding author.
E-mail address: bruce.cohen@orthocarolina.com (B.E. Cohen).
1083-7515/07/$ - see front matter 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.fcl.2007.03.002
foot.theclinics.com

288

COHEN & OGDEN

of stage III. This patient is unable to perform a single or double limbheel


rise. Instability of the rst MTC joint should be included in the course of the
physical examination. Dorsal midfoot pain, accompanied by a symptomatic
bony prominence, is the hallmark symptom of MTC involvement in the
acquired atfoot.
On radiographs, the atfoot has a characteristic appearance: the lateral
column is short compared to the medial column. Specically, the TN joint
is approximately 3 to 5 mm distal to the calcaneocuboid joint on an anteroposterior (AP) radiograph [9]. This relationship is in contrast to the parallel
joints on a normal radiograph. A radiographic feature that correlates to the
severity of the deformity is the TN coverage angle, which predicts the
amount of forefoot abduction. Another radiographic measure that specically relates to the medial column position and joint alignment is the lateral
talo-rst metatarsal angle (Mearys angle). This is measured on the standing
lateral radiograph and is formed by the intersection of the longitudinal axes
of the talus and the rst metatarsal. A measurement of 1530 indicates the
presence of moderate pes planus, and a measurement greater than 30 is
classied as severe [10]. Finally, the radiographic location of weight-bearing
sag may be indicative of joint involvement and instability. MTC instability
may result in plantar gapping on a weight-bearing lateral radiograph. To
this end, degenerative changes at any of the medial column articulations
may be seen. These ndings should be used to guide surgical treatment
and stabilization options.
The treatment of the acquired atfoot has numerous options ranging
from conservative, nonsurgical treatments to more extensive procedures involving soft tissue reconstructions with bony procedures. Surgical options
have included tenodesis, tendon transfers, osteotomies, fusions, and multiple combinations of these procedures. Most of the combined bone and
soft tissue procedures begin with a medial soft tissue reconstruction, most
commonly an FDL tendon transfer to the navicular. The imbrication of
an attenuated spring ligament also is considered in the soft tissue reconstruction. A contracted gastrocsoleus complex is often associated with the
acquired atfoot deformity. The lengthening of the Achilles tendon, or
preferentially the gastrocnemius, serves to lessen the hindfoot valgus
moment and the transfer of the FDL substitutes for the dysfunctional PTT.
These soft tissue procedures can restore some of the function of the decient PTT complex; however the presence of the attenuated medial column
ligamentous restraints eventually will cause attenuation and insuciency of
the transfer/reconstruction. Because of the joint forces involved, the addition of bony reconstructive procedures is imperative. The lateral procedures,
such as calcaneal osteotomies and calcaneocuboid arthrodesis, are outside
the scope of this article.
Although there are many options available to the orthopedic surgeon,
this article focuses on the medial column procedures for the treatment of
the acquired atfoot. When performing a medial column procedure alone,

ACQUIRED FLATFOOT DEFORMITY

289

in combination with soft tissue reconstruction, or with a concomitant lateral


bony procedure, the deformity being addressed should include signicant
abduction or collapse through the TN, NC, or tarsometatarsal (TMT) joints
[11]. The procedures covered in this article include arthrodesis of the TN,
NC, and MTC joints, together or in isolation. Osteotomies of the cuneiforms will also be addressed.
Rationale behind isolated medial column stabilization procedures
The contribution of medial column stability to the entire atfoot deformity is still somewhat unclear. Hansen and his colleagues have popularized
the phrases dorsilateral peritalar subluxation and forefoot-driven hindfoot
valgus. Dorsilateral peritalar subluxation occurs as the medial column support of the talus fails, the TN joint abducts, and the subtalar joint falls into
eversion. This peritalar subluxation is measured best radiographically by the
TN coverage angle popularized by Sangeorzan and colleagues [12]. This
leads to the principle of the resulting hindfoot deformity being linked to
the loss of forefoot stability caused by the loss of medial column support.
Consequently, the hindfoot deformity is driven by the loss of medial column
support [11].
The goal of an isolated medial column stabilizing procedure is to maintain hindfoot motion. An isolated arthrodesis of the TN joint reduces hindfoot motion by approximately 80%90% while NC or MTC arthrodeses
will maintain motion of the hindfoot/subtalar joints [13]. Because the atfoot deformity is frequently complex and multifactorial, these medial column procedures are often performed in conjunction with lateral
procedures such as calcaneal lengthening or medial displacement calcaneal
osteotomy.
Naviculocuneiform arthrodesis
Miller procedure and others
Miller [14] rst published his proposed procedure to correct the symptomatic at or relaxed foot in the adolescent or pre-adolescent child in
1927. He hypothesized that attenuation of the medial supporting structures
of the foot resulted in the progression of the atfoot deformity. His procedure was designed to correct the medial alignment of the foot by fusion of
the medial column; specically, an arthrodesis of the NC and rst MTC
joints was proposed (Figs. 1A, B). In addition to the medial column fusion,
an advancement of the PTT complex as an osteoperiosteal ap was performed. Miller noted symptomatic improvement in 16 adolescent patients
at up to 2.5 years follow-up.
The Miller atfoot procedure proceeds as follows: after an incision is
made along the medial column of the foot, the anterior edge of the PTT

290

COHEN & OGDEN

Fig. 1. (A, B) Standing AP and lateral radiographs of a patient 40 years after the Hoke-Miller
atfoot procedure.

is exposed, and the anterior tibialis tendon insertion is identied. An osteoperiosteal ap is created, which leaves the plantar attachments of the PTT
complex. This ap includes the tendon and bony components of the navicular, medial cuneiform and rst metatarsal base. This is performed with
a small osteotome. Arthrodesis of the NC and rst TMT joints is performed
with bony preparation that removes slightly more bone on the medial and
plantar surfaces, which allows correction of the planus and abduction components of the deformity. The next step is to advance the osteoperiosteal
ap and pass this beneath the insertion of the anterior tibialis tendon and
attach it to the adjacent soft tissue structures.
A second procedure to correct a symptomatic atfoot was proposed by
Hoke [15] in 1931. Hoke reported the use of a NC arthrodesis in both adolescents and adults. He believed that the important components of the adolescent atfoot were the contracted Achilles tendon and the extreme
exibility of the midtarsal joints. His procedure consisted of an advancement of the PTT complex, similar to that described by Miller, coupled
with a bone-block arthrodesis of the NC joint and an open Achilles
lengthening.
Duncan and Lovell [16] combined these two procedures and described it
as the modied Hoke-Miller procedure. This procedure was characterized
by an NC arthrodesis, an opening wedge osteotomy of the medial cuneiform, and the advancement of the spring ligament to support the reconstruction. Duncan and Lovell presented 10 adolescent patients who had been
treated by their procedure at an average of 1.7 years of follow-up. Results
of this procedure demonstrated symptomatic improvement in all of the
patients at follow-up.
In 1953, Caldwell [17] described a third surgery that included an arthrodesis of the NC joint. Like the above procedures, this Durham-plasty

ACQUIRED FLATFOOT DEFORMITY

291

procedure was targeted for the adolescent atfoot. Specically, an NC


fusion was combined with the advancement of an osteoperiosteal ap and
a separate advancement of the PTT. Caldwell reported that 95% of subjects
experienced relief of symptoms at an average of 6 years of follow-up.
Coleman [18] also reported symptomatic relief in 95% of 33 feet with a
maximum of 26 years of follow-up.
Isolated naviculocuneiform arthrodesis
The above procedures were described primarily for use in the pediatric or
adolescent atfoot. Similar principles can be used in the adult acquired
atfoot.
If the location of the collapse of the longitudinal arch on the standing
lateral radiograph is at the NC joint, a stabilization procedure of this joint
should be considered. When there is signicant hindfoot valgus, the NC arthrodesis should be performed in conjunction with a calcaneal osteotomy.
Mild hindfoot valgus with signicant medial column involvement can be
stabilized by an isolated arthrodesis. Greisberg and colleagues [11] reviewed isolated arthrodeses of the NC in conjunction with rst TMT arthrodesis in 19 patients. The patients had excellent correction of the
lateral talo-rst metatarsal angle and signicant improvement in the TN
coverage angle.
The surgical technique involves a medial longitudinal incision. The NC
joint is identied proximal and inferior to the insertion of the anterior tibialis tendon. The medial and middle facets of the joint are prepared for arthrodesis. Typically, the lateral cuneiform-navicular joint is not included
in the arthrodesis. After the standard joint preparation, the joint is reduced
by plantarexion and derotation. The appropriate rigid internal xation is
placed. Screw, compression staple, and/or plate xation is used based on
surgeon preference (Figs. 2A, D). The use of local or other autologous
bone graft is advised, because of the signicant risk of nonunion.

Cuneiform osteotomy
In 1935, Cotton [19] reported promising results of an opening wedge medial cuneiform osteotomy as a component of the treatment for the depressed
medial arch in pes planus. The medial cuneiform osteotomy typically has
been used in the reconstruction of forefoot adductus and residual clubfoot
[20]. Recently, there is renewed interest and increasing adjunctive use of
the Cotton osteotomy in the treatment of the forefoot varus component
of atfoot deformity [21]. Hirose and Johnson [21] used the Cotton osteotomy and reported favorable clinical and radiographic results in 16 feet. Average time for radiographic union was 12 weeks in their adult patients.
Postoperative complications were minimal and included screw removal in
one patient [21].

292

COHEN & OGDEN

Fig. 2. (A, B) Preoperative standing AP and lateral radiographs demonstrating NC collapse


contributing to loss of medial column support. (C, D) Postoperative standing AP and lateral
radiographs following isolated NC arthrodesis with gastrocnemius recession.

Surgical technique
After completing concomitant procedures for atfoot correction, the
amount of residual forefoot varus and the correction of the lateral talo-rst
metatarsal angle are assessed. If signicant deformity is present, the cuneiform osteotomy is performed. A dorsal longitudinal incision is used to expose the medial cuneiform. The extensor hallucis longus (EHL) tendon is
retracted medially and the medial cuneiform is osteotomized with a microsagittal saw from dorsal to plantar. The exact location of the osteotomy is veried by uoroscopy. Hirose and Johnson [21] describe the optimal location
for the osteotomy on the AP projection to be at the second TMT joint.
An osteotome is used to lever open the cuneiform and measure the desired
correction on a lateral uoroscopic view. Typically 46 mm of correction is
necessary. Interpositional graft is then placed in the osteotomy. Hirose and
Johnson used tricortical autograft; however, allograft or graft substitutes
can be substituted. Fixation is based on surgeons preference. Commonly
used implants include a single screw, compression staple, or a small plate.

ACQUIRED FLATFOOT DEFORMITY

293

The postoperative protocol depends on the entire reconstruction


performed; the cuneiform osteotomy typically can allow weight bearing
at 6 weeks.
A cuneiform osteotomy is easy to perform, is reliable, and has minimal
morbidity. The opening wedge component is relatively small and fusion
rates seem encouraging. Cotton [19] described a triangle of support for
the foot, which also has been described as the tripod of the foot. This is comprised of the rst metatarsal head, fth metatarsal head, and plantar heel.
This procedure allows further restoration of the tripod. Restoration of the
radiographic alignment of the foot is pronounced most on the lateral talorst metatarsal angle. The level of radiographic alignment restoration is
slightly less than that seen with the Hoke procedure, but there is less concern
and risk for nonunion [21]. This procedure is an excellent adjunct to a comprehensive atfoot correction, and it should be considered when there is
signicant forefoot varus without degeneration of the rst TMT joint
(Figs. 3A, D).

Fig. 3. (A, B) Preoperative AP and lateral radiographs of a 38-year-old patient with continued
progressive pes planovalgus 24 years after the Miller atfoot procedure. (C, D) Postoperative
standing AP and lateral radiographs following PTT reconstruction with FDL tendon transfer,
medial displacement calcaneal osteotomy, lateral column lengthening, and opening wedge
osteotomy.

294

COHEN & OGDEN

Talonavicular arthrodesis
Whereas the triple arthrodesis remains the gold standard procedure in
the correction of a atfoot [22], the TN joint is at the apex of this deformity
[13]. Using an in vitro model, OMalley and colleagues [13] showed that an
isolated TN fusion was as powerful in correcting the components of the atfoot as both a triple and double arthrodesis. All three procedures provided
more signicant correction than an isolated subtalar arthrodesis [13]. TN
arthrodesis provides stable powerful correction, while addressing the dynamic and static components of the acquired atfoot deformity.
Joint sparing procedures for atfoot correction have increased in popularity as an alternative to triple arthrodesis in younger patients who have supple
deformities [23]. Harper and Tisdel reported on 26 patients with an average of
27 months of follow-up who were treated with an isolated TN arthrodesis for
painful atfoot. At follow-up, 24 of 26 patients were either satised with minor reservations or completely satised. The same number reported either no
pain or pain only after heavy use. Harper [24] followed this initial study with
another that showed similar results with 25 of 29 patients satised with no or
minor reservations at an average of 26 months of follow-up. In these series,
the incidence of nonunion was less than 10%. At approximately 2 years of follow-up, mild asymptomatic adjacent joint arthrosis was noted and tibiotalar
motion was reduced in plantarexion by an average of 10 [24].
Fortin outlined several exclusion criteria for isolated TN fusion: excessive
hindfoot valgus, valgus ankle instability, ankle arthrosis, and instability of
the more distal midfoot joints [25]. Thus, the indications for the isolated
TN fusion are subjective, and the use of this isolated fusion remains incompletely dened.
Nonunion rates for TN fusion vary signicantly. Mann and Beaman [26]
reported up to 25% nonunion of the TN joint in the presence of double arthrodesis for PTT insuciency. This series included dorsal staple xation
without lag screw xation. Adjacent joint degeneration has been reported
in the ankle, subtalar, and midfoot joints [25]. Although most of these procedures are clinically asymptomatic, it is important to consider adjacent
joint arthrosis when choosing an arthrodesis procedure.
Surgical technique
Before performing the arthrodesis, the patient is examined for any evidence
of contracture of the gastrocsoleus complex. If present, a gastrocnemius recession or Achilles lengthening is performed. If excessive hindfoot valgus is present,
the addition of a medial displacement calcaneal osteotomy may be considered.
The TN joint is approached through a dorsal or dorsomedial incision.
The interval can be either just medial to the anterior tibialis tendon or between the tibialis anterior and the EHL tendon. The joint is prepared in
the standard fashion for arthrodesis. The addition of autograft or bone graft
substitute is recommended. The authors (Cohen, BE) preference is either

ACQUIRED FLATFOOT DEFORMITY

295

proximal tibial or iliac crest autograft. Distraction of the joint during the
preparation is facilitated easily with a laminar spreader or other similar device. The reduction of the joint is the most critical aspect of the procedure.
The abduction of the navicular, the plantar exion of the talus, and the rotation of the joint must be corrected and provisionally xed. Manual pressure forcing the talar head laterally is critical in correcting the abduction.
Fixation of the joint requires more than one point of xation to control rotation. Either two screws or a combination of a screw and a dorsally placed
staple (or similar device) is sucient. The intraoperative use of a at plate
(or other surface) is important to verify a plantigrade foot without excessive
pronation or supination. Postoperatively, the patient is kept non-weight
bearing for 6 weeks, followed by 46 weeks of weight bearing in a removable
boot or until there is radiographic evidence of osseous union.

First tarsometatarsal arthrodesis


In 1934, Lapidus [27] originally described a procedure that included an
arthrodesis from the rst and second metatarsal bases to the medial cuneiform, a distal soft tissue plication, and a bunionectomy for the correction of
hallux valgus. Multiple modications of his procedure have resulted in the
modied Lapidus procedure, in which the rst TMT joint is fused to the medial and middle cuneiforms [28]. When medial column stabilization is performed in atfoot reconstruction, it is not uncommon to include an
arthrodesis of the rst TMT joint [29].
Initial reticence to perform this procedure stems from reports of 10%
20% nonunion [29]. A recent review by Thompson, Bohay, and Anderson
[29] included 182 patients who underwent rst TMT joint arthrodesis as
a component of atfoot reconstruction or correction of hallux valgus. The
nonunion rate in all patients was only 4%, and for atfoot patients (21 of
182), there were no incidences of nonunion. Potential explanations of their
remarkable fusion rate include meticulous bone preparation, the use of bone
graft, and frequent use of Achilles lengthening or gastrocnemius recession.
The indications for including a rst TMT arthrodesis in the reconstruction of PTT insuciency are similar to that of the medial cuneiform osteotomy. Some indications include signicant forefoot varus or an abnormal
lateral talo-rst metatarsal angle on a standing radiograph. Other indications include secondary long-standing rst ray elevation and the presence
of hypermobility of the rst ray. Also, any signicant degenerative change
in the rst TMT joint with the presence of xed forefoot varus is a strong
indication for this procedure.
Surgical technique
The surgical approach includes a dorsal incision medial to the EHL
tendon. The joint is prepared in the standard manner. It is important to

296

COHEN & OGDEN

Fig. 4. (A) Preoperative standing AP radiograph of bilateral feet before rst TMT joint
arthrodesis. (B) Preoperative standing lateral radiograph of bilateral feet before rst TMT joint
arthrodesis, which demonstrates plantar gapping at rst TMT joint. (C, D) Postoperative standing AP and lateral radiographs following rst TMT joint arthrodesis. (Courtesy of John G.
Anderson, MD, Grand Rapids, MI.)

remember that the rst TMT joint is very deep, measuring up to 30 mm


from dorsal to plantar. Leaving unprepared plantar bone will lead to dorsiexion, which is undesirable. After the joints are prepared, a small wedge
can be removed from the plantar aspect to create a plantarexed position
of the arthrodesis. The use of autograft or bone graft substitutes is recommended. Fixation is typical with a dual compression screw technique
(Figs. 4A, D). The combination of compression screw xation and a dorsal
plate or staple also is adequate. Postoperative weight bearing is delayed
until approximately 6 weeks and immobilization is discontinued at
1012 weeks, or at the time evidence of osseous union is present
radiographically.

Summary
The contribution of the medial column to the adult acquired atfoot deformity can be substantial. Attenuation of the static structures resulting in
loss of support of the TN joint as well as the NC and TMT joints can result

ACQUIRED FLATFOOT DEFORMITY

297

Fig. 5. (A, B) Preoperative standing AP and lateral radiographs demonstrating extensive medial
column involvement and collapse. (C, D) Postoperative AP and lateral radiographs following
extensive medial column stabilization of NC and midfoot joints. (Courtesy of John G.
Anderson, MD, Grand Rapids, MI.)

in signicant deformity. Coupled with the loss of the dynamic function of


the PTT, attenuation can lead to signicant peritalar subluxation, rst ray
elevation, and midfoot/midtarsal sag or collapse, with or without degenerative change.
Medial column stabilization procedures are an important component of
the atfoot reconstruction. Whether in isolation or in combination with lateral bony procedures or medial soft tissue reconstructions, these procedures
are powerful techniques for correcting forefoot varus and TN uncoverage
(Figs. 5A, D). In isolation, these techniques are less eective in correcting
hindfoot valgus. The stepwise analysis and correction of these complex deformities are extremely important. No single procedure can be applied to
all patients who have an acquired atfoot. Attention to each component of
the deformity, including the gastrocsoleus complex, is imperative in providing and maintaining the best correction of this increasingly common
deformity.

298

COHEN & OGDEN

Acknowledgment
The authors thank John G. Anderson, MD, Grand Rapids, MI for Figs.
4AD, 5AD.

References
[1] Gould N, Schneider W, Ashikaga T. Epidemiological surgery of foot and ankle problems in
the continental United States 19781979. Foot Ankle 1980;1:8.
[2] Mann DA. Acquired at foot in adults. Clin Orthop 1983;181(12):4651.
[3] Anderson RB, Davis WH. Management of the adult atfoot deformity. In: Myerson MS,
editor. Foot and ankle disorders. Philadelphia: WB Saunders; 2000. p. 101739.
[4] Myerson MS. Adult acquired at foot deformity. J Bone Joint Surg Am 1996;78(5):78092.
[5] Hansen ST. Progressive symptomatic at foot (lateral peritalar subluxation). In: Hansen ST,
editor. Functional reconstruction of the foot and ankle. Philadelphia: Lippincott Williams &
Wilkins; 2000. p. 195205.
[6] Funk DA, Cass JR, Johnson KA. Adult acquired at foot deformity secondary to posterior
tibial tendon pathology. J Bone Joint Surg 1986;68A:95102.
[7] Pedowitz WJ, Kovatis P. Flat foot in the adult. J Am Acad Orthop Surg 1995;3:293302.
[8] Johnson KA, Strom DE. Tibialis posterior tendon dysfunction. Clin Orthop 1989;239:
196206.
[9] Stephens HM, Walling AK, Solmen JD, et al. Subtalar repositional arthrodesis for adult acquired atfoot. Clin Orthop 1999;365:6973.
[10] Chi TD, Toolan BC, Sangeorzan BJ, et al. The lateral column lengthening and medial column stabilization procedures. Clin Orthop Relat Res 1999;365:8190.
[11] Greisberg J, Assal M, Hansen ST Jr, et al. Isolated medial column stabilization improves
alignment in adult-acquired atfoot. Clin Orthop 2005;435:197202.
[12] Sangeorzan BJ, Mosca V, Hansen ST. Eect of calcaneal lengthening on the relationship
among the hindfoot, midfoot and forefoot. Foot Ankle 1993;14:13641.
[13] OMalley MJ, Deland JT, Lee KT. Selective hindfoot arthrodesis for the treatment of adult
acquired at foot deformity: an in vitro study. Foot Ankle Int 1995;16(7):4117.
[14] Miller OL. A plastic at foot operation. J Bone Joint Surg 1927;9A:8491.
[15] Hoke M. An operation for the correction of extremely relaxed at feet. J Bone Joint Surg
1931;13A:77383.
[16] Duncan JW, Lovell WW. Modied Hoke-Miller at foot procedure. Clin Orthop 1983;181:
247.
[17] Caldwell GD. Surgical correction of relaxed at foot by the Durham foot plasty. Clin Orthop
1953;2:2216.
[18] Coleman SS. Severe, exible planovalgus foot (at foot). In: Complex foot deformities in
children. Philadelphia: Lea & Febinger; 1992. p. 193222.
[19] Cotton F. Foot statistics and surgery. Trans N Engl Surg Soc 1935;18:1818.
[20] Fowler B, Brooks AL, Parrish TF. The cavovarus foot. J Bone Joint Surg 1959;41A:75761.
[21] Hirose CB, Johnson JE. Plantarexion opening wedge medial cuneiform osteotomy for correction of xed forefoot varus associated with atfoot deformity. Foot Ankle 2004;28(8):56874.
[22] Sangeorzan BJ, Smith D, Veith R, et al. Triple arthrodesis using internal xation in the treatment of adult foot disorders. Clin Orthop 1993;294:299307.
[23] Harper MC, Tisdel CL. Talonavicular arthrodesis for the painful adult acquired at foot.
Foot Ankle 1996;17:65861.
[24] Harper MC. Talonavicular arthrodesis for the acquired at foot in the adult. Clin Orthop
1999;365:658.
[25] Fortin PT. Posterior tibial tendon insuciency. Isolated fusion of the TN joint. Foot Ankle
Clin 2001;6(1):13751.

ACQUIRED FLATFOOT DEFORMITY

299

[26] Mann RA, Beaman DN. Double arthrodesis in the adult. Clin Orth 1999;365:7480.
[27] Lapidus PW. The operative correction of the metatarsus primus varus in hallux valgus. Surg
Gynecol Obstet 1934;58:18391.
[28] Sangeorzan B, Hansen ST Jr. Modied Lapidus procedure for hallux valgus. Foot Ankle
1989;9:2626.
[29] Thompson IM, Bohay DR, Anderson JG. Fusion rate of rst tarsometatarsal arthrodesis in
the modied Lapidus procedure and atfoot reconstruction. Foot Ankle 2005;26:698703.

Medial Column Procedures in the


Correction of Adult Acquired
Flatfoot Deformity
Jeremy J. McCormick, MD*, Jeffrey E. Johnson, MD

KEYWORDS
Adult acquired flatfoot defomity Cotton osteotomy Forefoot varus
Medial column osteoarthritis
KEY POINTS
Forefoot varus is a common finding associated with the adult acquired flatfoot deformity
(AFFD).
Physical examination of the foot, when the hindfoot has been corrected to neutral, is the
best method to determine if forefoot varus exists and whether surgical correction is
needed.
Forefoot varus owing to mild instability of the naviculocuneiform joint or the first
tarsometatarsal joint is treated with a plantarflexion opening wedge osteotomy (Cotton
osteotomy) of the first cuneiform.
Forefoot varus owing to medial column osteoarthritis or significant instability is treated with
arthrodesis of the involved joint(s) to restore the tripod of the foot.

Adult acquired flatfoot deformity (AAFD) is a global term that applies to patients with
varying degrees of hindfoot valgus, forefoot abduction, and forefoot varus. Most
commonly, a patient complains of medial hindfoot pain and swelling with a variable
degree of progressive pes planovalgus. The most common cause of AAFD is posterior
tibial tendon (PTT) dysfunction.1 Function of the PTT is critical because it contributes
to the static stability of the foot along with the spring ligament complex.2 Failure of the
PTT to maintain the arch of the foot can lead to progressive hindfoot valgus and
collapse of the medial column through the first metatarsocuneiform joint (also known
as the first tarsometatarsal [TMT] joint), the naviculocuneiform (NC) joint, and/or the
talonavicular (TN) joint.2 Other etiologies of AAFD include pathology such as posttraumatic deformity, osteoarthritis, inflammatory arthritis, Charcot neuroarthropathy,
and neuromuscular disorders.3,4 Numerous combinations of bone and soft-tissue

Department of Orthopedic Surgery, Foot and Ankle Service, Washington University School of
Medicine, 660 South Euclid Avenue, St Louis, MO 63110, USA
* Corresponding author.
E-mail address: mccormickj@wudosis.wustl.edu
Foot Ankle Clin N Am 17 (2012) 283298
http://dx.doi.org/10.1016/j.fcl.2012.03.003
1083-7515/12/$ see front matter 2012 Published by Elsevier Inc.

foot.theclinics.com

284

McCormick & Johnson

procedures have been described to address the different types of pes planovalgus
deformity without a clear consensus on a treatment algorithm.2,515
BACKGROUND

The typical history given by a patient with an acquired flatfoot deformity secondary to
PTT insufficiency is a progressive change in the shape of their foot over time. The
patient describes increasing pain and difficulty with weight-bearing activities. Typically, there is no specific instance where the foot changed form, or collapsed.
Physical examination begins with an observation of the standing alignment of the
foot and ankle to evaluate the extent of hindfoot valgus, midfoot pronation, and
forefoot abduction. Assessment of neurovascular and motor function is performed to
rule out other causes of tendon dysfunction and ensure that the remaining muscle
function in normal. The PTT is examined with the patient seated by asking them to
invert the foot while maintaining a plantarflexed and everted posture to the foot so as
to reduce inversion by the anterior tibial tendon that may mask PTT weakness. This
maneuver typically demonstrates weakness of the PTT compared with the contralateral side.
In the early stages of disease, deformity of the foot remains flexible and
correctible; however, in later stages the deformity remains fixed. The best way to
evaluate the ability to correct a foot deformity to a more normal posture is on
seated examination. Hindfoot valgus is typically corrected through the subtalar
joint by grasping the calcaneus with one hand and internally rotating the calcaneus
while stabilizing the ankle and talus with the other hand. If the hindfoot can be
corrected to a normal posture through the subtalar joint and the midfoot can be
corrected through the transverse tarsal joint, then it is felt to be a correctible
deformity. At this time, while holding the hindfoot in neutral, any gastrocnemius or
soleus contracture can be assessed with the use of the Silfverskild test.
Persistent hindfoot valgus and collapse of the medial column eventually results
in forefoot varus, which may be exhibited by significant first ray elevation or global
forefoot varus at the transverse tarsal joint.16 Deformity through the medial column
should be evaluated with the patient in the seated position. The hindfoot is placed
in a neutral position by centering the navicular over the talar head.17 The
deformity is confirmed by palpating the medial border of the foot and the
relationship of the navicular tuberosity to the talar head. As the hindfoot is held in
this position with one hand, the opposite hand is used to passively bring the ankle
to neutral dorsiflexion by placing force on the plantar aspect of the fourth and fifth
metatarsal heads. At this point, the relationship of the first and fifth metatarsals is
evaluated by viewing the foot head-on to determine the degree of elevation of
the first ray relative to the fifth ray which is known as the degree of forefoot varus
(Fig. 1).18
Stability of the first ray is assessed with the patient in a seated position by
stabilizing the lesser four metatarsals with one hand and then using the opposite hand
to manipulate the first metatarsal. The dorsiflexion and plantarflexion of the first ray in
relation to the rest of the foot is assessed to determine if there is excessive motion or
if there is pain or crepitus with range of motion.
In addition to the physical examination, radiographs play an important role in
determining the etiology of AAFD as well as assessing the deformity of a patients
foot. Weight-bearing radiographs of the foot and ankle should be obtained for proper
evaluation. On a normal weight-bearing anteroposterior and lateral radiograph, the
calcaneocuboid joint should be at the same level as the TN joint. With AAFD, the foot
radiographs show the appearance of a short lateral column compared with the medial

Medial Column Procedures

Fig. 1. Photograph comparing (A) forefoot varus deformity with the first metatarsal head
elevated above the fifth metatarsal head in the coronal plane and (B) normal forefoot
alignment.

column, with the TN joint approximately 3 to 5 mm distal to the calcaneocuboid


joint.19 Additionally, the anteroposterior radiograph helps to quantify forefoot abduction through evaluation of the TN coverage angle. This is measured by determining the
amount of the talar head that is covered by the navicular, with a smaller degree of
coverage corresponding with increased forefoot abduction.
On the weight-bearing lateral radiograph, the medial column of the foot can be
carefully evaluated. The lateral talofirst metatarsal angle, also known as Mearys
angle, is measured between the line drawn representing the longitudinal axis of the
talus and that of the first metatarsal. Normal is considered measurements from 5 to
5 degrees. Measurements less than 5 degrees are considered to have pes planus.
Another angular measurement that can be measured on a weight-bearing lateral
radiograph is the calcaneal pitch. This is the angle formed by the intersection of a line
drawn along the plantar surface of the calcaneus and a line drawn parallel to the
ground. Normal measurement for calcaneal pitch is 10 30, thus less than 10 would
be considered abnormally flat (Fig. 2).

Fig. 2. Lateral weight-bearing radiograph of a foot demonstrating Mearys angle (normal


5 to 5) and calcaneal pitch angle (normal, 10 to 30).

285

286

McCormick & Johnson

In addition to angular measurements, the joints of the medial column should be


carefully evaluated on the weight-bearing lateral radiograph. Instability at the TMT,
NC, or TN joints may result in asymmetry at the joint where there may be a relative
plantar gap at the joint space or subluxation. Further, the medial column joints may
show evidence of degenerative change such as joint space narrowing, subchondral
sclerosis, or osteophytes. Identifying the unstable or arthritic joint along the medial
column is important because this information will help guide the surgical treatment
plan if nonoperative modalities fail.
Nonoperative treatment for medial column deformity or instability with a foot
orthosis may be helpful, especially if the patient has a flexible or correctible deformity
as characterized on physical examination. A custom-molded, semi-rigid foot orthosis
for arch support is posted at the medial heel and the lateral forefoot to help correct
foot pronation and forefoot varus. If the forefoot varus is rigid, then an accommodative, semi-rigid molded orthosis with medial forefoot posting is used to support the
forefoot in the varus posture. If these options are unsuccessful, or if the patient has
a more rigid deformity that is not easily corrected with an orthosis, the patient can use
a custom molded brace such as a plastic and leather composite lace up brace (ie, an
Arizona brace) or a rigid anklefoot orthosis with a molded foot orthosis inside the
brace.
For AAFD, there are numerous surgical options that include tendon repair, tendon
transfer, osteotomies, fusions, and combinations of these procedures.2,514 The
operative procedures chosen must address the etiology of the AAFD and, most
important, must address all the components of the deformity. If the deformity is
flexible or passively correctible on physical examination and not associated with
degenerative joint disease, then a reconstructive option utilizing osteotomies can be
considered. If the deformity is rigid or not passively correctible, then one must employ
a triple arthrodesis (fusion of the subtalar, TN, and calcaneocuboid joints) to correct
the deformity appropriately.
With flexible deformity, most commonly PTT dysfunction, the reconstructive
procedure includes a medial soft tissue reconstruction that involves a flexor digitorum
longus tendon transfer to the navicular to augment or replace the PTT in addition to
repair or imbrication of the spring ligament if needed. Additionally, the gastrocnemius
and soleus complex is addressed with lengthening as needed to decrease the
hindfoot valgus force and resultant influence on the forefoot deformity.20,21 Reconstruction or repair of the medial soft tissue structures helps to restore function to the
foot; however, the underlying flatfoot deformity would eventually result in attenuation
and failure of the repair.22 As such, it is important to correct any underlying deformity
with bony procedures.
Options for bony correction commonly include a medial displacement calcaneal
osteotomy to primarily correct the hindfoot valgus deformity or a lateral column
lengthening to correct forefoot abduction. However, these procedures do not address
the forefoot varus component of the AAFD.23 Even a triple arthrodesis may not
completely correct forefoot varus if the deformity is distal to the TN joint. Without
appropriate correction of the flatfoot deformity, the patient retains some level of
deformity and has altered biomechanics with excessive weight-bearing on the lateral
border of the foot and possible overload of the deltoid ligament of the ankle.
Numerous options exist for correction of forefoot varus through the medial
column.16 The appropriate choice is based on proper identification of the location of
deformity and identification of abnormal or arthritic joints. The focus of this article is
on two such medial column procedures: the Cotton osteotomy (plantarflexion
opening wedge medial cuneiform osteotomy) and the first TMT fusion.

Medial Column Procedures

Fig. 3. Lateral weight-bearing radiographs demonstrating (A) normal forefoot posture, (B)
the contralateral foot of the same patient with AAFD whose lateral radiograph is taken in a
weight-bearing stance with hindfoot valgus, and (C) the same patient with AAFD whose
lateral radiograph was taken with the hindfoot repositioned into neutral. Note the forefoot
varus, which becomes quite evident as the heel is positioned in neutral, thereby elevating the
medial column. The first metatarsal head is now elevated and the fifth metatarsal head is
now able to be visualized on the lateral radiograph.

THE COTTON MEDIAL CUNEIFORM OSTEOTOMY

In 1936, Cotton5 described a procedure to assist in correction of the flatfoot deformity


that used an opening wedge medial cuneiform osteotomy to plantarflex the first ray.
With this procedure, he theorized that the triangle of support would be restored to
the foot and allow the patient to have improved function by restoring the mechanics
of weight-bearing.5 Since Cottons original report, little has been written on the use of
this medial cuneiform osteotomy as part of flatfoot deformity correction.5,10
Indications

A plantarflexion opening wedge medial cuneiform osteotomy should be used as an


adjunct to reconstruction of a flatfoot deformity. Primarily it is used to correct forefoot
varus with elevation of the medial column where the deformity is located at the first
TMT joint or the NC joint.17 This forefoot varus is typically identified preoperatively on
physical examination and is further evaluated with preoperative weight-bearing
radiographs (see Fig. 1; Fig. 3).
The Cotton osteotomy is a joint-sparing procedure; thus, the first TMT joint should
be stable or have only mild instability on physical examination. Further, there should
not be any radiographic evidence of joint abnormality such as plantar gapping on the
lateral weight-bearing radiograph or findings of significant osteoarthritis. If the lateral
x-ray demonstrates marked dorsal subluxation of the first metatarsal in relation to the
medial cuneiform or if there are findings of arthritis, cuneiform osteotomy is contraindicated.17 In these cases, arthrodesis of the first TMT joint is indicated. Further
contraindications to the Cotton osteotomy include deformity that is greater than what
a 5- to 8-mm plantarflexion bone block can correct.17 In these situations, more
proximal procedures such as reduction and fusion of the NC or TN joints are

287

288

McCormick & Johnson

Fig. 4. A dorsal longitudinal incision is used that is centered over the medial cuneiform. Care
is taken to stay medial to the dorsalis pedis artery and deep peroneal nerve.

indicated. Additionally, a fixed forefoot deformity through the transverse tarsal joints
should be addressed with correction at this more proximal level rather than through
a cuneiform osteotomy.
Most commonly, the Cotton osteotomy is performed as a last step in surgical
reconstruction of the AAFD. Typically, the other hindfoot osteotomies would have
been completed and fixed in their corrected position. After the medial soft-tissue
repair and tendon transfers are prepared for fixation, the static posture of the newly
corrected foot is evaluated. The amount of residual forefoot varus is assessed both
clinically and with a lateral intraoperative x-ray; and if significant deformity remains, it
should be corrected to restore the triangle of support.4
Technique

The patient is positioned supine on the operating room table for the flatfoot
reconstruction procedure. A bump is placed under the ipsilateral buttock to
internally rotate the affected leg and a tourniquet is placed around the thigh. If the
surgeon elects to use an autograft, the donor iliac crest is prepped and draped
along with the operative leg. Under tourniquet, a dorsal longitudinal incision is
made over the level of the medial cuneiform joint (Fig. 4). The skin and
subcutaneous tissues are carefully dissected down to the level of the extensor
hallucis longus (EHL) and extensor hallucis brevis (EHB) tendons. The EHL is
retracted medially and the EHB is retracted laterally. Care is taken to be sure the
dorsalis pedis artery and deep peroneal nerve are lateral to the field of work. The
dorsal portion of the medial cuneiform is exposed.

Medial Column Procedures

Fig. 5. A metallic instrument is used to identify the proposed location of the transverse
osteotomy of the medial cuneiform.

At this point, the C-arm fluoroscopy is used to identify the mid-portion of the
cuneiform. It is at this level that the osteotomy is performed (Fig. 5). A microsagittal
saw is then used to create a transverse osteotomy from dorsal to plantar through the
midportion of the medial cuneiform (Fig. 6). On lateral C-arm images, this tends to be
at or just proximal to the level of the second TMT joint. The saw is used to cut the bone
up to, but not through, the plantar cortex of the cuneiform. An osteotome is then used
to propagate the cuneiform osteotomy in a greenstick manner. It is important to leave
the plantar periosteum and ligamentous attachments intact so that the osteotomy
does not risk displacement or distraction. That same osteotome is then pulled distally
and used as a lever to plantarflex the first ray through the newly created osteotomy
(Fig. 7). The foot is then evaluated both clinically and with lateral C-arm imaging to
determine the amount of correction that is needed to restore Cottons triangle of
support and the width of the osteotomy gap is measured. Generally, a 5- to 8-mm
wedge of bone is needed to plantarflex the first metatarsal so that clinically and
radiographically it rests in line with the fifth metatarsal head.
The graft is obtained next. A tricortical iliac crest wedge has been described,
although many types of bone grafts may suffice, and it is surgeons preference
whether to employ autograft or allograft. The graft is cut into a triangular wedge with
a microsagittal saw to the appropriate size, so that the grafts cancellous surfaces
opposes the cancellous surfaces of the native cuneiform. The cortical surfaces of the
graft should rest dorsally, medially, and laterally. A narrow osteotome is then used to
lever open the graft site and the graft is gently impacted in place with a bone tamp.
Fixation of the graft is the surgeons preference and may include a K-wire, a single
cannulated screw, a compression staple, or a small plate (Fig. 8). The authors
preference is a single, 0.062-inch smooth K-wire for percutaneous fixation across the
osteotomy, which is removed at 4 to 6 weeks. Some surgeons use no internal fixation.
Recently, a wedge plate as well as a metallic wedge have been developed and
marketed; however, we have no experience using these devices.24 The wounds are
closed and the patient is splinted. The postoperative protocol may depend on the
other concomitant procedures performed at the time of the Cotton osteotomy.
Typically, the patient remains nonweight-bearing and casted for a total of 6 weeks
from the time of surgery. At that point, assuming the radiographs show signs of

289

290

McCormick & Johnson

Fig. 6. (A) A transverse osteotomy is made through the midportion of the cuneiform using a
microsagittal saw. (B) Drawing demonstrating the typical location of the cuneiform osteotomy at or just proximal to the level of the second tarsometatarsal joint. (C) Drawing
demonstrating the location of the cuneiform osteotomy. Osteotomy is made perpendicular
to the long access of the bone.

healing and the patient has no significant tenderness, the patient is allowed to bear
weight in a protective boot. With standard progression, they begin weaning from the
boot at about 10 weeks after surgery (Fig. 9).
Results

In his original article, Cotton stated that the operation is simple, not painful, and . . .
in the short series of cases done since I devised this operation, there has been no
trouble in any.5 More recent literature has supported this claim and shown the
procedure to be straightforward and predictable, with minimal morbidity. In one

Medial Column Procedures

Fig. 7. An osteotome is inserted into the cuneiform osteotomy and used to lever the first ray
into plantarflexion, correcting forefoot varus. The first metatarsal head is corrected until it
comes in line with the fifth metatarsal head in the coronal plane.

clinical review of the Cotton procedure, Hirose and Johnson reviewed a series of 16
patients who underwent correction of their residual forefoot varus with a plantarflexion medial cuneiform osteotomy.17 They utilized iliac crest autograft and internal
fixation and found no incidence of nonunion or malunion. At the time of follow-up, all
patients had mild to no pain with ambulation. Further, comparison of preoperative to
postoperative weight-bearing radiographs showed an average improvement in
Mearys angle of 14 (13 preoperative, 1 postoperative), demonstrating significant
correction of forefoot varus.17
In another recent clinical series, Lutz and Myerson25 reported on 101 medial
cuneiform osteotomies performed in conjunction with a comprehensive flatfoot
reconstructive procedure. This series did not report any nonunions, supporting the
thought that the Cotton osteotomy heals predictably. Further, they demonstrated a
statistically significant rate of improvement in Mearys angle from a preoperative
average value of 23 to an average postoperative value of 1.
Biomechanically, research has examined the weight-bearing characteristics of feet
before and after Cotton osteotomy. Scott and colleageus26 used cadaveric specimens to evaluate plantar pressures and loading characteristics after a medial
displacement calcaneal osteotomy, lateral column lengthening, and Cotton osteotomy.26 They found that the plantarflexion medial cuneiform osteotomy did increase
average plantar pressure within the medial forefoot, but did not create off-loading of
the lateral forefoot where the plantar pressures remained similar before and after the
Cotton osteotomy. Benthien and associates27 found conflicting results in a similar
cadaveric study by showing that the lateral forefoot pressure decreased with the
addition of a medial cuneiform osteotomy.
There have been some complications found with the use of this procedure. In
Hirose and Johnsons series, 1 patient had a symptomatic screw that was removed,
but there were no problems with nonunion or residual pain.17 In Lutz and Myersons
larger series,25 there were 10 postoperative complications attributed at least in part to
the Cotton osteotomy: 3 symptomatic screws, 2 bony exostosis, 1 sesamoid pain, 1
plantar fasciitis, 2 lateral column overload symptoms, and 1 recurrence of flatfoot
deformity. Overall, the Cotton osteotomy has proven straightforward and useful in
correcting residual forefoot varus in AAFD and should be carefully considered when

291

292

McCormick & Johnson

Fig. 8. (A) Intraoperative photograph of the medial cuneiform osteotomy with graft in place.
(B) Diagram of the foot from a dorsal view demonstrating the location of the interposition
graft with screw fixation. (C) Diagram of the lateral foot demonstrating the insertion of the
interposition bone graft with an oblique partially threaded screw for fixation.

there is no abnormality at the first TMT joint. Further, it allows for preservation of the
first TMT joint and NC joint, permitting normal motion through the foot after
reconstruction of Cottons triangle of support.
TMT JOINT FUSION

In 1934, Lapidus28 described a procedure for use in correction of hallux valgus where
a distal soft-tissue procedure and medial eminence resection was performed along
with arthrodesis of the first and second TMT joints. Since then, it has evolved to
include arthrodesis of only the first TMT joint.29 Although its original intent was for use
in correcting hallux valgus, surgeons have expanded the indications of the first TMT
arthrodesis by using it to correct medial column deformity in AAFD.20

Medial Column Procedures

Fig. 9. (A) Preoperative lateral weight-bearing radiograph of a foot with AAFD. (B) Immediate postoperative C-arm image after surgical correction. The patient has undergone a medial
displacement calcaneal osteotomy, a lateral column lengthening with interposition graft, a
medial soft tissue reconstruction with flexor digitorum longus transfer to the navicular, and
a Cotton osteotomyfixed with a single, 0.062-inch K-wire. (C) Lateral weight-bearing
radiograph after reconstruction has healed and patient is ambulatory.

Indications

Much like the Cotton osteotomy, the first TMT arthrodesis should be used as an
adjunct to a comprehensive flatfoot reconstruction. It is used to correct residual
forefoot varus in the AAFD when the deformity is identified to be primarily through the
first TMT joint. As with the Cotton osteotomy, indications for this procedure are
determined after careful preoperative physical examination and evaluation of preoperative weight-bearing radiographs. The first TMT arthrodesis should be employed
when forefoot varus remains after correction of the hindfoot and there is clinical or
radiographic evidence of significant joint arthrosis or instability. On physical examination, this is identified as gross instability or hypermobility of the first TMT joint or
pain and crepitus with range of motion of the joint. Radiographically, abnormality of
the first TMT joint is seen as asymmetric plantar gapping on weight-bearing lateral
view, joint malalignment, or subluxation on weight-bearing lateral view (Fig. 10), or
evidence of arthritic changes at the first TMT joint, such as joint space narrowing or
sclerosis. With these findings, a joint-sparing procedure such as the Cotton osteotomy is no longer a predictable option because of the concern for residual joint
instability or pain. As with the Cotton osteotomy, however, it is important to assess
whether a first TMT fusion will correct the deformity sufficiently or whether a more
proximal procedure such as a NC or TN fusion is necessary.
The first TMT fusion is typically performed as a last step in the reconstruction of the
AAFD. After the hindfoot correction has been completed, the posture of the foot
should be evaluated both clinically and radiographically. If significant forefoot varus
deformity remains and there is gross instability, hypermobility, or degenerative
change at the first TMT joint, then a first TMT fusion can be used to correct the
deformity.

293

294

McCormick & Johnson

Fig. 10. Weight-bearing lateral radiograph demonstrating instability at the first TMT joint.

Technique

The patient is positioned supine, with a bump under the ipsilateral buttock and a
tourniquet around the thigh. The leg is prepped and draped in exactly the same
manner as for the Cotton osteotomy. A dorsal incision is made over the level of the
first TMT joint and the skin and subcutaneous tissues are carefully dissected until the
EHL and EHB tendons are visualized and retracted medially and laterally, respectively. Care is taken to ensure the dorsalis pedis artery and deep peroneal nerve
remain lateral to the surgical site, and, if they are visualized, they are retracted laterally
with the EHB tendon. At this point, the first TMT joint is identified and a longitudinal
incision is made through the joint capsule, which is then elevated medially and
laterally to expose the joint. If there is any concern about identification of the first TMT
joint, a C-arm fluoroscopy unit can be used for verification.
Once the joint is identified and exposed, the joint surfaces are prepared by
removing all of the cartilage from the proximal aspect of the first metatarsal and the
distal aspect of the medial cuneiform. This is completed with the use of instruments
such as osteotomes, curettes, and rongeurs. It should be noted that the first TMT joint
is quite deep from dorsal to plantar, measuring on the order of 3 cm. As such, care
should be taken to prepare the full extent of the joint and avoid dorsiflexion through
the arthrodesis, which would be created by incomplete preparation of the most
plantar aspect of the joint. A lamina spreader or distraction apparatus using pins in the
bone is helpful to gain access to the entire first TMT joint. Once all of the cartilage has
been removed, the subchondral joint surface is further prepared by drilling or petaling
the bone to increase the bleeding surface area for fusion.
After the surfaces are prepared, the first TMT joint must be reduced to the
appropriate position of fusion. Clinical examination and C-arm fluoroscopy images
are used to ensure that the medial column is being appropriately plantarflexed to
correct the forefoot varus. As described for the Cotton osteotomy, this is accomplished by ensuring that the first metatarsal head rests in line with the fifth metatarsal
head clinically and radiographically. If anatomic reduction of the first TMT joint does
not correct the forefoot varus, resection of a small plantar wedge from the metatarsal
base or the use of an interposition wedge graft much like a Cotton osteotomy may
augment the correction. As with the plantarflexion cuneiform osteotomy, if a tricortical
iliac crest autograft or allograft is used, the joint surfaces are flattened with a
microsagittal saw and a wedge is cut to the size that creates the appropriate amount
of plantarflexion through the first TMT fusion site. An osteotome is used to lever open
the joint into an improved, plantarflexed posture and the graft is gently impacted into
place. As an alternative to an opening wedge graft, one might consider a plantar

Medial Column Procedures

Fig. 11. (A) Preoperative lateral weight-bearing radiograph of a patient with AAFD and first
TMT instability. (B) Lateral weight-bearing radiograph of a patient after reconstruction has
healed. The patient has undergone a medial displacement calcaneal osteotomy, a lateral
column lengthening with interposition graft, a medial soft tissue reconstruction with flexor
digitorum longus transfer to the navicular, and a first TMT joint arthrodesisfixed with
crossed lag screws.

closing wedge osteotomy at the level of the first TMT joint. The authors have minimal
experience with plantar closing wedge osteotomy of the cuneiform for the correction
of forefoot varus and have been able to correct the majority of medial column
deformities with a 5- to 8-mm dorsal bone wedge. However, for a more severe
dorsiflexion deformity of the medial column with a normal first TMT and TN joint, we
would favor either a double bone block opening wedge osteotomy as has been
described for a dorsal bunion correction30 or a correction and stabilization through
the naviculo-first cuneiform joint as similar to the Miller flatfoot procedure, the
modified HokeMiller or the Durham plasty procedures.7,9,31
Fixation for the first TMT fusion is most commonly dual compression crossed
screws or a single compression screw with a dorsal plate. The soft tissues and
incision are then carefully closed and the patient is splinted. The postoperative
protocol varies based on concomitant procedures that have been performed;
however, typically the patient is casted and nonweight-bearing for 6 weeks from the
time of their operation. At that point, they are typically transitioned to a boot for
progressive weight bearing, assuming that the clinical and radiographic examinations
do not raise concern for delayed healing. At 10 to 12 weeks after their surgery, the
patient is usually transitioning from their boot to a shoe and progressing with activities
as clinically tolerated (Fig. 11).
Results

In the past, there has been apprehension about performing the first TMT arthrodesis
because of concerns for rates of nonunion. Some authors have cited rates as high as
12%; however, a recent paper by Thompson and colleagues32 had a better union
rate. In their review of 182 patients who had undergone fusion of the first TMT joint for
correction of hallux valgus or as a component to reconstruction of AAFD, the
nonunion rate was only 4%. Of those, only half were symptomatic and required
revision surgery. Further analysis shows that, of those patients who underwent first
TMT fusion as part of a flatfoot reconstruction, there were no instances of nonunion.
They attributed their low rate of nonunion to careful bone preparation, addition of
bone graft, and the use of Achilles or gastrocnemius lengthening procedures as
indicated to unload stresses through the forefoot.
Cadaveric biomechanical studies have investigated various constructs for Lapidus
arthrodesis. One study showed no significant difference between locked plate fixation

295

296

McCormick & Johnson

and dual crossed lag screw fixation in terms of rigidity.33 Two studies, however,
showed greater rigidity at first TMT fusion site with a locked plate as compared with
dual crossed lag screws.34,35 DeVries and associates36 compared the clinical
outcome of these fixation methods and found a significant increase in union rate when
a locked plate was used as compared with the dual crossed screw technique. The
dual screw technique did have an acceptable union rate of almost 90%, however,
suggesting it to be a viable option. In our practice, we customarily use a lag screw and
dorsal locked plate.
Although Thompson and colleagues32 showed a high rate of union that may not be
attainable by all surgeons, at the least it suggests that the procedure is safe and
reasonable. Outside of the cited nonunions, there were 6 superficial wound infections
treated with oral antibiotics and 2 local wound breakdowns requiring wound care in
addition to antibiotic treatment. No deep infections or incidence of more severe
complications were identified. The procedure does sacrifice mobility of the first TMT
joint; however, in the face of first TMT hypermobility or arthritic change at the first TMT
joint, Lapidus fusion is a procedure that does quite well as an adjunct to a
comprehensive flatfoot reconstruction.
SUMMARY

AAFD is a complex problem with a wide variety of treatment options. No single


procedure or group of procedures can be applied to all patients with AAFD because
of the variety of underlying etiology and grades of deformity. As the posture of the foot
progresses into hindfoot valgus and forefoot abduction through attenuation of the
medial structures of the foot, the medial column begins to change shape. The first ray
elevates and the joints of the medial column may begin to collapse. Careful physical
examination and review of weight-bearing radiographs determines which patients
have an associated forefoot varus deformity that may require correction at the time of
flatfoot reconstruction.
Correction of an AAFD requires a combination of soft-tissue procedures to restore
dynamic inversion power and bony procedures to correct the hindfoot and midfoot
malalignments. If after these corrections forefoot varus deformity remains, the
surgeon should consider use of a medial column procedure to recreate the triangle
of support of the foot that Cotton described.5 If the elevation of the medial column
is identified to be at the first NC or the first TMT joint, then the joint should be carefully
examined for evidence of instability, hypermobility, or arthritic change. If none of
these problems exist, then the surgeon can consider use of the joint-sparing Cotton
medial cuneiform osteotomy to correct residual forefoot varus. However, if instability,
hypermobility, or arthritic change is present, then the surgeon should consider use of
an arthrodesis of the involved joint to correct residual forefoot varus. Either procedure
provides a safe and predictable correction to the medial column as part of a
comprehensive surgical correction of AAFD.
REFERENCES

1. Funk DA, Cass JR, Johnson KA. Acquired adult flat foot secondary to posterior
tibial-tendon pathology. J Bone Joint Surg Am 1986;68:95102.
2. Pedowitz WJ, Kovatis P. Flatfoot in the Adult. J Am Acad Orthop Surg 1995;3:293
302.
3. Mann DA. Acquired flat foot in adults. Clin Orthop 1983;181:46 51.
4. Myerson MS. Acquired flat foot deformity. J Bone Joint Surg Am 1996;78:780 92.
5. Cotton FJ. Foot statics and surgery. N Engl J Med 1936:353 62.

Medial Column Procedures

6. Duncan JW, Lovell WW. Modified Hoke-Miller flatfoot procedure. Clin Orthop Relat
Res 1983:24 7.
7. Hoke M. An operation for the correction of extremely relaxed flat feet. J Bone Joint
Surg 1931:773.
8. Johnson JE, Cohen BE, DiGiovanni BF, et al. Subtalar arthrodesis with flexor digitorum longus transfer and spring ligament repair for treatment of posterior tibial tendon
insufficiency. Foot Ankle Int 2000;21:7229.
9. Miller OL. A plastic flatfoot operation. J Bone Joint Surg 1927:84.
10. Mosca VS. Calcaneal lengthening for valgus deformity of the hindfoot. Results in
children who had severe, symptomatic flatfoot and skewfoot. J Bone Joint Surg Am
1995;77:500 12.
11. Myerson MS, Corrigan J, Thompson F, et al. Tendon transfer combined with calcaneal osteotomy for treatment of posterior tibial tendon insufficiency: a radiological
investigation. Foot Ankle Int 1995;16:712 8.
12. Sands A, Grujic L, Sangeorzan BJ, et al. Lateral column lengthening through the
calcaneocubiod joint: an alternative to triple arthrodesis for correction of flatfoot. In:
AOFAS Annual Meeting. Orlando (FL), 1995.
13. Toolan BC, Sangeorzan BJ, Hansen ST Jr. Complex reconstruction for the treatment
of dorsolateral peritalar subluxation of the foot. Early results after distraction arthrodesis of the calcaneocuboid joint in conjunction with stabilization of, and transfer of
the flexor digitorum longus tendon to, the midfoot to treat acquired pes planovalgus in
adults. J Bone Joint Surg Am 1999;81:1545 60.
14. Hirose CB. The use of plantarflexion opening wedge medial cuneiform osteotomy for
correction of fixed forefoot varus associated with flatfoot deformity. In: AOFAS 33rd
Annual Meeting. New Orleans (LA), 2003.
15. Hiller L, Pinney SJ. Surgical treatment of acquired flatfoot deformity: what is the state
of practice among academic foot and ankle surgeons in 2002? Foot Ankle Int
2003;24:7015.
16. Cohen BE, Ogden F. Medial column procedures in the acquired flatfoot deformity.
Foot Ankle Clin 2007;12:28799.
17. Johnson JE. Plantarflexion opening wedge cuneiform-1 osteotomy for correction of
fixed forefoot varus. Techniques in Foot and Ankle Surgery 2004;3:2 8.
18. Alexander IJ. The foot: examination and diagnosis. New York: Churchill Livingstone;
1990.
19. Stephens HM, Walling AK, Solmen JD, et al. Subtalar repositional arthrodesis for adult
acquired flatfoot. Clin Orthop Relat Res 1999;365:69 73.
20. Habbu R HS, Anderson JG, Bohay DR. Operative correction of arch collapse with forefoot
deformity: a retrospective analysis of outcomes. Foot Ankle Int 2011;32:764 73.
21. Coughlin MJ, Kaz A. Correlation of Harris mats, physical exam, pictures, and radiographic measurements in adult flatfoot deformity. Foot Ankle Int 2009;30:604 12.
22. Michelson J,Conti S, Jahss MH. Surviving analysis of tendon transfer surgery for
posterior tibial tendon rupture Orthop Trans 1992;16:30 1.
23. Neufeld SK, Myerson MS. Complications of surgical treatments for adult flatfoot
deformities. Foot Ankle Clin 2001;6:179-91.
24. League AC, Parks BG, Schon LC. Radiographic and pedobarographic comparison of
femoral head allograft versus block plate with dorsal opening wedge medial cuneiform
osteotomy: a biomechanical study. Foot Ankle Int 2008;29:922 6.
25. Lutz M, Myerson M. Radiographic analysis of an opening wedge osteotomy of the
medial cuneiform. Foot Ankle Int 2011;32:278 87.

297

298

McCormick & Johnson

26. Scott AT, Hendry TM, Iaquinto JM, et al. Plantar pressure analysis in cadaver feet after
bony procedures commonly used in the treatment of stage II posterior tibial tendon
insufficiency. Foot Ankle Int 2007;28:114353.
27. Benthien RA, Parks BG, Guyton GP, et al. Lateral column calcaneal lengthening,
flexor digitorum longus transfer, and opening wedge medial cuneiform osteotomy for
flexible flatfoot: a biomechanical study. Foot Ankle Int 2007;28:70 7.
28. Lapidus PW. The operative correction of the metatarsus primus varus in hallux valgus.
Surg Gynecol Obstet 1934;58:18391.
29. Sangeorzan BJ, Hansen ST Jr. Modified Lapidus procedure for hallux valgus. Foot
Ankle 1989;9:262 6.
30. Johnson JE, Thomson AB, Yu JR. Double bone-block arthrodesis for the correction of
dorsal bunion deformity. Techniques in Foot and Ankle Surgery 2007;6:170 4.
31. Caldwell GD. Surgical correction of relaxed flatfoot by the Durham flatfoot plasty. Clin
Orthop 1953;2:221 6.
32. Thompson IM, Bohay DR, Anderson JG. Fusion rate of first tarsometatarsal arthrodesis in the modified Lapidus procedure and flatfoot reconstruction. Foot Ankle Int
2005;26:698 703.
33. Gruber F, Sinkov VS, Bae SY, et al. Crossed screws versus dorsomedial locking plate
with compression screw for first metatarsocuneiform arthrodesis: a cadaver study.
Foot Ankle Int 2008;29:92730.
34. Klos K, Gueorguiev B, Muckley T, et al. Stability of medial locking plate and compression
screw versus two crossed screws for lapidus arthrodesis. Foot Ankle Int 2010;31:15863.
35. Scranton PE, Coetzee JC, Carreira D. Arthrodesis of the first metatarsocuneiform
joint: a comparative study of fixation methods. Foot Ankle Int 2009;30:3415.
36. DeVries JG, Granata JD, Hyer CF. Fixation of first tarsometatarsal arthrodesis: a
retrospective comparative cohort of two techniques. Foot Ankle Int 2011;32:158 62.

Minimizing the Role of Fusion in


the Rigid Flatfoot
Ross Taylor, MDa,*, V. James Sammarco, MD,b

KEYWORDS
Flatfoot Fusion alternatives Osteotomy Reconstruction
KEY POINTS
Equinus contracture, ankle and hindfoot valgus, midfoot abduction, and forefoot
varus must each be individually considered with the ultimate goal of not only
deformity correction, but preservation of function in the ankle, subtalar, hindfoot, or
midfoot joints.
Although triple arthrodesis affords predictable correction and pain relief, the longterm sequelae of extended hindfoot fusions include arthritis and often the need for
further, more extensive fusion procedures.
Satisfactory results can be achieved in the rigid flatfoot by limiting fusion to joints that
are arthritic, and correcting associated deformity with osteotomy and soft tissue
reconstruction.

There has been a proliferation of treatment options for flexible flatfoot deformity,
but in cases where the deformities are more rigid and fixed, triple arthrodesis
remained the most common surgical technique employed for treatment. Although
triple arthrodesis can afford excellent deformity correction, it can be associated
with disability owing to the stiffness created by the procedure. Patients may
complain of continued pain with ambulation and extended weight bearing or when
walking long distances. In addition, after triple arthrodesis, symptomatic arthritis
often develops with time in the ankle, midfoot, and intertarsal joints. Extending the
arthrodesis when progressive arthritis develops rarely improves the patients
function to their expectation. Converting a triple arthrodesis to a pan-talar
arthrodesis leads to a very stiff and often chronically painful leg and, in the
authors opinion, should be avoided. Recognizing both the early and late problems
associated with extended hindfoot fusions, increased interest in using the
techniques developed for correction of flexible flatfoot has occurred. This dialogue

Coastal Orthopedics, 2376 Cypress Circle Suite 300, Conway, SC 29526, USA; b Cincinnati Sports
Medicine & Orthopaedic Center, 10663 Montgomery Road, Cincinnati, OH, USA
* Corresponding author.
E-mail address: rtaylor897@sc.rr.com
Foot Ankle Clin N Am 17 (2012) 337349
http://dx.doi.org/10.1016/j.fcl.2012.03.010
1083-7515/12/$ see front matter 2012 Elsevier Inc. All rights reserved.

foot.theclinics.com

338

Taylor & Sammarco

is intended to provide alternatives to fusion surgery for the treatment of rigid


flatfoot deformity while discussing the various osteotomy and soft tissue reconstructive options that may allow the surgeon to either avoid fusion or perform a
more limited fusion than might otherwise be required.
DEFORMITY AND ARTHRITIS

Deformity correction without arthrodesis requires a conceptual change for most


surgeons, and for patients as well. As a surgeon, one wants to offer the patient a
reliable way to fix the problem at hand. In both the rigid and flexible flatfoot,
arthrodesis is undoubtedly the most predictable way to achieve a plantigrade foot.
When done correctly, there is little chance of deformity recurrence, and pain relief is
typically good. The longer term problems associated with arthrodesis can be easily
overlooked, but need to be balanced with the need for successful treatment today.
This balance needs to be explained to the patient, who must be aware that surgery in
the future may be inevitable whether the triple arthrodesis is done primarily or whether
arthrodesis is deferred to a later time in life when the patient is not as active. Further
individualization of the procedure needs to be done based on the patients age,
activity level, and medical comorbidities.
As the treatment paradigm shifts from fusion to osteotomy, the surgeon must
distinguish between components of the flatfoot deformity that are arthritic versus
those that are supple and nonarthritic. Once determined, the surgeon may restrict
fusion only to those joints that are arthritic while correcting nonarthritic components of the deformity by osteotomy and soft-tissue reconstruction. Equinus
contracture, ankle and hindfoot valgus, midfoot abduction, and forefoot varus
must each be individually considered with the ultimate goal of not only deformity
correction, but preservation of function in the ankle, subtalar, hindfoot, or midfoot
joints.
Ankle deformity must be defined before any surgery done to correct flatfoot.
Coronal plane valgus deformity of the mortise need to be addressed and may require
osteotomy of the tibia to correct bony malalignment, or possibly reconstruction of the
attenuated deltoid ligament complex when the talus is tilted in a neutral mortise.
Almost all flatfoot correction is associated with equinus owing to posterior muscular
contracture, and this is addressed by muscular or tendinous lengthening. Subtalar
joint valgus may be rigid even in the absence of arthritis. Conditions such as tarsal
coalition or other congenital deformities of the calcaneus may be present with little
arthritis of the subtalar joint and may be salvageable without fusion by resection,
osteotomy, or both (Fig. 1). The traverse tarsal joint is comprised of the talonavicular
and calcaneocuboid joints, and deformity here is usually defined by collapse of the
arch in the sagittal plane and abduction of the foot through the joint in the axial plane.
Correction of both sagittal and axial deformities of the talonavicular joint may be
possible by spring ligament reconstruction and digital flexor tendon transfer to the
navicular. Further correction can be achieved by lengthening the lateral column
through an osteotomy in the calcaneus. Fixed varus deformity in the medial column
can be corrected by plantar flexion osteotomy of the medial column. In this way, each
deformity segment can be defined, and the surgeon can use what tools are available
for correction of the deformity. In the foot, if arthritis of a joint is present, fusion is
usually recommended at that segment. When ankle arthritis as present, it is
increasingly possible for the surgeon to correct the underlying foot deformity, and
perform a staged ankle replacement.

Minimizing Fusion in the Rigid Flatfoot

Fig. 1. (A, B) A 15-year-old boy with rigid pes planovalgus owing to a middle facet talarcalcaneal coalition. (C, D) Surgery included resection of the coalition, MDCO, lateral column
lengthening, and advancement of the posterior tibial tendon and reefing of the spring
ligament complex. (E, F) The patient underwent bilateral procedures and was able to return
to high school athletics. X-rays 4 years after procedure demonstrate durable correction
without progression of arthritis.

MINIMIZING FUSION: THE ROLE OF OSTEOTOMY AND SOFT-TISSUE


RECONSTRUCTION

The contributions of equinus contracture, ankle and hindfoot valgus, midfoot abduction,
and forefoot varus in rigid flatfoot deformity are presented. Although many of these
deformities have historically been treated with fusion, osteotomy, and soft-tissue reconstructive options for each of these elements of the rigid planovalgus foot are reviewed.

339

340

Taylor & Sammarco

Equinus Contracture

Harris and Beath1 were the first to identify the role of equinus contracture in flatfoot
deformity. With the hindfoot pathologically everted, the Achilles tendon insertion is
not only lateralized, but the triceps surae is relatively shortened. Clinical examination
may paradoxically suggest a relative increase in dorsiflexion as the foot subluxes
laterally and dorsally around the talus with dorsally directed forefoot pressure.
Often, it is only with manual reduction of hindfoot valgus that deficits in ankle
dorsiflexion become apparent. Rigid hindfoot deformity precludes clinical assessment of equinus contracture until the subtalar joint is reduced at the time of fusion
surgery. Numerous authors have supported either tendo-Achilles lengthening
(TAL) or gastrocnemius recession as a component of hindfoot reconstruction.2,3
Lengthening of the triceps surae is, therefore, nearly universally employed by the
authors as a component of flatfoot reconstruction both with and without arthrodesis of the subtalar joint. Although there is little controversy regarding the need
for this procedure to obtain adequate deformity correction there are 2 differing
approaches: Gastrocnemius recession and TAL.
The decision to perform either gastrocnemius recession or TAL at the time of
surgical reconstruction of the rigid flatfoot classically depends on the ability to
neutralize equinus contracture by flexing the knee (Silfverskiold test).4 Knee flexion
relaxes the gastrocnemius, but not the soleus; the former originates proximal to the
knee and the latter distally. An equinus contracture eliminated by knee flexion
indicates an isolated gastrocnemius contracture. Conversely, persistence of equinus
contracture with knee flexion indicates contracture of the entire triceps surae or a
shortened Achilles tendon. Although TAL lengthens the entire triceps surae group,
gastrocnemius recession provides selective lengthening of the superficial muscles
only. Proponents of TAL cite its simplicity, ease, and cosmesis versus a gastrocnemius recession. Unfortunately, TAL has been associated with tendon rupture.5 It is our
experience that overlengthening of the tendon with loss of push-off is common with
TAL in adults, and that the incised Achilles tendon can be a source of persistent pain.
Although gastrocnemius recession has been shown to provide predictable correction
of equinus deformity,6 risks include sural nerve injury and unacceptable cosmesis.
Ankle Valgus

One of the major goals in treatment of flatfoot deformity is the restoration of coronal
plane alignment to the ankle and foot. Coronal plane malalignment may occur either
as a result of valgus angulation of the tibial plafond secondary to congenital tibial
deformity or trauma, or as a result of valgus tilt of the talus due to progression of
deltoid ligament insufficiency. As is the case with hindfoot deformity, ankle valgus
may be either flexible or rigid. This distinction is reflected in Myersons expanded the
classification of posterior tendon insufficiency to include stage IV. Stage IV-A
describes reducible ankle valgus whereas stage IV-B describes rigid ankle valgus.7
Traditionally, arthrodesis of the ankle was used to correct significant coronal or
sagittal plane malalignment. Although deformity correction is assured and the risk of
recurrence is eliminated with this approach, the long-term effects of arthrodesis of the
ankle are concerning, particularly where foot deformity requires additional fusion for
correction. Progressive arthritis of the peritalar joints after ankle arthrodesis may lead
to the eventual need for pantalar arthrodesis, a risk that may be compounded by
preexisting flatfoot deformity. We therefore recommend that in cases of significant
pes planovalgus, all attempts at preserving the ankle with a joint-sparing procedure
be made. With the advent of more successful ankle replacement designs, our

Minimizing Fusion in the Rigid Flatfoot

philosophy has changed from one where the patient is simply braced until fusion is
necessary to one where alignment and ligamentous stability are surgically restored in
preparation for a staged joint replacement. With this being said, it is impressive how
many patients have significant improvement in their symptoms with deformity
correction and restoration of stability despite radiographically significant arthritis of
the ankle.
In cases where the distal tibial plafond has a significant valgus deformity,
osteotomy of the tibia needs to be considered as part of the reconstruction. There are
many techniques to accomplish this including those that use direct realignment by
opening or closing wedge osteotomy, and those that favor progressive correction
with an external fixator. In cases where valgus exists owing to fracture malunion,
deformity correction is straightforward and often can be accomplished by osteotomy
of the tibia at the level of the deformity (Fig. 2). Congenital cases or cases owing to
growth plate disturbance can be more complex as the deformity may be multiplanar.
Computed tomography can be particularly helpful in planning these osteotomies and
helps to define the amount of valgus and plantar flexion requiring correction. We
prefer a closing wedge osteotomy when reconstructing congenital deformity in adults
because metaphyseal osteotomies in this area heal quickly with bone on bone
apposition, and the slight shortening of the limb that occurs with removal of a wedge
of bone decreases soft tissue tension and promotes postoperative motion. The distal
tibial articular angle should be corrected to 3% to 5% of valgus, although a neutral
angle is acceptable. It is important to avoid overcorrection with varus angulation of
the tibial plafond as this may result in unacceptable lateral ankle instability or lateral
column overload of the foot. In cases where significant arthritis is present, or in cases
where rigid equinus accompanies the deformity, an anterior closing wedge osteotomy
can be incorporated to restore neutral sagittal plane alignment of the ankle.
Patients with a normal distal tibial articular angle and reducible valgus tilting of the talus
within the ankle mortis (stage IV-A deformity) may be candidates for joint sparing
reconstruction. Jeng and colleagues8 outlined criteria for joint sparing reconstruction as:
Persistent ankle pain refractory to conservative care;
Rigid hindfoot valgus and forefoot abduction consistent with stage IV adult
acquired flatfoot deformity;
Valgus tilting of the talus within the mortise of at least 3 that is passively
correctible under fluoroscopy; and
No less than 2 mm of lateral tibiotalar cartilage remaining as measured on
standing anteroposterior ankle radiographs.
Joint-sparing procedures involve deltoid ligament reconstruction combined with
distal realignment. Although numerous joint-sparing procedures have been described, they uniformly involve reconstruction of the incompetent deltoid ligament.
Although procedures involving repair and advancement of the native deltoid ligament
have been described, these rely on the repair of diseased tissue and may be
predisposed to failure. Reconstruction of the deltoid ligament has been described
using both host and native tissues. Deland and associates9 performed deltoid
ligament reconstruction using peroneus longus autograft transferred through bone
tunnels to recreate both the superficial and deep components of the deltoid ligament.
Jeng and colleagues8 recently described minimally invasive deltoid ligament reconstruction in combination with triple arthrodesis. Successful outcomes occurred in 5 of
8 patients, underscoring the challenging nature of this condition and the limitations of
surgical reconstruction. Most recently, Haddad and colleagues10 described reconstruction of the superficial and deep deltoid ligaments using tibialis anterior tendon

341

342

Taylor & Sammarco

Fig. 2. (A, B) Preoperative x-rays of a 32-year-old woman who underwent previous surgery
for trauma at age 17 for an open ankle fracture. Note the deficiency of the fibula. She in the
intervening years she had undergone several procedures, including triple arthrodesis. She
presented with equinus and valgus. X-rays and physical examination revealed that the
deformity was at the ankle. (C, D) Postoperative films showing anterior and medial biplanar
closing wedge osteotomy. Despite radiographic arthritis, the patient experienced dramatic
improvement in symptoms.

allograft in 6 cadaveric specimens. Torque testing of reconstructed specimens


revealed equivalent resistance to eversion and external rotation to matched controls.
Although the initial study was cadaveric, this technique has been performed in eight
patients using semitendonosis allograft.
Whereas the goal of restoration of ankle alignment is long-term preservation of
ankle function, often arthritis of the ankle progresses despite successful deformity
correction. Nonetheless, a well-aligned ankle not only improves overall gait mechanics, but may delay the onset of arthritic symptoms as well. Ultimately, conversion to

Minimizing Fusion in the Rigid Flatfoot

ankle replacement or even fusion is greatly facilitated by the normal ankle alignment
imparted by either osteotomy or deltoid ligament reconstruction.
Hindfoot Valgus

Medial displacement calcaneal osteotomy (MDCO) has been used as an adjunctive


procedure in the correction of flexible flatfoot for more than a century. Gleich in 1893
described medializing osteotomy of the calcaneus by excising a medial wedge.
Almost a century later in 1960, Dwyer11 introduced a lateral closing wedge osteotomy
of the calcaneus in the treatment of pes cavus deformity. Koutsogiannis12 in 1971
introduced the idea of a pure medial translational osteotomy of the calcaneus for
flatfoot deformity. Since that time, MDCO has gained near universal acceptance as a
fundamental component of flexible flatfoot reconstruction. Although studies specifically addressing the use of MDCO in the treatment of rigid flatfoot are lacking,
biomechanical principles support its application when combined with both soft-tissue
reconstruction and hindfoot fusion.
Coronal plane alignment of the foot is typically restored at the time of hindfoot
fusion via internal rotation of calcaneus beneath the talus and with simultaneous
reduction of the talonavicular joint. Nonetheless, the degree of obtainable reduction
by this method is limited. Although the amount of acceptable hindfoot valgus after
subtalar arthrodesis has never been precisely quantified, persistent eversion of the
heel has been correlated with ankle degeneration after malaligned hindfoot fusion.
Valgus tibiotalar arthritis was observed after valgus malunion of 6 hindfoot fusions in
a series of patients reviewed by Fitzgibbons.13 Furthermore, after hindfoot fusion,
preexisting deltoid ligament attenuation as well as posterior tibial tendon insufficiency
may act synergistically with residual hindfoot valgus to accelerate postoperative ankle
deformity and arthritis. Resnick and colleagues14 noted a 76% increase in deltoid
ligament strain with a valgus displaced hindfoot in a cadaveric triple arthrodesis model.
Others have noted marked progression of ankle osteoarthritis after triple arthrodesis.15,16
It is plausible to assume that residual hindfoot valgus may play a role in deltoid ligament
failure and progression to valgus arthritis of the ankle after triple arthrodesis. Medialization
of the posterior tuberosity of the calcaneus translates the limb center of gravity toward the
medial malleolus, which not only reduces lateral tibiotalar contact forces, but also
decreases deltoid ligament strain.
Complete reduction of hindfoot valgus at the time of surgery is necessary for rigid
flatfoot deformity, whether surgery includes a subtalar arthrodesis or not. In cases of
subtalar arthrodesis, MDCO should be done if residual hindfoot valgus is greater than
7 after provisional fixation. When hindfoot reconstruction is combined with deltoid
ligament reconstruction for stage IV deformity, correction of the calcaneus to neutral
or 0 is desirable. Inclusion of MDCO as a component of either soft-tissue reconstruction or fusion surgery for the rigid flatfoot may protect the ankle joint from
subsequent valgus deformity and degenerative arthritis, ultimately mitigating the risk
of requiring ankle fusion.
Midfoot Abduction

In the acquired flatfoot, dorsolateral peritalar subluxation of the tarsus occurs as the
posterior tibial tendon and spring ligament fail resulting in midfoot abduction and
hindfoot valgus. Evans17 first demonstrated in the pediatric flatfoot that lateral column
lengthening through the anterior body of the calcaneus improved not only midfoot
abduction but hindfoot valgus as well. Sangeorzan and colleagues18 applied lateral
column lengthening to the correction of adult acquired flatfoot. In 2003, Anderson
subclassified Johnson and Strom stage II into subtypes A and B to quantify the

343

344

Taylor & Sammarco

degree of lateral navicular peritalar subluxation as measured on standing AP


radiographs.19 Subgroup B represents patients with greater than 50% uncovering of
the talar head and those whom require lateral column lengthening for correction of
midfoot abduction.
Much of the controversy surrounding this procedure centers on whether to
lengthen the lateral column of the foot with osteotomy of the calcaneus or by
distraction arthrodesis of the calcaneocuboid joint. Proponents of the former cite the
risk of calcaneocuboid arthritis after lateral column calcaneal lengthening. Cooper
and colleagues20 used a cadaveric model to demonstrate an 8-fold increase in
calcaneocuboid joint pressure after lateral column calcaneal lengthening. Proponents
of osteotomy cite the potential drawbacks of arthrodesis including not only the
diminished hindfoot motion but risk of nonunion as well. Deland and colleagues21
demonstrated a 52% reduction in talonavicular motion and a 30% reduction in
subtalar motion after distraction arthrodesis of the calcaneocuboid joint. Toolan and
colleagues22 reported a 20% rate of nonunion in their series of 41 feet undergoing
distraction arthrodesis of the calcaneocuboid joint combined with medial column
fusions and FDL transfer. In addition to nonunion, lateral pain owing to overcorrection
is more common with distraction arthrodesis of the calcaneocuboid joint than with
lateral column calcaneal lengthening; therefore, we reserve distraction arthrodesis of
the calcaneocuboid joint for cases where significant arthritis of that joint is present.
Forefoot Varus

Hindfoot valgus and midfoot abduction require compensatory supination of the


forefoot to maintain floor contact while standing. Initially flexible, forefoot supination
with time may become rigid, resulting in fixed forefoot varus. Malalignment, instability,
and arthritis may occur through the talonavicular, naviculocuneiform, or first tarsometatarsal joints. Whether forefoot varus is a primary or secondary deformity in
acquired flatfoot deformity is of ongoing debate. There is, however, little controversy
that failure to address fixed forefoot varus at the time of flatfoot reconstruction may
result in a supinated forefoot, lateral column overload, and even fracture of the fifth
metatarsal.
The distinction between rigid and flexible forefoot varus is important in determining
the scope of flatfoot reconstruction. Flexible deformity seldom requires correction
unless medial column instability is present on lateral radiographs. The flexibility of
forefoot varus is easily evaluated preoperatively when hindfoot deformity is supple.
Simultaneous reduction of heel eversion while the ankle is plantar flexed should
reduce the forefoot to neutral with minimal coaxing if forefoot varus is flexible. Fixed
forefoot varus is more difficult to detect when hindfoot deformity is rigid, because
hindfoot reduction may not be possible until correction is done at the time of surgery.
Because correction of coronal plane alignment in the flatfoot requires inversion of the
subtalar joint, hindfoot correction may exacerbate forefoot varus, particularly when
combined with lateral column lengthening. This can be compensated for to a certain
extent by increasing plantarflexion of the medial column at the talonavicular joint if
arthrodesis of that joint is planned. If the talonavicular joint can be spared, however,
forefoot varus can be corrected by medial column osteotomy.
Cotton23 first described the utility of plantar flexion osteotomy of the medial
cuneiform in correcting depression of the medial arch in pes planus. Since then, this
osteotomy has been readily adapted to the reconstruction of congenital flatfoot.
Hirose and Johnson24 combined the Cotton osteotomy with hindfoot reconstruction
in 16 feet with flatfoot deformity, reporting improved radiographic parameters
including lateral talo-first metatarsal angle, calcaneal pitch and medial cuneiform

Minimizing Fusion in the Rigid Flatfoot

height. Benthien and colleagues25 demonstrated not only correction of forefoot


varus but also diminished lateral column overload after lateral column lengthening
with the addition of opening wedge osteotomy of the medial cuneiform in a severe
flatfoot model. Unless gross instability or arthritis of the naviculocuneiform or
tarsometatarsal joints is present, the authors utilize plantar flexion osteotomy of
the medial cuneiform to correct rigid forefoot varus rather than an extended medial
column arthrodesis.
LIMITED HINDFOOT FUSIONS

Osteotomy and soft tissue reconstruction can provide surprisingly good correction of
even rigid planovalgus deformity; however, the presence of arthritic changes of the
subtalar or transverse tarsal joint are a contraindication to this type of joint sparing
reconstruction. Triple arthrodesis remains the procedure of choice when longstanding deformity has progressed to severe arthritis, although this procedure is not
without its own complications. Most notably, the reported progression of arthritis at
the ankle joint after triple arthrodesis has been alarming. Wetmore and Drennan15
reported a radiographic prevalence of ankle arthritis of 77% and a clinical prevalence
of 63% after triple arthrodesis. Pell and colleagues16 examined 132 feet after triple
arthrodesis and noted radiographic progression of ankle osteoarthritis in 60% of their
study population. Other studies have demonstrated significant complications including infection, nonunion, malunion, and hardware related problems. Although triple
arthrodesis may be the only appropriate surgical procedure in some patients with rigid
flatfoot, more limited fusions, including isolated subtalar arthrodesis and double
arthrodesis of the subtalar and talonavicular joints, may be reasonable alternatives,
particularly when all associated deformities are corrected with soft-tissue reconstruction and osteotomy.
Patients with rigid flatfoot deformities who have a rigid hindfoot valgus deformity
yet retain flexibility at the transverse tarsal joint are well-suited for isolated subtalar
arthrodesis. Some examples include posttraumatic deformity owing to fracture of the
os calcis and rigid pes planovalgus associated with tarsal coalition of the subtalar
joint. Mann and colleagues26 described 5 patients in a larger series of 48 feet who
underwent isolated subtalar arthrodesis for posterior tibial tendon dysfunction. The
authors noted that subtalar arthrodesis is indicated in the older, less-active patient
when the primary deformity is hindfoot valgus and/or subtalar inversion is absent. If
forefoot varus of greater than 10 to 15 or transverse tarsal joint hyper-mobility is
present, a subtalar fusion is contraindicated, and a double or triple arthrodesis is
indicated.26 It has been our experience that if the transverse tarsal joint is nonarthritic, it may be possible to correct the forefoot abduction by spring ligament repair
and plantar flexion osteotomy of the medial cuneiform. The addition of a flexor
digitorum longus transfer to subtalar arthrodesis has been advocated not only for
treatment of the flexible flatfoot, but may improve both deformity and function in
patients requiring subtalar arthrodesis for rigid deformity.26 Spring ligament reconstruction also can correct sagittal and axial abduction in these patients if arthritis is
absent. Although this patient population may be limited, those with acquired pes
planus who retain a supple midfoot in the setting of isolated subtalar arthritis may
benefit from isolated subtalar versus triple arthrodesis.
For the patient with arthritis of the subtalar and talonavicular joints, a more
extensive arthrodesis is indicated. Traditionally, these patients have undergone a
triple arthrodesis, which includes the calcaneocuboid joint into the fusion mass
regardless of the presence or absence of calcaneocuboid arthritis. Surgeons have
argued that fusion of the calcaneocuboid joint is necessary to achieve adequate

345

346

Taylor & Sammarco

Fig. 3. (A, B) Preoperative radiographs showing rigid flatfoot deformity in a 48-year-old


woman with disabling pain. Intraoperatively, the subtalar and talonavicular joints were
noted to be arthritic, but the calcaneocuboid joint had normal articular cartilage. (CE)
Postoperative radiographs demonstrating excellent correction with double arthrodesis and
preservation of the calcaneocuboid joint.

deformity correction and to improve the likelihood of successful arthrodesis of the


talonavicular joint. The absence of calcaneocuboid joint motion after subtalar and
talonavicular joint fusion has been supported by biomechanical studies.27 Still, these
studies do not account for smaller motion, which can help to dissipate force across
the lateral column of the foot. It has been these authors experience that, despite
successful realignment with triple arthrodesis, some patients continue to experience
pain and discomfort along the lateral aspect of the foot. Preserving the nonarthritic
calcaneocuboid joint diminishes this problem. Although reported nonunion rates after

Minimizing Fusion in the Rigid Flatfoot

Fig. 4. (AC) A 62-year-old woman with symptomatic arthritis of the subtalar joint after calcaneal
fracture. Rigid pes planovalgus was present despite the absence of arthritis in the TN and CC joints.
(D, E) Subtalar arthrodesis was combined with lateral column lengthening and FDL transfer to spare
the transverse tarsal joint. Iliac crest allograft was secured with a locking plate. (F, G) Two-year
postoperative radiographs showing correction of deformity and successful arthrodesis of the subtalar
joint without progression of arthritis at the transverse tarsal joint.

347

348

Taylor & Sammarco

modern triple arthrodesis are low, nonunion not uncommonly involves the calcaneocuboid joint16,28; hence, excluding the calcaneocuboid joint from the fusion eliminates
the risk of symptomatic calcaneocuboid nonunion. Furthermore, a relative lateral
column lengthening is created via talonavicular arthrodesis and calcaneocuboid
preservation, directly improving midfoot abduction. Sammarco and colleagues2
described not only good deformity correction, but also improved function and high
rates of patient satisfaction with isolated subtalar and talonavicular arthrodesis in 15
patients with rigid flatfoot deformity (Fig. 3).
Although hindfoot fusion is often required when rigid flatfoot deformity is complicated by arthritis, the authors routinely employ osteotomy and soft-tissue reconstruction to correct ankle deformity, hindfoot valgus, midfoot abduction, and forefoot varus
while confining fusion only to the subtalar and/or the talonavicular joint if necessary.
Judicious use of hindfoot fusion combined with nonfusion reconstructive options
offers the patient not only deformity correction, but optimizes function while reducing
the risk of subsequent arthritis in the ankle and midfoot joints (Fig. 4).
SUMMARY

The goals of surgery for the rigid flatfoot are to achieve a painless, stable, functional
plantigrade foot. Although triple arthrodesis affords predictable correction and pain
relief, the long-term sequelae of extended hindfoot fusions include arthritis and often
the need for further, more extensive fusion procedures. We propose that satisfactory
results can be achieved in the rigid flatfoot by limiting fusion to joints that are arthritic,
and correcting associated deformity with osteotomy and soft tissue reconstruction.
REFERENCES

1. Harris RI, Beath T. Hypermobile flat-foot with short tendo Achilles. J Bone Joint Surg
Am 1948;30:116 41.
2. Sammarco VJ, Magur EG, Sammarco GJ, et al. Arthrodesis of the subtalar and
talonavicular joints for correction of symptomatic hindfoot malalignment. Foot Ankle
Int 2006;27:661 6.
3. Kadakia AR, Haddad SL. Hindfoot arthrodesis for the adult acquired flatfoot. Foot
Ankle Clin 2003;8:569 94.
4. Silver CM, Simon SD. Gastrocnemius-muscle recession (Silfverskiold operation) for
spastic equinus deformity in cerebral palsy. J Bone Joint Surg Am 1959;41:1021 8.
5. Nishimoto GS, Attinger CE, Cooper PS. Lengthening the Achilles tendon for the
treatment of diabetic plantar forefoot ulceration. Surg Clin North Am 2003;83:70726.
6. Sammarco GJ, Bagwe MR, Sammarco VJ, et al. The effects of unilateral gastrocsoleus recession. Foot Ankle Int 2006;27:508 11.
7. Myerson MS. Adult acquired flatfoot deformity: treatment of dysfunction of the
posterior tibial tendon. Instr Course Lect 1997;46: 393 405.
8. Jeng CL, Bluman EM, Myerson MS. Minimally invasive deltoid ligament reconstruction
for stage IV flatfoot deformity. Foot Ankle Int 2011;32:2130.
9. Deland JT, de Asla RJ, Segal A. Reconstruction of the chronically failed deltoid
ligament: a new technique. Foot Ankle Int 2004;25:7959.
10. Haddad SL, Dedhia S, Ren Y, et al. Deltoid ligament reconstruction: a novel technique
with biomechanical analysis. Foot Ankle Int 2010;31:639 51.
11. Dwyer FC. Osteotomy of the calcaneum for pes cavus. J Bone Joint Surg Br
1959;41-B:80 6.
12. Koutsogiannis E. Treatment of mobile flatfoot by displacement osteotomy of the
calcaneus. J Bone Joint Surg Br 1971;53:96 100.

Minimizing Fusion in the Rigid Flatfoot

13. Fitzgibbons TC. Valgus tilting of the ankle joint after subtalar (hindfoot) fusion:
complication or natural progression of valgus hindfoot deformity? Orthopedics 1996;
19:41523.
14. Resnick RB, Jahss MH, Choueka J, et al. Deltoid ligament forces after tibialis posterior
tendon rupture: effects of triple arthrodesis and calcaneal displacement osteotomies.
Foot Ankle Int 1995;16:14 20 [Erratum in: Foot Ankle Int 1995;16:314].
15. Wetmore RS, Drennan JC. Long-term results of triple arthrodesis in Charcot-MarieTooth disease. J Bone Joint Surg Am 1989;71:41722.
16. Pell RF 4th, Myerson MS, Schon LC. Clinical outcome after primary triple arthrodesis.
J Bone Joint Surg Am 2000;82:4757.
17. Evans D. Calcaneo-valgus deformity. J Bone Joint Surg Br 1975;57:270 8.
18. Sangeorzan BJ, Mosca V, Hansen ST Jr. Effect of calcaneal lengthening on relationships among the hindfoot, midfoot, and forefoot. Foot Ankle 1993;14:136 41.
19. Haddad SL, Myerson MS, Younger A, et al. SYMPOSIUM: Adult acquired flatfoot
deformity. Foot Ankle Int 2011;32:95111.
20. Cooper PS, Nowak MD, Shaer J. Calcaneocuboid joint pressures with lateral column
lengthening (Evans) procedure. Foot Ankle Int 1997;18:199 205.
21. Deland JT, Otis JC, Lee KT, et al. Lateral column lengthening with calcaneocuboid
fusion: range of motion in the triple joint complex. Foot Ankle Int 1995;16:729 33.
22. Toolan BC, Sangeorzan BJ, Hansen ST Jr. Complex reconstruction for the treatment
of dorsolateral peritalar subluxation of the foot. Early results after distraction arthrodesis of the calcaneocuboid joint in conjunction with stabilization of, and transfer of
the flexor digitorum longus tendon to, the midfoot to treat acquired pes planovalgus in
adults. J Bone Joint Surg Am 1999;81:1545 60.
23. Cotton F. Foot statistics and surgery. Trans N Engl Surg Soc 1935;18:181 8.
24. Hirose CB, Johnson JE. Plantarflexion opening wedge medial cuneiform osteotomy
for correction of fixed forefoot varus associated with flatfoot deformity. Foot Ankle Int
2004;25:568 74.
25. Benthien RA, Parks BG, Guyton GP, et al. Lateral column calcaneal lengthening,
flexor digitorum longus transfer, and opening wedge medial cuneiform osteotomy for
flexible flatfoot: a biomechanical study. Foot Ankle Int 2007;28:70 7.
26. Mann RA, Beaman DN, Horton GA. Isolated subtalar arthrodesis. Foot Ankle Int
1998;19:5119.
27. Walker N, Stukenborg C, Savory KM, et al. Hindfoot motion after isolated and
combined arthrodesis: measurements in anatomic specimens. Foot Ankle Int 2000;
21:9217.
28. Graves SC, Mann RA, Graves KO. Triple arthrodesis in older adults. Results after
long-term follow-up. J Bone Joint Surg Am 1993;75:355 62.

349

Foot Ankle Clin N Am


8 (2003) 91 104

Stage II f latfoot: what fails and why


Keith Hill, MDa, William E. Saar, DOb,
Thomas H. Lee, MDa, Gregory C. Berlet, MDa,c,*
a

Orthopedic Foot and Ankle Center, 6200 Cleveland Avenue, Columbus, OH 43231, USA
b
Orthopedic Associates of Northern Ohio, Inc., 4100 Warrensville Center Road,
Cleveland, OH 44122, USA
c
Department of Orthopaedic Surgery, The Ohio State University College of Medicine,
N1050 Doan Hall, 410 West 10th Ave., Columbus, OH 43210, USA

Dysfunction of the posterior tibial tendon (PTT) that results in acquired


flatfoot deformity continues to be a common clinical entity that seems to be
recognized more frequently [1 4]. Treatment options that are available for PTT
insufficiency vary considerably, depending upon the severity and stage of the
dysfunction and on each patients situation [1,4 7]. Although conservative
nonoperative treatment is often the initial management of choice, surgical
intervention sometimes becomes necessary. Complications can arise from either
the nonoperative or operative treatment of PTT dysfunction and must be
considered, recognized, and dealt with effectively.

Grading
The classification of PTT dysfunction was initially described in 1989 by
Johnson and Strom [1]. A three-stage scheme was assigned to grade the varying
disease patterns that are seen with this entity. Myerson [8] later added a fourth
stage to include the prolonged effects of the disease on the ankle joint.
Patients with Stage I disease typically present without deformity. Dynamic
examination shows normal alignment of the foot and ankle. Mild weak-

* Corresponding author. Orthopedic Foot and Ankle Center, 6200 Cleveland Avenue,
Columbus, OH 43231.
E-mail address: gberlet@aol.com (G.C. Berlet).
1083-7515/03/$ see front matter D 2003, Elsevier Science (USA). All rights reserved.
doi:10.1016/S1083-7515(03)00008-1

92

K. Hill et al / Foot Ankle Clin N Am 8 (2003) 91104

ness can be noted with tenderness along the course of the posterior tibial
tendon [5].
Stage II is characterized by disease progression and the development of a
flexible hindfoot valgus deformity. Physical examination of these patients reveals
the too many toes sign and an inability to perform a single heel rise [1].
Contracture of the gastrocnemius-soleus complex can also be present. A fixed
hindfoot valgus deformity with an accompanying fixed compensatory forefoot
varus characterizes Stage III disease [5]. With chronic Stage IV disease, a fixed
hindfoot deformity leads to eccentric loading and development of ankle arthrosis
due to ankle valgus deformity.

Treatment options and complications


Nonoperative treatment
Regardless of the disease stage at the time of presentation, a period of
conservative treatment is warranted [5]. Despite the dynamic deformity that is
present in patients with Stage II disease, nonoperative treatment, including the
use of bracing or orthotics and nonsteroidal anti-inflammatory medications, may
be beneficial.
The use of appropriate NSAIDs has been well-documented for the treatment of
conditions with an inflammatory component, including PTT dysfunction. With
the advent of Cyclooxygenase-2 (COX-II) selective anti-inflammatory medications, the side effect profile has improved; however, patients should still be
followed closely to avoid any liver, renal, or gastrointestinal complications.
Immobilization in an effort to rest an acutely swollen or chronically torn PTT
can assist in symptom resolution. A below-knee, rigid, walking cast that is worn
for approximately 6 to 8 weeks can substantially reduce pain levels and possible
further degeneration of the tendon [9], as can removable boot walkers or similar
devices. Conservative immobilization of an extremity can have deleterious effects,
including muscular atrophy, skin maceration and breakdown, and possible
increased incidence of deep venous thrombosis.
Patients who improve with a trial of immobilization may eventually
become long-term users of custom-made orthotics or orthoses. A custommolded, semi-rigid foot orthosis [9] with medial posting of the hindfoot and
forefoot may benefit patients with Stage II disease. Other orthoses that are
typically used to treat Stage II PTT dysfunction include the University of
California-Berkeley Laboratories (UCBL) orthosis or a rigid or articulated
ankle-foot orthosis [8,9]. Poorly-fitting orthoses can cause skin breakdown
with subsequent infection over bony prominences. In addition, bracing only
addresses the symptoms and attempts to control progression of deformity.
Braces do not improve the underlying deforming forces. Younger, active
patients may not tolerate bracing and may be considered candidates for surgical intervention.

K. Hill et al / Foot Ankle Clin N Am 8 (2003) 91104

93

Surgical treatment of Stage II disease


General risks of surgical intervention
As with any surgical procedure, there are certain inherent risks when surgically
correcting adult-acquired flatfoot deformity secondary to PTT insufficiency.
Particular attention should be directed toward avoiding excessive retraction
or inadvertent injury to neurovascular structures. Although the posterior tibial
artery, vein, and the tibial nerve are usually deep to the PTT, iatrogenic injury to
any part of the neurovascular bundle can cause long-term, serious consequences
for the patient.
Wound complications, including skin necrosis and infection, can occur with
the surgical incision. Predisposing factors for development of these complications
include patient tissue quality, lack of subcutaneous tissue in or around the medial
approach to the ankle, and postoperative swelling [10]. Meticulous handling of
skin edges, the creation of full-thickness flaps (if possible), and judicious use of
elevation and icing may assist in preventing these complications.
Tenosynovectomy
The current indication for tenosynovectomy is primarily in patients with early
Stage I disease. Complications of tenosynovectomy are relatively rare and primarily are related to the approach to this area. Recurrent synovitis with progressive
tendon dysfunction can develop if patient selection is improper (ie, if synovectomy
is performed for Stage II or III disease) [10]. Tendon debridement and tenosynovectomy alone for Stage II disease will often result in recurrent episodes of
inflammation and progressive degeneration of the tendon [11]. Tenosynovectomy
is not recommended for Stage II disease.
Flexor digitorum longus transfer
Flexor digitorum longus (FDL) transfer has been performed for the past 20 years
and has become a mainstay for soft-tissue procedures in patients with Stage II PTT
dysfunction [6,7,10,12,13]. It has been used as an isolated procedure and in
combination with bony procedures, as well. This is a technically demanding
procedure with pitfalls that can lead to failure. In general, whenever possible,
tendon reconstruction with or without osteotomy is preferable to arthrodesis [14].
Flexor digitorum longus transfer is performed by releasing the tendon distally near
the knot of Henry and rerouting it through the navicular or medial cuneiform to help
substitute for the diseased PTT.
Creation of a bony tunnel for transfer of the FDL tendon can have complications. Particular attention should be directed to the talonavicular and naviculocuneiform joint surfaces to avoid intra-articular penetration by the drill. Fluoroscopy
or intraoperative visualization of joint surfaces can help prevent iatrogenic articular damage. The bony bridge through the navicular can break if the tunnel is
drilled too far medially. This can be salvaged by transferring the FDL to the medial

94

K. Hill et al / Foot Ankle Clin N Am 8 (2003) 91104

cuneiform, if tendon length is adequate [10]. The navicular and cuneiform bones
have both been recommended as transfer sites for the FDL tendon [15]. Depending
upon the length of the harvested tendon, creation of a new bony tunnel in either
bone is possible if the bony bridge is fractured.
The necessity of tenodesing the FDL stump to the flexor hallucis longus (FHL)
tendon at the harvest site has been debated. Wapner et al [16] reported that the
multiple interconnections between the FHL and FDL eliminate the need for
tenodesis after harvesting the FDL. Mann [13] stated that connections between
the FDL and FHL are highly variable and suggested that suturing of the FDL to the
FHL tendon is warranted [13]. If tenodesis is performed, Dalton et al [10] stressed
the importance of performing the tenodesis with the ankle in neutral position.
Regardless, hammertoe deformities may develop if the ankle is not maintained in
neutral position when the FDL is sutured to the FHL in the distal forefoot.
Inadequate repair at the tenodesis site between the FDL and FHL can lead to
rupture with subsequent instability of the first ray; this will necessitate further
surgical intervention [10].
When performed alone, FDL transfers fail to restore normal biomechanics of
the foot. Significant correction of the planovalgus deformity should not be
anticipated [5 7]. In fact, some patients have a progressive planovalgus deformity despite initially successful FDL transfer [10,12,17].
Calcaneal osteotomy with flexor digitorum longus transfer
Because deformities have persisted despite successful FDL transfer, other
concurrent procedures have been recommended. Transfer of the FDL tendon to
treat PTT dysfunction was popularized by Mann and Johnson in the mid 1980s.
In follow-up studies, the deformities that were present in patients with Stage II
disease either remained uncorrected or recurred [18]. Myerson and Corrigan [17]
reported on the lack of correction and progressive planovalgus deformity that
occurred with a FDL transfer alone [17]. Therefore, FDL is not recommended as
a solitary procedure.
To improve the longevity of the transfer and avoid eventual failure of the
procedure, a medial displacement osteotomy of the calcaneus was advocated to
support the tendon transfer by off-loading the stress that is placed on the softtissue procedure. Initially, it was thought that medial displacement osteotomy
placed increased tension in the plantar fascia helping to preserve longitudinal
column height. Horton et al [19] determined that lateral column lengthening and
medial displacing calcaneal osteotomies actually reduced the plantar fascia
tension [19]. With chronic PTT disease, progressive formation of a valgus heel
displaces the Achilles tendon insertion lateral to the subtalar axis and adds to the
valgus-deforming forces. Following medial displacement of the calcaneus, the
vector of pull of the Achilles tendon is medial and assists in foot inversion, which
eliminates the deforming lateral force.
Calcaneal osteotomy has the risk of additional complications. The lateral
incision places the sural nerve and peroneal tendons at risk. To help avoid these

K. Hill et al / Foot Ankle Clin N Am 8 (2003) 91104

95

structures, the incision should be placed posterior to the peroneal tendons and the
sural nerve. Careful dissection and retraction superiorly of these structures can
help prevent iatrogenic injury. Closure of the incision places the sural nerve at
risk again and caution should be used [10].
Medial neurovascular structures can also be at risk at two critical steps during
the procedure. Completion of the osteotomy through the medial cortex must be
performed attentively with an osteotome to avoid injury to the posterior tibial
artery and veins and the lateral and medial plantar nerves. In a cadaveric study,
Greene et al [20] reported that an average of four, and a minimum of two,
neurovascular structures crossed the medial side of the osteotomy site. Branches
of the posterior tibial artery and lateral plantar nerve are at the greatest risk of
injury [20]. Another concern with large translocations of the posterior calcaneus
involves the possibility of troughing. Troughing is a condition that occurs when
cortex to cortex contact is lost after medial displacement of the osteotomy is
performed allowing only cortical to cancellous contact. Subsequent compression
across the osteotomy forces the cortical surface to penetrate into the cancellous
bone causing the osteotomy to collapse on itself. Although this complication has
not been reported in the literature, it is theoretically possible because of the soft
cancellous nature of the calcaneus [21]. Osteomyelitis is also a potential
complication of this procedure.

Fig. 1. Intraoperative view of medial displacement calcaneal osteotomy. Note lateral start position of
the guidewire relative to the tuberosity. Placing the guidewire laterally allows for proper position after
medial shift is accomplished and injury to medial neurovascular structures is avoided. Fixation should
be placed perpendicular to the osteotomy.

96

K. Hill et al / Foot Ankle Clin N Am 8 (2003) 91104

Fig. 2. The Bob osteotomy, which places the cut perpendicular to the posterior facet, allows the
osteotomy to translate proximally, and violates the subtalar joint.

Cannulated, partially-threaded screws should be used for fixation of the


osteotomy. Fluoroscopy should be used for placement of the guidewire to ensure
proper placement perpendicular to the osteotomy site (Fig. 1). The danger is to
place the guidewire too far medially which puts the medial neurovascular
structures at risk. This hazard can be reduced by placing the guidewire for the
screw into the lateral aspect of the osteotomy site under direct visualization
before performing the manual shift. This simplifies advancement of the guidewire

Fig. 3. Nonunion of calcaneal osteotomy despite proper placement and fixation.

K. Hill et al / Foot Ankle Clin N Am 8 (2003) 91104

97

across the osteotomy and into the distal calcaneus. Malunion and nonunion of the
osteotomy can occur [10].
The osteotomy should be placed at a 45 angle to the weight bearing surface of
the foot, parallel to the posterior facet of the subtalar joint (Fig. 2). Verticallyoriented osteotomies are more prone to complications of malunion and nonunion
(Fig. 3), because there is less bony contact for bone healing and increased
instability at the osteotomy site that is amplified by the proximal pull of the
Achilles tendon. Proximal migration of the posterior fragment can compromise
function by decreasing tension on the Achilles tendon and decreasing plantar
flexion power. Calf atrophy can also occur.
Middle-term results for the combined FDL transfer with a medial displacement
calcaneal osteotomy have a failure rate that is as high as 7% [18]. Early failures
have resulted from the pullout of the tendon from the navicular. Guyton et al [18]
reported reattaching two tendons after pullout, but function afterward was poor.
One patient required subtalar fusion as a salvage. One late failure occurred because
of rupture of the tendon transfer 3 months postoperatively. The patient with late
failure developed progressive collapse of the arch and loss of function. Therefore,
care must be taken to secure the tendon transfer to avoid pullout. Using multiple,
interrupted sutures to attach the tendon back onto itself provides solid fixation.
Placing a suture through bone at the entrance of the bony tunnel will add support.
Pullout of the tendon results in poor outcome and need for repeat procedures.
Patients who underwent the combined FDL transfer and medial displacement
osteotomy reported an average of a 10-month postoperative recovery period to
reach maximal clinical improvement. Patients should be counseled to expect
prolonged recovery times. In addition, although radiographic parameters improve
[17,18,21], patients may not see any noticeable change to the shape of the foot.
Approximately 50% of patients reported no improvement in the overall shape of
their foot and only 4% noted any major change. Therefore, patients should
be counseled preoperatively not to expect visible improvement in the arch of
their foot.
Lateral column lengthening
Calcaneal opening wedge osteotomy
Evans [22] reported that the columns must be equal in length to maintain the
balance between the medial and lateral sides of the foot. As evidenced by the
lateral position of the navicular on the talar head, the planovalgus foot, like that
seen in patients with Stage II PTT dysfunction, has a relatively shorter lateral
column that contributes to the deformity. Evans [22] described improving the
alignment of the symptomatic flatfoot by inserting a wedge of tibial bone into an
anterior calcaneal opening wedge osteotomy. Long-term results showed that most
of the correction that was gained with this procedure was maintained over time
[23]. The osteotomy was modified by directing the cut between the anterior and
middle facets of the calcaneus and using a trapezoid-shaped tricortical iliac crest
wedge instead of tibial bone [24]. This procedure has been combined with an

98

K. Hill et al / Foot Ankle Clin N Am 8 (2003) 91104

FDL transfer to treat PTT dysfunction. The addition of the modified Evans
procedure to the FDL transfer attempts to improve the longitudinal arch and
minimize the chances of planovalgus progression of the foot postoperatively
[10,24]. The addition of bony realignment also relieves the stress that is placed on
an isolated tendon transfer.
Some specific physical examination findings may preclude the use of lateral
column lengthening to correct the painful flatfoot. Thomas et al [25] reported on
failed calcaneal osteotomies that required revision to double or triple arthrodesis.
Two of these patients were noted preoperatively to have weakness of the PTT,
FDL, and FHL muscles. The patients also reported severe instability in the ankle.
Two of these patients reported persistent postoperative pain. The other patient
was initially pain-free but sustained frequent eversion injuries that subsequently
progressed to renewed collapse and pain. Thomas et al [25] concluded that
calcaneal osteotomy is likely to fail in patients with substantial muscular
imbalance in the hindfoot. To avoid a disappointing postoperative result, primary
triple arthrodesis may be more beneficial in this group.
Calcaneal osteotomy to lengthen the lateral column has potential problems that
can lead to failure. The range of motion in ankle extension is only moderately
reduced, but loss of subtalar motion can occur despite success of the surgery [22].
A 30% loss of subtalar motion has been reported [26]. In addition, degenerative
changes in the calcaneocuboid joint have been reported in long-term follow-up
[23], probably because of increased pressure in the calcaneocuboid joint, which
has occurred after Evans opening wedge calcaneal osteotomy [27]. The Evans
procedure was shown to increase the contact pressures across the posterior and
anteromedial facets [28]. These areas of increased pressure could potentially lead
to long-term pain and disability from the development of accelerated arthrosis.

Fig. 4. Calcaneal specimen demonstrating the confluence of anterior and middle facets found in two
thirds of the population. AF, anterior facet; MF, middle facet; PF, posterior facet;. The arrow denotes
the confluence of AF and MF.

K. Hill et al / Foot Ankle Clin N Am 8 (2003) 91104

99

Fig. 5. (A)Preoperative and (B) postoperative lateral radiographs of Evans lateral column lengthening
and combined medial displacement osteotomy. Tricortical graft in the anterior process is secured with
an H-plate. Note that the longitudinal arch is re-established.

In addition, the osteotomy for the Evans procedure may violate the anterior or
middle calcaneal facet. There has been no clear consensus on the starting position
for the osteotomy [22,25]. Originally, it was reported to begin 15 mm proximal to
the calcaneocuboid joint, but other investigators advocated starting from 4 mm to
12 mm proximal to the calcaneocuboid joint. Directing the cut between the
anterior and middle facets has been recommended; however, these facets are
distinctly separate in only one third of all patients (Fig. 4) [29,30]. Presumably
early arthrosis as the result of violating the anterior or middle facet of the subtalar
joint could lead to failure of the procedure.
After the osteotomy has been performed, bone graft must be inserted to
support the lateral column lengthening. Nonunion of the graft can occur in up to
20% of cases, which often leads to loss of correction, chronic lateral column pain,
and occasionally, to hardware failure [25,31]. Careful preparation of the graft and
the calcaneal surfaces is essential to provide good exposure of cancellous bone
with tight apposition of the graft. H-shaped cervical plates provide more rigid
stabilization of the graft than pins and staples and should be used when possible
(Fig. 5). Despite this, dorsal displacement of the graft or subluxation of the
anterior process of the calcaneus can occur [10,24,25].
Graft resorption can also occur, which leads to loss of correction, hardware
failure, and chronic lateral column pain. Postoperative, extended, protected
weight bearing is helpful in minimizing these potential complications. Morbidity
of the bone graft site is always a potential complication [10].
Calcaneocuboid distraction arthrodesis
For various reasons, including increased pressure across the calcaneocuboid
joint, some investigators are now recommending lateral column lengthening by
calcaneocuboid distraction arthrodesis (CCDA) [25,28]. This procedure fuses the
calcaneus and the cuboid by excising the articular cartilage from the joint and

100

K. Hill et al / Foot Ankle Clin N Am 8 (2003) 91104

inserting a bone graft into the prepared joint. This lengthens the lateral column
and creates a solid fusion between the calcaneus and the cuboid.
Failures of this procedure are often caused by failure of fixation. The cervical
H-plate has been shown to be substantially stronger than crossed screws for graft
fixation [32], which may explain the lower incidence of nonunion with the use of
this plate. In one study [25], fixation of the graft was accomplished initially with
a cervical plate and 2.7 mm screws. Early screw failure prompted a switch to
3.5 mm screws. The larger screws proved more durable and less likely to fail.
Thomas et al [25] compared their results with using Evans opening wedge
calcaneal osteotomy with CCDA. Postoperative, residual forefoot supination was
a common complaint from patients who underwent CCDA. Lateral foot overloading was also implicated in fifth metatarsal stress fractures, which occurred
months after lengthening. They suggested that these complications could be
overcome by performing the lateral column lengthening procedure and the softtissue procedure with the foot maintained in neutral position.
When assessing the short-term results of both methods of lateral column
lengthening, some loss of correction has been noted. Loss of the talonavicular
coverage angle has been reported as 2.6 with calcaneocuboid fusion and up 5.9
with calcaneal osteotomy. Loss of correction in long-term follow-up led some
investigators to suggest that overcorrection at the time of surgery is actually
desirable; some settling will occur with progressive weight bearing. A cadaveric
study, however, suggested that with CCDA, calcaneocuboid alignment is already
overcorrected in adduction, plantar flexion, and eversion [4].
Lateral column lengthening may lead to loss of subtalar motion. Restriction of
subtalar motion has been reported in 23% of patients and has been noted to be
primarily loss of eversion. Sagittal plane motion of the medial column can be
maintained, but stiffness occurs in subtalar motion [14]. Substantial loss of
subtalar motion can also occur because of fracture of the graft followed by bone
healing. Iatrogenic coalition has been reported, characterized by increasing bone
mass present on radiographs [25].
Wound complications can also contribute to failure of the procedure. Some
degree of dehiscence of the lateral incision has been noted in up to 20% of cases.
Lengthening of the lateral column increases tension on the operative incision that
can lead to dehiscence. Adequate exposure and gentle soft-tissue handling are
especially important to reduce the incidence of this complication. The medial
column may need to be shortened to relieve the tension that is placed on the
lateral skin. The lateral incision may injure the peroneal or sural nerve and cause
a painful neuroma [22,31].
Triple arthrodesis
Triple arthrodesis is a valuable surgical technique to provide stability and
normal alignment to the foot in patients with Stage II PTT dysfunction. The
procedure involves exposing the talonavicular, calcaneocuboid, and talocalcaneal
joints, removing the articular surfaces, and realigning the hindfoot. Triple

K. Hill et al / Foot Ankle Clin N Am 8 (2003) 91104

101

arthrodesis attempts to convert the hindfoot into one osseous unit which obviates
the need for medial and lateral extrinsic muscular stability at the ankle [33].
Triple arthrodesis has been regarded as the best treatment option for patients with
extensive hindfoot deformity [34 36]. The procedure has evolved from using a
large, bony resection to a bone-sparing procedure.
Performing a triple arthrodesis can be a technically challenging procedure that
can have many reasons for failure [36 42]. The main reasons for failed arthrodeses
are persistent deformity, nonunion, malunion, and lateral ankle instability
[33,34,39,42]. Other noted complications include skin necrosis, avascular necrosis
of the talus, sural nerve entrapment, A-V fistula of the posterior tibial artery;
chronic edema and phlebitis, calf muscle atrophy, limb shortening, arthritic
changes in other joints, and difficulty ambulating on uneven surfaces.
Understanding and obtaining the proper position of the arthrodesis is critical in
avoiding a failed triple arthrodesis. The ideal position of the hindfoot following
triple arthrodesis has been suggested to be between 3 to 10 of valgus [36,37,42].
Varus malalignment is unforgiving and leads to lateral column overload, metatarsal and supramalleolar pain, and possible stress fractures [34,37,42]. If there is
any question of alignment, obtaining intraoperative fluoroscopic or radiographic
images is necessary. Maenpaa et al [35] reiterated that to reduce the hindfoot into
the optimal 5 to 10 of valgus, the talar neck may have to be manually lifted
dorsally and laterally, bringing the forefoot medial and plantarward. This can be
accomplished before fixation by placing a lamina spreader between the lateral
shoulder of the talus and the anterior process of the calcaneus. Opening the
spreader will relalign the talar head with the navicular and reduce the posterior
facet to allow screw placement. Adequately correcting heel valgus and pes planus
intraoperatively can avoid varus malunion.
Development of pseudoarthrosis or nonunion can be another major contributor
to failure of a triple arthrodesis. One study reported that 66% of their failures were
malunions and 29% were nonunions [35]. The nonunion rate for triple arthrodesis
has been reported to be between 6% and 36% [33]. Of these, the talonavicular joint
has the highest rate of nonunion [33,35,36,37,40,41], and accounted for more than
90% of the nonunions in one study [41]. The susceptibility of the talonavicular
joint to vertical shear forces and bending levers from the anterior deltoid ligament
and PTT attachment may account for this predisposition. The frequency of mildto-moderate pain in patients with talonavicular nonunion can be as high as 40%,
but most cases are asymptomatic. Talocalcaneal nonunion carries a somewhat
poorer prognosis. The possibility of developing a symptomatic nonunion can be
minimized. Jahss et al [43] suggested an improved method of arthrodesis by fusing
the talus, calcaneus, navicular, and cuboid at their anatomical intersection, the
talocalcaneonavicular articulation. This produces a clinically stable foot, despite a
15% incidence of talonavicular nonunion.
The development of nonunions in triple arthrodesis is strongly related to
premature, postoperative weight bearing and loss of bony apposition because of
inappropriate internal fixation. Wilson et al [40] cited a 14.6% pseudoarthrosis rate
in patients who began weight bearing 2 to 5 weeks postoperatively and only and a

102

K. Hill et al / Foot Ankle Clin N Am 8 (2003) 91104

2.3% nonunion rate in those who were nonweight bearing for at least 9 weeks
postoperatively. Hallgrimsson [44] reported a 47% nonunion rate for joints that
were immobilized for less than 8 weeks and only a 7.7% pseudoarthrosis rate for
joints that were casted for more than 10 weeks. Many symptomatic pseudoarthroses demonstrated poor bony apposition when postoperative radiographs were
reviewed; these may have resulted from a lack of rigid internal fixation.
To minimize failure rates, rigid internal fixation should be accomplished
[36,39]. Fixation of the arthrodeses should provide good bony apposition and
resistance to shear forces that may disrupt a bony union [33,40]. Rigid internal
fixation, using AO/ASIF (Arbeitsgemeinschaft fur osteosynthesefragen/Association for the Study of Internal Fixation) techniques with screws or staples,
provides compression across the arthrodesis sites and allows the maximum
opportunity to fuse the joint.
Reported failures have also been attributed to osteonecrosis of the talar dome.
One case of talar osteonecrosis was caused by a suspected injury to the talar blood
supply from aggressive debridement [35]. Jones and Nunley [45] reported three
cases in which screws were placed from the anterior talar neck directed posteriorly
across the posterior facet of the calcaneus. They hypothesized that the blood
supply to the talus through the deltoid branches, as well as those through the talar
neck, were compromised during screw placement. The blood supply from the
sinus tarsi is compromised by the surgical approach which contributes to the risk
of osteonecrosis. Failure of the triple arthrodesis because of osteonecrosis of the
talus may be avoided by screw placement that is directed posterior to anterior.
Triple arthrodesis deprives the midtarsal and subtalar joints of their role in the
gait cycle, impaction, and translation of torsional forces in the lower leg. This is
believed to result in a largely asymptomatic arthrosis of the contiguous ankle and
anterior intertarsal joints in many patients, but triple arthrodesis is clearly related
to progression of degenerative changes in the foot and ankle [37,38]. Triple
arthrodesis may result in increased lateral ankle instability, perhaps related to
intraoperative violation of the anterior talofibular ligament [38]. This instability
may contribute to radiographic changes in the ankle joint. The presence of
arthrosis does not always correlate with symptoms. Drew [42] studied 27 triple
arthrodesis patients and found ankle arthritic changes on radiographs; however,
most patients were asymptomatic. Radiographic changes were unrelated to the
position of the foot, age of the patient, or bony union.
Patient satisfaction may lead to a perceived failure despite a technically
successful operation. Vertical pressure analysis of gait following triple arthrodesis
can reveal decreased dorsiflexion-plantarflexion range of pedal motion, difficulty
ambulating on irregular terrain, decreased mean step length and gait velocity, and
rapid forefoot abduction at propulsion. In addition, weight may be unequally
distributed between the feet, with lateral displacement at toe-off. Disturbances in
the normal gait patterns have an impact, in theory, on symptoms and joint stresses
in the extremity after triple arthrodesis and may be an unexpected result to the
patient [33]. Preoperative counseling must ensure that surgeon and patient
expectations of a successful outcome are the same.

K. Hill et al / Foot Ankle Clin N Am 8 (2003) 91104

103

Summary
Stage II PTT dysfunction is a complex problem that has multiple treatment
options. Whether it is treated nonoperatively or operatively, multiple factors
affect the success of treatment. The orthopedic literature reports many factors that
can lead to failure. Although it is virtually impossible to account for every
possible scenario, adhering to sound principles and understanding what can go
wrong and why, can help to avoid many of the pitfalls.

References
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]

[17]
[18]

[19]

[20]
[21]

Johnson KA, Strom DE. Tibialis posterior tendon dysfunction. Clin Orthop 1989;239:196 206.
Johnson KA. Tibialis posterior tendon rupture. Clin Orthop 1983;177:140 7.
Fowler AW. Tibialis posterior syndrome. J Bone Joint Surg 1953;37B:520.
Kitaoka HB, Kura H, Luo ZP, et al. Calcaneocuboid distraction arthrodesis for posterior tibial
tendon dysfunction and flatfoot. A cadaveric study. Clin Orthop 2000;381:241 7.
Beals TC, Pomeroy GC, Manoli A. Posterior tibial tendon insufficiency: diagnosis and treatment.
J Am Acad Orthop Surg 1999;7:112 8.
Funk DA, Cass JR, Johnson KA. Acquired adult flat foot secondary to posterior tibial-tendon
pathology. J Bone Joint Surg 1986;68A:95 106.
Mann RA, Thompson FM. Rupture of the posterior tibial tendon causing flat foot: surgical
treatment. J Bone Joint Surg 1985;67:556 61.
Myerson MS. Adult acquired flatfoot deformity. J Bone Joint Surg 1996;78A:780 92.
Sferra JJ, Rosenberg GA. Nonoperative treatment of posterior tibial tendon pathology. Foot
Ankle Clin 1997;2:261 73.
Dalton GP, Wapner KL, Hecht PJ. Complications of Achilles and posterior tibial tendon surgeries. Clin Orthop 2001;391:133 9.
Kitaoka HB, Luo ZP, An KN. Reconstruction operations for acquired flatfoot: biomechanical
evaluation. Foot Ankle Int 1998;19:203 7.
Mann RA, Specht LH. Posterior tibial tendon rupture: analysis of eight cases. Foot Ankle Int
1982;2:350.
Mann RA. Posterior tibial tendon dysfunction: treatment by flexor digitorum longus transfer.
Foot Ankle Clin 2001;6:77 87.
Neufeld SK, Myerson MS. Adult acquired flatfoot deformity: complications of surgical treatments for adult flatfoot deformities. Foot Ankle Clin 2001;6:1.
Pomeroy GC, Manoli II A. A new operative approach for flatfoot secondary to tibialis posterior
tendon insufficiency: a preliminary report. Foot Ankle Int 1997;18:206 12.
Wapner KL, Hecht PJ, Shea JR, et al. Anatomy of second muscular layer of the foot: considerations for tendon selection in transfer for Achilles and posterior tibial tendon reconstruction.
Foot Ankle Int 1994;15:420 3.
Myerson MS, Corrigan J. Treatment of posterior tendon dysfunction with flexor digitorum
longus tendon transfer and calcaneal osteotomy. Orthopedics 1996;19:383 8.
Guyton GP, Jeng C, Krieger LE, et al. Flexor digitorium longus transfer and medial displacement
calcaneal osteotomy for posterior tibial tendon dysfunction: a middle-term clinical follow-up.
Foot Ankle Int 2001;22:627 32.
Horton GA, Myerson MS, Park BG, et al. Effects of calcaneal osteotomy and lateral column lengthening on the plantar fascia: a biomechanical investigation. Foot Ankle Int 1998;
19:370 3.
Greene DL, Thompson MC, Gesink DS, et al. Anatomic study of the medial neurovascular
structures in relation to calcaneal osteotomy. Foot Ankle Int 2001;22:569 71.
Beasley B, Lee TL, Massa E, Berlet G. Medial displacement calcaneal osteotomy and tendon

104

[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]

[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]

K. Hill et al / Foot Ankle Clin N Am 8 (2003) 91104


transfer for treatment of posterior tibial tendon dysfunction: digital radiographic interpretation
of sagittal plane changes. Presented at American Orthopedic Foot and Ankle Association
18th Annual Meeting. Traverse City (MI), July 12 14, 2002.
Evans D. Calcaneo-valgus deformity. J Bone Joint Surg 1975;57B:270 8.
Phillips GE. A review of elongation of os calcis for flat feet. J Bone Joint Surg 1983;65B:15 8.
Mosca VS. Calcaneal lengthening for valgus deformity of the hindfoot. J Bone Joint Surg
1995;77A:500 12.
Thomas RL, Wells BC, Garrison RL, et al. Preliminary results comparing two methods of lateral
column lengthening. Foot Ankle Int 2001;22:107 19.
Deland JT, Otis JC, Lee IT, et al. Lateral column lengthening with calcaneocuboid fusion: range
of motion in the triple joint complex. Foot Ankle Int 1995;16:729 33.
Cooper PS, Nowak MD, Shaer J. Calcaneocuboid joint pressures with lateral column lengthening
(Evans) procedure. Foot Ankle Int 1997;18(4):199 204.
McCormack AP, Niki H, Kiser P, et al. Two reconstructive techniques for flatfoot deformity
comparing contact characteristics of the hindfoot joints. Foot Ankle Int 1998;19:451 61.
Sarrafian SK. Anatomy of the foot and ankle: descriptive, topographic, functional. Philadelphia:
J.B. Lippincott; 1983.
Hyer CF, Lee TH, Bloc AJ, et al. Evaluation of the anterior and middle talocalcaneal articular
facets and the Evans osteotomy. J Foot Ankle Surg 2002;41:389 93.
Toolan BC, Sangeorzan BJ, Hansen S. Complex reconstruction for the treatment of dorsolateral
peritalar subluxation of the foot. J Bone Joint Surg 1999;81A:1545 60.
Kimball HL, Aronow MS, Sullivan RJ, et al. Biomechanical evaluation of the calcaneocuboid
distraction arthrodesis: a cadaver study of two different fixation methods. Foot Ankle Int 2000;
21:845 8.
Johnson MK, Kanat IO. Complications of triple arthrodesis with comparison to select rearfoot
fusions. J Foot Surg 1987;26(5):371 9.
Hutchinson B. Reconstructive surgery of the foot and ankle: current concepts in triple arthrodesis. Clin Podiatr Med Surg 2000;17(1):1 18.
Maenpaa H, Lehto MU, Belt EA. What went wrong in triple arthrodesis? Clin Orthop 2001;
391:218 23.
Wapner KL. Triple arthrodesis in adults. J Am Acad Orthop Surg 1998;6:188 96.
Bednarz PA, Monroe MT, Manoli A. Triple arthrodesis in adults using rigid internal fixation: an
assessment of outcome. Foot Ankle Int 1999;20:356 63.
Pell RF, Myerson MS, Schon LC. Clinical outcome after primary triple arthrodesis. J Bone Joint
Surg 2000;82A:47 57.
Coetzee JC, Hansen ST. Surgical management of severe deformity resulting from posterior tibia
tendon dysfunction. Foot Ankle Int 2001;22:944 9.
Wilson FC, Fay GF, Lamotte P, et al. Triple arthrodesis. a study of the factors affecting fusion
after three hundred procedures. J Bone Joint Surg 1965;47A:340.
Patterson RL, Parrish FF, Hathaway EN. Stabilizing operations on the foot. a study of the
indications, techniques used, and the end results. J Bone Joint Surg 1950;32A:1 26.
Drew AJ. The late results of arthrodesis of the foot. J Bone Joint Surg 1951;33B:496 502.
Jahss MH, Godscik PA, Levin H. Quadruple arthrodesis with iliac bone graft. In: Bateman JE,
Trott AW, editors. The foot and ankle. New York: Thieme C. Decker; 1980. p. 93 102.
Hallgrimsson S. Studies on reconstructive and stabilizing operations on the skeleton of the foot.
Acta Chir Scand 1943;88:78.
Jones CK, Nunley JA. Osteonecrosis of the lateral aspect of the talar dome after triple arthrodesis: a report of three cases. J Bone Joint Surg 1999;81A:1165 9.

Subtalar
A r t h ro e re i s i s i n
Pediatric Flatfoot
Reconstruction
Pablo Fern
andez de Retana,
n Viladot, MDc
Ramo

MD

a,

*, Fern
ando Alvarez,

MD

KEYWORDS
 Arthroereisis  Subtalar implant  Flatfoot  Subtalar joint
 Pes planovalgus  Pediatric flatfoot

Pediatric and juvenile flatfoot is a common problem in childhood, present in one in nine
children.1 The morphologic characteristics of this condition are heel valgus and flattening of the medial longitudinal arch. Other characteristics are usually observed,
such as supination and abduction of the forefoot, tightening of the Achilles tendon,
and hypertonia of the peroneal muscles. Most children with flatfoot will undergo spontaneous correction or become asymptomatic; those that are symptomatic require
treatment.
Conservative treatment includes supportive footwear and physical therapy. Usually
flatfoot is overtreated. The prevalence of flat feet with diagnostic criteria was 2.7% in
a Spanish population of 4- to 13-year-old schoolchildren, with 14.2% of them were
receiving orthopedic treatment.2 Very little is known about the natural development
of flatfoot, and failure to treat this condition could lead to persistent pain in the medial
longitudinal arch, hallux valgus, degenerative arthritis, posterior tibial tendon dysfunction, metatarsalgia, and knee or low back pain.3,4
Surgery is rarely indicated in children with flatfeet. Surgical options are tendon
transfers, tarsal arthrodeses, calcaneal osteotomies, and subtalar joint motion blocking procedures. The final decision is usually based on the clinical and radiographic
assessment of the patient and surgeon preference. Flexible flatfoot should be treated
preferably with extra-articular procedures and avoiding tendon transfers. In 1946,
Chambers5 was the first to introduce the idea of restricting subtalar joint motion

Department of Orthopedic Surgery, Hospital San Rafael, Passeig Vall dHebron 107-117, 08035
Barcelona, Spain
b
Head of Foot and Ankle Unit, Department of Orthopedic Surgery, Hospital San Rafael,
Passeig Vall dHebron 107-117, 08035 Barcelona, Spain
c
Clnica Tres Torres, Barcelona, Spain
* Corresponding author.
E-mail address: pfernan@hsrafael.com
Foot Ankle Clin N Am 15 (2010) 323335
doi:10.1016/j.fcl.2010.01.001
1083-7515/10/$ see front matter 2010 Elsevier Inc. All rights reserved.

foot.theclinics.com

324

 ndez de Retana et al
Ferna

with an autologous bone block filling the sinus tarsi. In 1952, Grice6 observed that
extra-articular subtalar arthrodesis with cortical graft for flatfoot deformity in children
was associated with several problems, such as development of degenerative arthritis
in adjacent joints and inability of the hindfoot to adapt to uneven surfaces. Haraldsson7
and Lelievre8 pointed out it was most important to block the sinus tarsi restricting subtalar motion while avoiding arthrodesis. Since then, the term arthroereisis is used to
describe the limitation of subtalar motion.8
Shortening of the gastrocnemius or the Achilles tendon is very common in flatfoot,
and often these structures need to be lengthened. Osteoperiosteal flap of the posterior
tibial tendon has been advised to increase the strength of this muscle. In the past, this
procedure was recommended in addition to arthroereisis, whereas now arthroereisis
is performed without the osteoperiosteal flap.
SUBTALAR ARTHROEREISIS WITH SINUS TARSI IMPLANT

Arthroereisis has been used predominantly in recent years to treat flatfeet in the pediatric population.9,10 Many arthroereisis procedures have been described to limit subtalar joint motion and improve position. In 1970, Lelievre8 introduced the term lateral
arthroereisis using a temporary staple across the subtalar joint. In 1977, Subotnick11
first described a sinus tarsi implant using a block of silicone elastamer. Smith and
Millar3 used a polyethylene screw in sinus tarsi (STA-peg, Dow Corning Wright
Corporation, Arlington, TN, USA) and reported a 96% success rate. Vedantam and
colleagues12 used the STA-peg in 78 patients, most of whom had neurologic problems, and reported satisfactory results in 96%. Although no implant degradation
was encountered, they did not recommend the implant in more active children.
Viladot4 reported the use of a silicon implant with a cup shape, obtaining 99% good
results in 234 patients. Carranza-Bencano and colleagues13 obtained 90% good and
excellent results in 77 feet with Viladots silicon implant. Using Viladots device, Black
and colleagues14 were unable to obtain consistent improvements radiologically, and
73% of patients reported that their feet were significantly painful.
Currently, silicon implants are not usually used and have been replaced with troncoconical implants, such as Kalix (Newdeal, Saint Priest, France) and MBA (Kinetikos
Medical, Inc, Carlsbad, CA, USA). Zaret and Myerson9 reported 23 children who
had flexible flatfoot treated with MBA implants. They reported postoperative sinus
tarsi pain in 18% of patients that improved with rest and corticosteroid injection.
Two children required implant removal at a mean of 9 months after surgery. Subtalar
joint range of motion increased after implant removal. They also observed an interesting phenomenon: removal of the implant was not associated with a reversal of
the foot structure.
Giannini15 developed an expanding and bioabsorbable implant material. He
reported 94% good results in 50 cases using nonreabsorbable material and 20 cases
with bioabsorbable. He observed no difference in long-term follow-up between nonreabsorbable and bioabsorbable materials. Bioabsorbable implants have the advantage of a low rate of intolerance, which helps avoid a second procedure for removal.
The calcaneo-stop is a technique for limiting motion in the subtalar joint through
the insertion of a screw into the calcaneus. Alvarez16 was the first to describe this
procedure using a cancellous screw (Fig. 1). The entry point was located in the sinus
tarsi. The head of the screw limits talus motion and prevents talus displacement downward. In 475 patients with an average follow-up ranging from 12 to 112 months,
Magnan and colleagues17 reported 83% good results. Nogarin18 described the anterograde calcaneus-stop with a cancellous screw inserted into the talus. Recently, Roth

Pediatric Flatfoot Reconstruction

Fig. 1. Recaredos procedure.

and colleagues19 reported 94 feet operated in 48 children using a temporary cancellous screw placed percutaneously into the talus. At 5 years follow-up, 91.5% had
excellent results.
Vogler20 classified sinus tarsi implants into three types based on their biomechanical properties: self-locking wedge, axis-altering implant, and impact-blocking device.
Most sinus tarsi implants are self-locking wedge. All sinus tarsi devices restrict hindfoot valgus, and the calcaneus is vertically orientated beneath the ankle joint.10 The
talus is dorsiflexed and externally deviated. Talonavicular subluxation and forefoot
abduction are corrected.
INDICATIONS FOR SUBTALAR ARTHROEREISIS IN CHILDREN

Supportive insoles with a medial longitudinal arch are recommended in children who
have painful flatfoot. Surgery is indicated for symptomatic flatfoot, which persists
despite prolonged nonsurgical treatment. Surgical options are lateral column lengthening and sinus tarsi implant. Both operations need a flexible foot to correct malposition. If a rigid deformity of the foot is present (eg, tarsal coalitions or spastic disorders),
additional procedures must be associated.
Calcaneal lengthening is widely accepted in the United States, whereas sinus tarsi
implants have been used for many years in European countries. The advantages of
arthroereisis over osteotomies are that it is easy to perform, less immobilization is
required, and it has no associated risk for nonunion. However, its disadvantages
include a high percentage of implant intolerance and limitation of subtalar motion.
No controlled trials are comparing these techniques and the level of scientific evidence
is low. Practitioners using both techniques agree that a low percentage of children
who have flexible flatfoot require surgery.15,21
Pediatric flatfoot can have different origins, including flexible flatfoot, cerebral palsy,
myelomeningocele, tarsal coalitions, and rheumatic diseases. Not every childs flatfoot needs surgical treatment and, when surgery is necessary, not all cases require
subtalar arthroereisis. The main indications of subtalar arthroereisis in children are discussed in the following sections.
Flexible Flatfoot

Flexible flatfoot is a very common disorder in children. It rarely needs surgical treatment because patients are usually completely asymptomatic and most of these feet
correct spontaneously with growth. However, in a small group of these patients
(<3%), the feet do not correct and are symptomatic. Symptoms are pain (usually on
the medial side of the foot), fatigue, difficulty playing sports or walking on uneven
surfaces, and cramps. Surgery is indicated in these patients if, in addition to

325

326

 ndez de Retana et al
Ferna

symptoms, the child demonstrates most of the following inclusion criteria: severe flatfoot without clinical and radiologic improvement after at least 2 years of conservative
treatment; hindfoot valgus greater than 10 ; shortness of the Achilles tendon; Viladot
footprint type II, III, or IV4; talo-first metatarsal (Mearys) angle smaller that 170 ; Moreau-Costa-Bartani angle greater than 130 , or Kite angle greater than 25 .
The age of the patient is an important factor and is of wide concern.3,15 The authors
recommend surgery between 8 and 12 years. Before 8 years, many children may experience spontaneous correction. The authors also prefer not to operate on patients older
than 12 years of age because one goal of arthroereisis is to relocate the talus properly
over the calcaneus to allow remodeling of these bones and the subtalar joint during the
rest of the growing period, which they believe requires at least 2 years. The sinus tarsi
implant acts as a splint restoring the normal foot architecture. Bone growth will develop
in the right way. After 12 years of age, there is not enough time for remodeling hindfoot
bones and ligaments. Although many children have flexible flatfeet, very few meet the
inclusion criteria, and therefore surgery is unusual for childhood flexible flatfoot.
Flatfoot Associated to Navicular Bone Disorders

It is not unusual for childhood flatfoot to be related to an accessory or a prominent


navicular. In these cases, the tibialis posterior tendon has an abnormal distal insertion,
which may cause a functional insufficiency of this tendon. If the foot is symptomatic,
subtalar arthroereisis must be associated with partial removal of the navicular as
described by Kidner,22 or a suture between both tibialis tendons.
Flatfoot Associated With Tarsal Coalition

Talocalcaneal and calcaneonavicular coalitions can cause flatfoot. Usually this condition is asymptomatic until the age of 10 to 12 years, when the foot becomes painful,
the hindfoot is stiff, and a painful contracture of the peroneal tendons occurs. When
surgery is needed, removal of the coalition is indicated together with subtalar arthroereisis. If more than one third of the subtalar joint is affected, subtalar arthrodesis is
probably a better option.
Flatfoot Secondary to Clubfoot Overcorrection

Flatfeet that are secondary to clubfoot overcorrection are difficult to correct because
they are stiff from previous treatments. However, in some cases the foot is still flexible
and subtalar arthroereisis may be indicated. Additional surgical procedures are often
necessary to correct forefoot deformities (hallux flexus is frequent in overcorrected
clubfeet).
Spastic Paralytic Flatfoot

Children who have cerebral palsy often develop severe planovalgus feet with Achilles
and peroneal tendons contracture. Arthrodesis or tendon transfers are accepted treatments. When the foot is not very rigid, subtalar arthroereisis can be an alternative to
arthrodesis.23
Subtalar arthroereisis should only be indicated in patients who have severe symptomatic flatfoot that has not improved with orthotic treatment and who fulfill the inclusion criteria for flexible flat foot.
COMPLICATIONS

Complications may be divided into general and implant-specific.10 General complications include persistent sinus tarsi pain, malposition, overcorrection, undercorrection,

Pediatric Flatfoot Reconstruction

wrong implant size, and loss of position.10 The most common complication is intolerance of the implant because of pain. Black and colleagues14 reported 73% painful feet
using Viladots initial design, requiring implant removal in 36% of their patients with
rlittle resolution of the symptoms. This initial design by Viladot is no longer used. Gutie
rez and Herrera24 reported that 7.5% of implant removals involved the Giannini nonreabsorbable implant. Overcorrection is a cause of pain in sinus tarsi and implant
removal has been recommended with good results (Fig. 2).12
Implant-specific complications are related to the properties of each implant, and
include debris from wear, foreign body reaction, implant degradation, and implant
fracture.10 For Roth and colleagues,19 12% of complications were related to the calcaneo-stop technique, and all were associated with the screws (9 screw breakages
and 2 incorrectly positioned screws). Giannini15 reported two complications involving
implant degradation (4.8%) in his series with a bioreabsorbable implant. Small fragments migrated under the skin, causing local irritation.
Arthritis, synovitis, implant fracture, and implant loosening have been reported with
the STA-peg implant.3 Case reports have shown intraosseous cysts within the talus,25
osteonecrosis of the talus,26 peroneal muscle spasms, and small fracture of the
calcaneus.27
rrez and Herrera24 reported temporary postoperative supination in 27% of
Gutie
cases. This condition resolved spontaneously in all feet, but resolution of symptoms
took 1 month in 35%, 3 months in 41%, and 4 months in 24% of cases.
SURGICAL TECHNIQUE

The authors usually perform foot surgery under general anesthesia in children. The
patient is placed in a supine position with a pillow under the ipsilateral buttock to avoid
excessive external rotation of the foot and facilitate approach to the sinus tarsi. A

Fig. 2. Preoperative (A) and postoperative (B) lateral radiographs showing overcorrection
after subtalar arthroereisis.

327

328

 ndez de Retana et al
Ferna

pneumatic tourniquet is applied at the thigh and the surgical field is prepared leaving
the leg exposed.
In most cases, the first step of the operation is the lengthening of the Achilles
tendon. To evaluate if this tendon is retracted, the authors hold the foot with the hindfoot in neutral position and the knee completely extended. Then we passively dorsiflex
the foot. If we cannot achieve 10 of dorsiflexion we consider that the Achilles tendon
is retracted and needs to be lengthened, which will help in correcting valgus of the
calcaneus and implanting the endorthesis. They recommend percutaneous lengthening. Two 5-mm incisions are performed on the lateral border of the tendon (at 2
and 6 cm from its insertion in the calcaneus) and one 5-mm incision on the medial
border (at 4 cm from the insertion). Through each incision, the tendon is divided transversely approximately half of its width. The foot is dorsiflexed so that the tendon is
lengthened until approximately 15 of dorsiflexion is obtained while the knee is
extended.
If an accessory or prominent navicular bone must be removed, the procedure is performed as described by Kidner.22 A 3-cm longitudinal skin incision is performed at the
medial side of the foot, over the navicular tuberosity. The tibialis posterior tendon is
split and partially detached from the navicular bone. The bony prominence or accessory bone is removed and the tendon sutured. Subcutaneous tissue and skin are also
sutured.
A slightly curved 2-cm skin incision is made on the lateral side of the hindfoot
centered over the sinus tarsi, just anterior and plantar to the tip of the lateral malleolus.
Care must be taken not to damage the peroneal tendons, saphenous nerve, and intermediate dorsal cutaneous nerve (a branch of the superficial peroneal nerve). Direct
approach to the sinus tarsi is obtained. The sinus tarsi is debrided to remove its
contents (fatty tissue with abundant nerve endings) (Fig. 3). This step is important to
eliminate the irritative stimuli of the endorthesis in the sinus, which could originate
the contracture of the peroneal tendons. The powerful interosseous talocalcaneal ligament must not be damaged, because it is an important hindfoot stabilizer.
The next step consists of restoration of the foot arch and correction of calcaneus
valgus deviation. This is achieved using a blunt lever introduced through the sinus tarsi
and under the neck of the talus. The correcting procedure is then performed, with the
lever pushed in distal direction so that the hindfoot is supinated at the same time the
assistant pronates the forefoot. The aim of this procedure is to move the head of the
talus upwards, backward, and outwards, so that it is repositioned in its physiologic

Fig. 3. Sinus tarsi is emptied.

Pediatric Flatfoot Reconstruction

position and hindfoot pronation is corrected. One must note that in flatfoot, the talus
has slipped forward, downward, and inward.
To conserve the correction obtained, an endorthesis is inserted in the sinus tarsi.
First, the trial implants (with increasing diameters) are inserted until the appropriate
implant size is determined. The authors choose the smallest implant that corrects
the deformity and remains stable in the sinus tarsi while moving the subtalar joint. It
is advisable to check the correct position of the trial implant with fluoroscopy. Then,
the definitive endorthesis is implanted, which is the same size as the trial implant
(Fig. 4).
Closure of the wound is performed in a routine fashion and a below-knee compression cast is applied. The whole procedure takes 20 to 30 minutes.
If necessary, additional procedures can be associated, such as removal of accessory navicular bone, removal of tarsal coalition, and peroneal tendons lengthening.
Sutures are removed 10 to 12 days postoperatively and a below knee walking cast is
applied. This cast is removed 6 weeks after surgery and the patient is advised to wear
rigid orthopedic insoles for 1 year to support the correction. Weight-bearing is allowed
when the skin has healed.
Endorthesis

Since 1998, the authors have been using the Kalix endorthesis (Newdeal SA, Vienne,
France) for child and adult flatfoot surgical treatment (Fig. 5). This implant consists of
a metal cone trunk introduced into another polyethylene cone trunk that expands. It is
manufactured with the biocompatible materials titanium and high-density polyethylene. Its conical shape fits well into the sinus tarsi, and the lateral fins prevent the
implant from moving out of the sinus tarsi. It can be visualized radiographically
because of its metallic component.

RESULTS OF TREATMENT

In 2006, the authors performed a descriptive retrospective study of children who had
flatfoot treated with subtalar arthroereisis between 1998 and 2005 in their institution,
Hospital San Rafael in Barcelona, Spain. The total number of feet operated was 116 in
68 patients. Etiology of the deformity showed flexible flatfoot in 97 feet; tarsal coalition
in 10 feet; spastic flatfoot in 3 feet; Downs syndrome in 2 feet; clubfoot overcorrection
in 2 feet; and juvenile arthritis in 2 feet. To have a homogeneous sample, the study only
included patients who had flexible flatfoot (56 patients and 97 feet).

Fig. 4. (A) Endorthesis mounted on its handle and ready to be implanted. (B) Endorthesis
implanted in sinus tarsi.

329

330

 ndez de Retana et al
Ferna

Fig. 5. Kalix endorthesis.

The average age at surgery was 10.5 years (range, 714 years), and the mean
follow-up was 4.5 years (range, 1.18.0 years). The patients consisted of 34 boys
(60.1%) and 22 girls (39.9%), and side distribution was 50 right feet (51.5%) and 47
left (48.5%).
All 97 feet underwent subtalar arthroereisis with Kalix endorthesis and the Achilles
tendon was lengthened in 85 feet (87.6%).
The study included clinical and radiologic evaluation, and the American Orthopaedic
Foot & Ankle Society (AOFAS) scale was used. The footprint evolution was also evaluated. The radiological study consisted of weight-bearing anteroposterior and lateral
views of both feet. In the anterior-posterior view, the angle between the main axes of
the talus and calcaneus (Kite angle, which should be between 15 and 25 ) was
measured. In the lateral view, the Moreau-Costa-Bartani angle, formed by the lowest
points of the calcaneus, talonavicular joint, and head of the first metatarsal (normal
values range from 120 to 130 ) was also measured. These angles are greater in
subjects who have flatfeet.
AOFAS average score improved from 72.2 points (range, 5881) to 96.4 points
(range, 87100). Patient satisfaction was as follows: very high in 23 (41%); high in
28 (50%); low in 3 (5.4%); and very low in 2 (3.6%). When considering patient satisfaction, it is helpful to take into account that most patients had bilateral procedures.
Clinical correction of the deformity (Fig. 6) was evaluated using the Viladot footprint
classification.4 Most feet (94.8%) were classified as grade III or IV flatfoot before treatment, whereas 91.8% were normal or grade I or II flatfoot after surgery (Fig. 7). Preoperative and postoperative values are shown in Table 1.

Fig. 6. Preoperative (A) and postoperative (B) aspect of a childs flatfoot treated by subtalar
arthroereisis and Achilles tendon lengthening.

Pediatric Flatfoot Reconstruction

Fig. 7. (A) Preoperative footprint: grade III flatfoot. (B) Postoperative footprint: normalgrade I flatfoot.

Radiologic results were as follows: Mean Moreau-Costa-Bartani angle changed


from 144.9 preoperative (136 161 ) to 127.5 postoperative (119 135 ) (Fig. 8).
Mean Kite angle improved from 32.2 preoperative (26 42 ) to 19.3 postoperative
(12 28 ). By the time the study was performed, the endorthesis had been removed
in 37 feetin 6 because of intolerance to the implant and in 31 it was removed without
complications because the patients had reached the end of the growing period. The
average angle loss in these feet after endorthesis removal was 0.8 in MoreauCosta-Bartani angle and 0.3 in Kite angle (not statistically significant for both values)
(Fig. 9).
Complications observed in this study included undercorrection in eight feet, overcorrection in five, Achilles tendon contracture in four, pain at the sinus tarsi in four,

Table 1
Preoperative and postoperative footprints according to Viladot
Preoperative

Postoperative

Grade 0

0 feet

31 feet

Grade I

0 feet

42 feet

Grade II

5 feet

16 feet

Grade III

39 feet

6 feet

Grade IV

53 feet

2 feet

From Viladot A. Surgical treatment of the childs flatfoot. Clin Orthop 1992;283:348; with
permission.

331

332

 ndez de Retana et al
Ferna

Fig. 8. (A, B) Preoperative lateral radiographs. (C, D) Postoperative lateral radiographs.

and peroneal tendon contracture in two. No cases of infection occurred. All cases with
undercorrection had grade IV footprint before surgery and all improved clinically. The
overcorrected cases were all asymptomatic. Four cases needed Achilles tendon
lengthening (none had it lengthened when subtalar arthroereisis was performed). In

Fig. 9. (A) Preoperative lateral radiograph of an 8-year-old boys flatfoot. (B) Postoperative
lateral radiograph at 14 years of age. (C) Lateral radiograph at 17 years of age, 3 years after
endorthesis removal. Note that there is no loss of correction.

Pediatric Flatfoot Reconstruction

all four feet with pain at sinus tarsi and the two with peroneal tendons contracture,
symptoms disappeared after endorthesis removal.
DISCUSSION

Flexible flatfoot in children is a common condition that does not require surgery in
most cases. However, some patients are symptomatic and, if they fulfill the inclusion
criteria, require surgical treatment. Although numerous treatment options have been
proposed, probably no single treatment is appropriate for every patient. Subtalar
arthroereisis has proved to yield good mid- to long-term results.3,4,11,13,15
The purpose of arthroereisis is to relocate the talus properly over the calcaneus to
allow remodeling of these bones and the subtalar joint during the remainder of the
growing period.9,24 This remodeling period should last at least 2 years. The foot reaches its bone maturity at age 14 to 15 years. Therefore, arthroereisis should best be
performed before the age of 12, so that there are still two years for the remodeling
period.
Achilles tendon lengthening is a surgical procedure that is very common in flatfoot
treatment. Although some surgeons do not consider it necessary,19 most publications
about subtalar arthroereisis for flatfoot include Achilles tendon lengthening.4,9,24,28,29
Gastrocnemius contracture is part of the pathologic anatomy of flatfoot and, in the
authors experience, is present in most cases. These investigators always decide
preoperatively if Achilles lengthening is necessary. If ankle dorsiflexion is less than
10 while the knee is extended, they believe Achilles tendon lengthening is indicated.
Subtalar joint motion is usually diminished after arthroereisis, but the authors believe
this should not be considered a complication. In fact, the purpose of arthroereisis is to
limit the displacement of the talus over the calcaneus. Further investigation is required
to evaluate if subtalar motion restriction recovers after implant removal. They also
believe that supination of the foot in the early postoperative period is not a complication, because it resolves spontaneously 3 to 4 months after surgery.24
The authors advise endorthesis removal in all cases after the foot has achieved bone
maturity (1516 years of age). By that time, the implant has already performed its function and is no longer needed. Removal of the implant is a very simple procedure. In the
authors study, 38% of the feet had the endorthesis removed after completion of the
foots growth period. The average radiologic correction loss in these cases was
minimal and not statistically significant.
In addition to the mentioned advantages of subtalar arthroereisis for flatfoot treatment (simple procedure, good results), this procedure has another important advantage: it allows further future surgery if needed. Osteotomies, arthrodesis, and soft
tissue procedures are not hindered by subtalar arthroereisis.
SUMMARY

Subtalar arthroereisis, often combined with Achilles tendon lengthening, is a simple


and effective way to treat flexible flatfoot in children. Mid- and long-term results are
good, and the procedure does not prevent future treatments.
REFERENCES

1. Gould N, Schneider W, Ashikaga T. Epidemiological survey of foot problems in


the continental United States: 19781979. Foot Ankle 1980;1:810.
2. Garca-Rodriguez A, Martin-Jimenez F, Carnero-Varo M, et al. Flexible flat feet in
children: a real problem? Pediatrics 1999;103:8495.

333

334

 ndez de Retana et al
Ferna

3. Smith SD, Millar EA. Arthroereisis by means of the subtalar polyethylene peg implant
for correction of hindfoot pronation in children. Clin Orthop 1983;181:1525.
4. Viladot A. Surgical treatment of the childs flatfoot. Clin Orthop 1992;283:348.
5. Chambers EF. An operation for the correction of flexible flat feet of adolescents.
West J Surg Obstet Gynecol 1946;54:6034.
6. Grice DS. An extra-articular arthrodesis of the subastragalar joint for correction of
paralytic flat feet in children. J Bone Joint Surg Am 1952;34A(4):92740.
7. Haraldsson S. Operative treatment of pes planovalgus staticus juvenilis. Acta
Orthop Scand 1962;32:4928.
8. Lelievre J. The valgus foot: current concepts and correction. Clin Orthop 1970;70:
4355.
9. Zaret DI, Myerson MS. Arthroereisis of the subtalar joint. Foot Ankle Clin 2003;8:
60517.
10. Needleman RL. Current topic review: subtalar arthroereisis for correction of flexible flatfoot. Foot Ankle Int 2005;26:33646.
11. Subotnick S. The subtalar joint lateral extra-articular arthroereisis: a follow-up
report. J Am Podiatry Assoc 1977;32:2733.
12. Vedantam R, Capelli AM, Schoenecker PL. Subtalar arthroereisis for the correction of planovalgus foot in children with neuromuscular disorders. J Pediatr
Orthop 1998;18:2948.
ndez-Velazques JR. Viladots oper13. Carranza-Bencano A, Zamora-Navas P, Ferna
ation in the treatment of the childs flatfoot. Foot Ankle Int 1997;18:5449.
14. Black PR, Betts RP, Duckworth T, et al. The Viladot implant in flatfoot children.
Foot Ankle Int 2000;21:47881.
15. Giannini S. Operative treatment of the flatfoot: why and how. Foot Ankle Int 1998;
19:528.
cnica personal original para el tratamiento quiru
rgico
16. Alvarez R. Calcaneo stop. Te
ven. In: Epeldegui T, editor. Pie pladel pie plano-valgo del nino y del adolescente jo
. Madrid, Spain: Madrid Vicente; 1995. p. 1747.
no y anomalas del antepie
17. Magnan B, Baldrighi C, Papadia D. Flatfeet: comparison of surgical techniques.
Result of study group into retrograde endorthesis with calcaneus-stop. Ital J Pediatr Orthop 1997;13:2833.
18. Nogarin L. Retrograde endorthesis. Ital J Pediatr Orthop 1997;13:349.
19. Roth S, Sestan B, Tudor A, et al. Minimally invasive calcaneo-stop method for
idiopathic, flexible pes planovalgus in children. Foot Ankle Int 2007;28:9915.
20. Vogler H. Subtalar joint blocking operations for pathological pronation
syndromes. In: McGlamery ED, editor. Comprehensive textbook of foot surgery.
Baltimore (MD): William & Wilkins; 1987. p. 46682.
21. Mosca VS. Flexible flatfoot and tarsal coalition. In: Canale ST, editor. Orthopedic
knowledge update: pediatrics 2. Chicago (IL): American Academy of Orthopaedic Surgeons; 2002. p. 2157.
22. Kidner FC. The prehallux (accessory scaphoid) in relation to flat foot. J Bone Joint
Surg Am 1929;11:8317.
23. Molayem I, Persiani P, Marcovici LL, et al. Complications following correction of
the planovalgus foot in cerebral palsy by arthroereisis. Acta Orthop Belg 2009;
75:3749.
rrez PR, Herrera M. Giannini prosthesis for flatfoot. Foot Ankle Int 2005;26:
24. Gutie
91826.
25. Rocket AK, Mangum G, Mendicino SS. Bilateral intraosseous cystic formation in
the talus: a complication of subtalar arthroereisis. J Foot Ankle Surg 1996;37:
4215.

Pediatric Flatfoot Reconstruction

26. Siff TE, Granberry WM. Avascular necrosis of the talus following subtalar arthroereisis with a polyethylene endoprosthesis: a case report. Foot Ankle Int 2000;21:
2479.
27. Kuwada GT, Dockery GL. Complications following traumatic incidents with STApeg procedures. J Foot Surg 1988;27:2369.
28. Cicchinelli LD, Pascual-Huerta J, Garca-Carmona FJ, et al. Analysis of gastrocnemius recession and medial column procedures as adjuncts in arthroereisis for
the correction of pediatric pes planovalgus: a radiographic retrospective study.
J Foot Ankle Surg 2008;47:38591.
29. Nelson SC, Haycock DM, Little ER. Flexible flatfoot treatment with arthroereisis:
radiographic improvement and child health survey analysis. J Foot Ankle Surg
2004;43:14455.

335

Tendon Transfer Options in


Managing the Adult Flexible
Flatfoot
Michael S. Aronow, MD

KEYWORDS
Posterior tibial tendon dysfunction Flatfoot Tendon transfer Posterior tibial tendon
Tibialis posterior Flexor digitorum longus Flexor hallucis longus
Peroneus brevis
KEY POINTS
Posterior tibialis function may be compromised or associated with pain in adult
patients with symptomatic flexible flatfoot deformity.
Surgical treatment may include debridement and augmentation or replacement of the
diseased posterior tibial tendon using a tendon transfer to decrease pain and attempt
to restore functional hindfoot inversion strength.
The most common tendon transferred in this situation is the flexor digitorum longus
because of its anatomic proximity, in phase firing during the gait cycle, and limited
functional loss if sacrificed.
Other tendon transfers that have been described in place of the flexor digitorum
longus are the flexor hallucis longus and peroneus brevis.

Posterior tibial tendon (PTT) dysfunction is a disorder in which there is a symptomatic


pathologic condition involving the PTT and/or spring ligament complex. Some
patients with PTT dysfunction previously had a normal or slightly cavus arch before
acquiring a flexible flatfoot during adulthood. Other patients with PTT dysfunction
may still have a normal or slightly cavus arch. However, more commonly there is a
preexisting flatfoot that may have developed increased deformity that may be flexible
or fixed. Per the Johnson and Strom classification1 as modified by Myerson,2 stage
1 PTT dysfunction has an intact arch, stage 2 has a flexible flatfoot deformity, stage
3 has a fixed flatfoot deformity, and stage 4 has an underlying stage 2 or stage 3

Commercial relationships: The author is a nonpaid consultant for Arthrex, Inc, which makes an
interference screw shown in this article.
Department of Orthopaedic Surgery, University of Connecticut School of Medicine, Medical Arts
and Research Building, 263 Farmington Avenue, Farmington, CT 06034 4037, USA
E-mail address: aronow@nso.uchc.edu
Foot Ankle Clin N Am 17 (2012) 205226
doi:10.1016/j.fcl.2012.02.001
1083-7515/12/$ see front matter 2012 Elsevier Inc. All rights reserved.

foot.theclinics.com

206

Aronow

deformity with the addition of valgus tilt at the ankle joint secondary to deltoid
ligament laxity or lateral ankle arthritis. In many patients with stage 2 PTT dysfunction
the deformity may only be correctable with the ankle in equinus or the gastrocnemius
relaxed by knee flexion.
The initial treatment of patients with PTT dysfunction is usually nonoperative and
includes supportive orthoses, physical therapy, antiinflammatory medications, and
potentially selective injections. Should that treatment be unsuccessful, surgery may
be performed to address any pathologic condition in the PTT, which is the focus of
this article. Additional soft tissue procedures are often also performed. Patients with
spring and/or deltoid ligament pathology may benefit from direct repair, imbrication,
or graft reconstruction.35 Significant triceps surae contractures should be addressed
with gastrocnemius recession with a modified Baker partial soleus aponeurosis lengthening added or a tendo-Achilles lengthening substituted if there is significant equinus with
the knee flexed.6 How much triceps surae contracture is considered to be significant is
somewhat controversial and beyond the scope of this article. Relatively hindfoot
motion-sparing procedures including osteotomies, limited fusions, and subtalar arthroereisis7 may be added to improve the arch and decrease biomechanical strain on the PTT
repair, tendon transfer, and spring and deltoid ligaments. A comprehensive discussion of
these procedures is also beyond the scope of this article, but they include medial
displacement calcaneal osteotomy,8,9 intracalcaneal lengthening osteotomies such as
the Evans procedure,10 Cotton medial cuneiform dorsal opening wedge osteotomy,11
combined calcaneal osteotomies,12 calcaneocuboid distraction arthrodesis,13 and medial column naviculocuneiform and/or first tarsometatarsal fusions.14 In low demand
patients or patients with significant subtalar arthritis tibialis posterior it may be appropriate
to add a subtalar fusion.15
BIOMECHANICS

The tibialis posterior is the principal supinator of the subtalar joint, an adductor of the
midfoot, and an ankle plantarflexor. Through its multiple attachments on the plantar
aspect of the foot the PTT also helps support the ligaments that maintain the arch of
the foot. During the stance phase of normal gait the tibialis posterior fires eccentrically
after heel strike allowing the hindfoot to pronate in a controlled fashion, which in turn
unlocks the talonavicular and calcaneocuboid joints allowing the foot to better absorb
shock. During midstance and toe-off the tibialis posterior fires concentrically to bring the
foot back into supination, locking the talonavicular and calcaneocuboid joints to provide
a rigid arch so that the gastrocnemius and soleus can effectively push off the forefoot.
Because the PTT has a limited excursion of only 1 to 2 cm, any insult, no matter how
minor, that lengthens this tendon may have an adverse effect on its function.16 In addition
to decreased hindfoot eversion strength, the diminished ability of the lengthened tibialis
posterior to lock the hindfoot and stabilize the arch may lead to decreased push-off
strength and difficulty with gait. However, when a pathologic condition weakens the PTT
with its antagonist the peroneus brevis intact, the arch usually does not immediately
become flat unless there is a concurrent significant spring ligament tear. This delay is
because it takes time for the cumulative increased forces on the spring ligament and the
other plantar ligaments of the foot secondary to the loss of protective effect of the
eccentric firing of the tibialis posterior after heel strike to stretch out these ligaments.
Therefore, theoretically it may be beneficial with respect to arch preservation to maintain
or restore tibialis posterior function and provide sufficient external orthotic support to the
ligaments of the foot early on in the disease process.
With respect to the biomechanical principals of tendon transfer, an excellent review
was recently published by Bluman and Dowd.17 An ideal tendon transfer would

Tendon Transfer for Adult Flexible Flatfoot

provide sufficient strength to meet the patients needs and goals without excessive
sacrifice and be performed technically in a way to maximize the transferred tendons
function and minimize morbidity. Based on the concept that relative muscle strength
is proportional to cross-sectional area, Silver and colleagues18 calculated the relative
strength of the tibialis posterior to be 6.4, the tibialis anterior 5.6, the peroneus longus
5.5, the flexor hallucis longus (FHL) 3.6, the peroneus brevis 2.6, and the flexor
digitorum longus (FDL) 1.8. Therefore, even without taking into account the expected
associated loss of a grade of strength for a transferred tendon, the muscles typically
used to reconstruct the tibialis posterior are much weaker than the one that eventually
failed against at least the same biomechanical stress an isolated tendon transfer
would face. Whereas the FDL and FHL are closer in strength to the principal
antagonist of the PTT, the peroneus brevis, the stronger peroneus longus also acts as
an evertor of the subtalar joint, and its activity begins earlier and ends later during
stance phase than the peroneus brevis. Therefore, the long-term functional results of
tendon transfers to the PTT might be theoretically better if adjunctive procedures
were performed to decrease the biomechanical stress on the tendon transfer by
improving the longitudinal arch of the foot, decreasing antagonistic hindfoot eversion
force, and/or increasing total hindfoot inversion force by increasing the transferred
tendons moment arm and/or preserving the tibialis posterior muscle function when
possible.
Kitaoka and colleagues19 created a cadaver flatfoot deformity model and applied
load to the plantar aspect of the foot and the Achilles, peroneus longus, peroneus
brevis, posterior tibial, and FDL tendons to simulate the midstance phase of gait. They
found that subtalar fusion provided significantly greater deformity correction of foot
position including arch height than a Johnson and Strom1 side-to-side FDL tendon
transfer to the PTT. In a similar study,20 the same investigators found that foot
position including arch height improved substantially after deltoid ligament reconstruction but not after FDL tendon transfer. In a multisegment biomechanical flatfoot
model Arangio and Salathe21 noted that a simulated 10-mm medial displacement
calcaneal osteotomy substantially decreased the load on the first metatarsal and the
moment at the talonavicular joint and increased the load on the fifth metatarsal and
the calcaneocuboid joint. The subsequent addition of an FDL transfer had only a small
additional effect.
CLINICAL DECISION-MAKING

When operating on patients with PTT dysfunction, the surgeon needs to make two
important decisions with respect to preserving PTT function. First a decision needs to
be made as to whether or not the diseased PTT should be preserved or completely
excised. The second decision is whether or not a tendon transfer is indicated to
augment or replace the pathologic PTT and if so, which tendon and technique is used.
If the PTT has minimal tendinopathy it generally should be preserved. On the other
hand, most surgeons would excise a nonfunctional PTT with severe tendinosis and
scarring. There are also chronic complete ruptures in which the proximal stump of the
PTT has retracted well into the leg and cannot be re-approximated to the distal
tendon stump. However, controversy exists with respect to whether or not to excise
or retain the PTT when there is functional tendon with associated areas of tendinosis
or large chronic but repairable tear or tears. One school of thought is that the diseased
PTT is a pain generator that will likely cause persistent discomfort if maintained. The
other school of thought believes that the tibialis posterior is stronger than the muscle
tendon units used to replace it and therefore it is in the patients best interest not to

207

208

Aronow

unnecessarily sacrifice it when just addressing the areas of symptomatic tendinopathy may be sufficient.
Supporting tendon excision are the histopathologic findings originally described by
Trevino and colleagues22 and most recently described and reviewed by Mosier and
colleagues23,24 for patients with stage 2 PTT dysfunction as compared with presumably normal cadaver PTTs. In the latter studies, the PTT dysfunction tendons were all
enlarged or bulbous with loss of normal color and sheen, giving them a dull white
appearance. On microscopic examination there was no evidence of acute or chronic
inflammation but instead degenerative tendinosis with a nonspecific reparative
response characterized by mucinous degeneration, fibroblast hypercellularity, chondroid metaplasia, and neovascularization resulting in marked disruption in collagen
bundle structure and orientation. Therefore, even after debridement a retained PTT
will likely always be histopathologically abnormal. Furthermore, the residual bulk of
the retained often enlarged PTT, especially if combined with the additional bulk of a
tendon transfer, may occasionally lead to discomfort secondary to constriction under
a tight flexor retinaculum or alternatively, inability to adequately close the tendon
sheath possibly leading to discomfort and decreased mobility secondary to scarring
between the tendon and subcutaneous tissue. With respect to concerns about
residual strength and function, if the proximal tibialis posterior muscle seems healthy
and contractile and the more proximal PTT seems normal, consideration can be made
toward tenodesing it to the adjacent FDL transfer. Furthermore, if the distal PTT
stump tendon is relatively normal, then it may also be anastomosed to the tendon
transfer to preserve the pull on the more distal insertions of the PTT on the plantar foot
that help support the arch and adduct the foot.
In addition there may be increased residual strength of the transferred muscle and
tendon secondary to hypertrophy, as well as compensation for decreased muscle
mass by adjunctive procedures including medial displacement calcaneal osteotomy
that increase the moment arm of the transferred tendon relative to the subtalar joint
axis. In 12 patients with unilateral stage 2 PTT dysfunction, magnetic resonance
imaging (MRI) showed that as compared with the asymptomatic leg, there was mean
10.7% atrophy of the tibialis posterior muscle and 17.2% hypertrophy of the FDL
muscle.25 Eight of these patients were reassessed in a subsequent study26 an
average of 14 months after FDL transfer to the navicular, medial displacement
calcaneal osteotomy, and in 4 patients, PTT excision. In the 4 patients with retained
PTTs, the PTT muscle volume decreased to 23% of the contralateral side and the FDL
muscle volume increased to 11% of the contralateral side. In the 4 patients with
resected PTTs, the contralateral side FDL muscle volume increased to 44% of the
contralateral side.
Supporting tendon retention is the clinical observation that not all histopathologically abnormal tendons are symptomatic. Patients with tendinopathy involving other
tendons in the body including the Achilles tendon often do quite well after treatment
including physical therapy, extracorporeal shock wave therapy, platelet-rich plasma
injections, or surgical debridement despite residual areas of tendon thickness and
nodularity. Many patients with stage 2 PTT dysfunction respond to conservative
treatment27 and have minimal if any tenderness over their abnormal residual PTTs.
Gould28 believed that the strength of the PTT was not fully replaceable by a transfer
and that tendon transfer is not necessary if collagen replacement is all that is
necessary to restore muscle function. Finally, whereas based on the relative muscle
strength data generated by Silver and colleagues18 the muscles most commonly
transferred are significantly weaker than the tibialis posterior, the combined strength
of the tibialis posterior and either the FDL or FHL exceeds not only that of the

Tendon Transfer for Adult Flexible Flatfoot

peroneus brevis by itself, but also the combined strength of the peroneus brevis and
the other principal evertor of the subtalar joint, the peroneus longus.
The authors personal preference is to debride, repair, and/or augment areas of
limited tendinosis and tearing in young or active patients and excise similar tendons
in more elderly, low demand patients. A persistently symptomatic retained PTT can
still be resected at a later date, but a resected PTT cannot be subsequently reinserted
in a patient with symptomatic residual functional weakness.
ALTERNATIVES TO TENDON TRANSFER

A dynamic tendon transfer may not be necessary if the patients symptoms are not
associated with pathology in the PTT or if the PTT can be retained with resolution of
pain and restoration of function. Patients with a symptomatic adult-acquired flatfoot
deformity, a significant spring ligament tear, and a normal PTT on careful exploration
can be treated with a spring ligament repair without any additional intervention to the
PTT. Similarly, posterior tibial tendinitis secondary to triceps surae contractures with
no PTT abnormalities on MRI or, if believed to be necessary intraoperative exploration, may respond to an isolated gastrocnemius recession. Posterior tibial tenosynovitis surrounding an otherwise normal tendon can be treated with tenosynovectomy.29,30 An acute PTT laceration may undergo primary repair if the ends can be
brought together and small partial-thickness tears in an otherwise normal tendon may
be repaired. Symptomatic type 2 accessory naviculars are typically surgically treated
with fusion of the synchondrosis31 or a Kidner procedure32 with excision of the
accessory navicular and distal advancement of the PTT if lax. Patients with symptomatic flatfeet but no PTT tenderness or abnormalities on MRI or intraoperative
exploration may undergo an arch-restoring procedure such as arthroereisis, osteotomy, or fusion either without surgical intervention to the PTT, or alternatively PTT
advancement33 or shortening28 should the tendon be believed to be excessively lax.
The Cobb procedure34 and Youngs tenosuspension3537 augment the PTT using
the anterior tibial tendon. In the Cobb procedure the medial half of the distal anterior
tibial tendon is left attached to its normal insertion on the medial cuneiform and first
metatarsal base and the proximal end of the graft sutured to the debrided PTT. In the
Youngs tenosuspension the distal anterior tibial tendon is rerouted through the
navicular, maintaining the normal medial cuneiform and first metatarsal base insertion. Neither of these procedures recreates the function of the tibialis posterior
muscle, and in the Cobb procedure whereas the transferred anterior tibial tendon
provides healthy tendon for structural support, it has no muscle attached. However,
both procedures reinforce the plantar naviculocuneiform ligament to resist the
naviculocuneiform sag commonly seen in PTT dysfunction. Both procedures also
make the anterior tibialis a less effective antagonist to the peroneus longus, thereby
also improving arch height by increasing plantarflexion of the first metatarsal, but at
the expense of weakening hindfoot inversion.
With recent complete rupture of the PTT Gould28 noted that several minutes of
traction to the proximal tendon end might pull the contracted muscle out to length and
allow direct repair. With more chronic ruptures, if traction to the proximal tendon end
produced no or minimal stretch or elasticity, he recommended a tendon transfer.
However, if traction produced elasticity, then repair was recommended using one of
three techniques: a z-lengthening of the PTT could be performed proximal to the
medial malleolus to allow direct repair of the distal tendon ends, the distal gap could
be bridged using a plantaris or second toe long extensor tendon interposition graft
using a Pulvertaft interweave to both ends of the PTT, or the gap could be bridged by
the Cobb method using half of the distal anterior tibial tendon. Other investigators

209

210

Aronow

have augmented the debrided PTT using an extensor digitorum longus tendon to the
fourth toe graft38 or either GraftJacket or OrthAdapt biografts39 instead of tendon
transfers, or performed an endoscopic gastrocnemius recession, percutaneous
medial displacement calcaneal osteotomy, Evans lateral calcaneal lengthening osteotomy, and in some cases a medial column fusion without surgically addressing the
pathologic PTT.40
TENDON TRANSFERS

In stage 1 and 2 PTT dysfunction, a dynamic tendon transfer should be considered


when active hindfoot inversion motion is still desired and the tibialis posterior is
absent, is no longer functional, needs to be excised for significant tendinopathy, or
despite consideration of the previously described PTT-preserving procedures is not
strong enough to adequately perform its function. Whereas there are indications for
tendon transfer in stage 3 and 4 PTT dysfunction, this topic is beyond the scope of
this article. The FDL is the most common tendon transfer used in the treatment of PTT
dysfunction, although the FHL and peroneal tendons have also been used to
reconstruct the PTT.
FDL Transfer

There are several advantages of transferring the FDL to the PTT over the other
available tendons. The FDL originates from the posterior tibia adjacent to the tibialis
posterior and runs anteromedial and then posterolateral to the PTT without any
significant intervening structures from above the flexor retinaculum to the navicular
tuberosity. Therefore, the same incision that is used to expose the PTT can be used
to harvest and transfer the FDL tendon, although if a longer tendon graft is desired
either the incision needs to be lengthened or a second plantar incision added.
Furthermore, unlike the FHL and peroneal tendons, the FDL transfer does not need to
cross the tibial nerve and posterior tibial vessels. The function of the FDL is more
expendable than the other available tendons and, as will be further discussed, is
usually at least partially preserved because of the FHL and quadratus plantae
attachments on the distal FDL tendon stump. The FDL is also in phase with the tibialis
posterior during the gait cycle, although it fires for a shorter percentage of stance
phase. The peroneus longus fires during stance phase close to the same amount of
time as the tibialis posterior, followed in decreasing order by the FHL, the peroneus
brevis, and then the FDL. The main disadvantage of using the FDL is that it is the
weakest of the potential donor tendons.
FDL Transfer Surgical Technique

With FDL transfer, if indicated, a triceps surae lengthening procedure is performed


first. An initial incision is made overlying the PTT from the level of the medial malleolus
to the naviculocuneiform joint (Fig. 1). The PTT sheath is opened leaving the flexor
retinaculum intact. The PTT is carefully inspected visually and by palpation including
the deep surface and distal insertion for any tears, tenosynovitis, thickening, nodularity, or discoloration. An accessory navicular, if present, is excised. If there is any
suspicion of a more proximal tendon pathologic condition based on preoperative
tenderness, MRI, ultrasound, or intraoperative observation including potential tendoscopy up the proximal tendon sheath, then the incision may be extended more
proximally with division and subsequent repair of the flexor retinaculum if needed. A
decision is then made as to whether or not the PTT will be preserved and whether or
not a tendon transfer will be performed. If an FDL tendon transfer is to be performed,

Tendon Transfer for Adult Flexible Flatfoot

Fig. 1. FDL transfer. (A) An incision is made from the naviculocuneiform joint along the PTT
to the posterior inferior medial malleolus, which can be extended distally or proximally as
needed. The PTT (clamp) and FDL (scissors) are identified in their sheaths. If desired the FDL
could be transected here and transferred to the distal PTT in Pulvertaft fashion. (B) The
incision has been extended distally to the first tarsometatarsal joint level, the abductor
hallucis muscle retracted plantarly, and the FDL and if necessary, deeper FHL identified at the
master knot of Henry. In this picture the FDL is more superficial and closer to the tip of the
clamp than the FHL. (C) After making a drill hole in the navicular with an appropriate size
reamer over a guide wire, the FDL tendon is passed plantar to dorsal through navicular tunnel
and with appropriate tension applied secured with an interference screw to bone and sutures
to the distal PTT. (D) If the distal PTT has been excised or has ruptured and retracted, the
incision is extended proximally or a separate proximal incision made over the tibialis posterior
myotendinous junction. The proximal FDL muscle and tendon may be tenodesed to the
tibialis posterior muscle and proximal tendon if the latter muscle is functional. If the PTT
sheath is healthy-appearing, the distal end of the FDL is brought into the proximal wound
and then brought back distally through the PTT sheath under the flexor retinaculum. If not,
the FDL tendon is kept in its own sheath. The FDL is then placed through the navicular tunnel
and secured as previously described.

the FDL is identified posterolateral to the distal PTT by palpating its tendon sheath
while passively flexing and extending the lesser toes. With a knife or dissecting
scissor in the FDL sheath deep to the tendon, the sheath is cut distally toward the
master knot of Henry, which requires detachment and plantar retraction of the
abductor hallucis muscle from the plantar medial aspect of the medial cuneiform,
taking care to avoid bleeding vessels that typically cross the field at that level. The
medial plantar nerve is often found on the superficial aspect of the FDL tendon,41
making it important during distal release of the tendon sheath to stay on the deep and
not superficial aspect of the FDL tendon. With the ankle plantarflexed, the hindfoot
supinated, and the lesser toes plantarflexed, the FDL tendon is carefully transected

211

212

Aronow

distally. The length of FDL tendon required depends on where the tendon is being
transferred to and the tendon fixation technique used.
Panchbhavi and colleagues42 described an alternative technique for harvesting the
FDL tendon distally through a plantar approach. The FDL tendon sheath was exposed
at the level of the medial malleolus and a malleable metal probe with a smooth bulb
at the tip introduced into the sheath and passed gently into the midfoot. A separate
vertical incision was made through the plantar skin and plantar aponeurosis over the
probe and the fibers of the flexor digitorum brevis split to expose the FDL tendon. The
location of the plantar incision was found to be typically midway between the base of
the heel and the base of the second toe and about two-thirds the width of the foot
medially from the lateral border of the foot. After confirming its identity by applying
tension to the proximal FDL, the FDL was transected distally and pulled into the
proximal wound. In 11 of 83 specimens a slip between the FDL and FHL tendons had
to be cut through the distal incision. This connecting slip was exposed by pulling the
FDL tendon distally and plantarflexing the hallux in order to pull the FHL tendon
distally. There was no injury noted to either the medial or plantar nerve in any of the
83 cadaver specimens, although the medial plantar nerve was noted to be on average
0.7 mm from the FDL division site and the lateral plantar nerve 6.6 mm. Oddy and
colleagues43 compared this limited plantar approach with the traditional medial
midfoot exposure and found that the limited plantar approach allowed a mean 22.9
mm additional length of FDL tendon harvest through a mean 15.6 mm decreased
incision length.
Some investigators have recommended after harvesting the FDL44 or FHL45 tendon
for transfer, suturing the distal stump of the tendon to the intact other tendon to
maintain plantarflexion strength of the lesser toes or hallux, respectively. Multiple
investigators have shown that there are usually, but not always, interconnections
between the two tendons near the master knot of Henry.43,46 50 In one study49 these
interconnections from the FHL to more distal FDL tendon were present in 83 of 95
specimens, with 26 of the 85 specimens having an additional FDL to more distal FHL
reciprocal interconnection. In three other studies46 48 proximal to distal interconnections from the FHL to the FDL were also more common than proximal to distal
interconnections from the FDL to FHL, suggesting that preservation of FHL function
after transection of the FDL proximal to the master knot of Henry is more likely than
preservation of FDL function after transection of the FHL proximal to the master knot
of Henry. No interconnections between the FDL and FHL were noted in 4 of 24,46 3
of 24,47 and none of 1648 specimens. Preoperatively, if the patient can plantarflex the
hallux interphalangeal (IP) joint without plantarflexing the lesser toe distal interphalangeal joints, then the surgeon can assume that there is an interconnection from the
FHL to the more distal FDL and as long as the FDL tendon is transected proximal to
this interconnection at the master knot of Henry, then there should be no need to
tenodese the distal stump of the FDL to the FHL. Even if this interconnection is not
present, decreased lesser toe distal interphalangeal plantarflexion strength is not
functionally limiting for most patients and partially compensated by the quadratus
plantae insertion onto the distal FDL and intact metatarsophalangeal and proximal
interphalangeal joint plantarflexion from the interossei, lumbricals, and flexor digitorum brevis. Therefore, the author does not extend the midfoot incision distally to
perform a tenodesis of the distal FDL stump to the FHL in the absence of this
interconnection.
Once harvested, the FDL tendon is stripped of its distal paratenon where it will be
placed into bone or tenodesed to tendon or periosteum and a #1 Kessler passing
suture placed at the distal end of the tendon. If a decision is made to resect the distal

Tendon Transfer for Adult Flexible Flatfoot

PTT or if there has been a previous complete rupture with proximal retraction of the
tendon, the initial incision is either carried more proximally or a new proximal incision
is made over the myotendonous junction of the tibialis posterior and the FDL. If
functional and healthy-appearing, the tibialis posterior muscle and proximal-most
PTT are tenodesed to the proximal FDL with nonabsorbable suture while physiologic
tension is applied to both. Alternatively, this tenodesis can be deferred until later in the
procedure after distal fixation of the FDL tendon. If there is not significant scarring in
the PTT sheath, particularly under the flexor retinaculum, the FDL tendon is brought
into the proximal wound using the passing suture and then rerouted distally through
the PTT sheath. If the PTT sheath has significant scarring or adhesions, the FDL
tendon is left in its own sheath. If the PTT is being preserved and the FDL tendon is
only being transferred to bone or distal tendon, then no exposure of the FDL tendon
proximal to the ankle is necessary and the FDL is typically left in its own sheath.
The FDL tendon may be transferred to the navicular, the medial cuneiform, an intact
PTT, or should the more proximal tendon require excision, the distal PTT stump.
Transferring the FDL tendon into just the distal PTT requires less FDL tendon length
and therefore avoids the potential morbidity of a longer midfoot or second plantar
incision. Another advantage of transfer directly into the distal PTT or its stump is that
it best recreates the normal biomechanical function of the tibialis posterior by acting
on all nine bone insertions as opposed to just one. Therefore, even if the FDL is being
transferred to the navicular or medial cuneiform, consideration should be given to also
tenodesing it to the distal PTT or its stump. Bone fixation of the FDL tendon transfer
is required when there is minimal residual distal PTT stump or the distal PTT needs to
also be reattached to bone because of partial or complete avulsion, significant laxity,
or accessory navicular excision. Biomechanically more secure fixation to bone can
also be obtained, which may allow earlier or more aggressive postoperative motion.
Of the two bone options, the navicular is the more principal insertion of the PTT and
also requires less FDL tendon length to reach it than the medial cuneiform does.
However, the theoretical benefit that the FDL transfer to the medial cuneiform
provides is it may help support the naviculocuneiform joint, which often develops
lateral sag in patients with PTT dysfunction, by buttressing the plantar medial aspect
of the joint and by pulling the medial cuneiform plantarly.12 Mann44 believed that the
placement of the FDL tendon always should be into the navicular bone, not the medial
cuneiform because the navicular is positioned more medial to the subtalar joint axis
and therefore due to the increased moment arm would be a biomechanically more
effective inverter of the subtalar joint and adductor of the foot. A subsequent cadaver
study by Hui and colleagues51 found that transferring the FDL tendon from its native
site to the navicular significantly decreased its moment arm 46% and transferring the
FDL tendon to the medial cuneiform decreased its moment arm 56%.
Fixation to the distal PTT can be performed using a Pulvertaft weave or a
side-to-side repair. The latter may be performed with nonabsorbable suture
between the paratenon-free edges of the two tendons or by making a partialthickness slit in the distal PTT through which debridement can take place and then
suture the FDL tendon in the PTT slit like a hot dog in a bun. When transferred
to the navicular or the medial cuneiform, the FDL tendon is typically placed from
plantar to dorsal through a bone tunnel that is drilled protecting the proximal and
distal articular surfaces and medial cortical wall. In particular, care needs to be
taken to avoid violating the concave proximal articular surface of the navicular and
damaging the important talonavicular joint. The author will initially place a guide
wire through the bone and confirm its position fluoroscopically. A commercially
available reamer can then be used over the guide wire to make a tunnel wider than

213

214

Aronow

the FDL tendon diameter and appropriate for potential interference screw fixation.
A suture passer or thin metal suction tip attached to suction is used to place the
passing suture pass through the bone tunnel, which in turn is used to place the
FDL tendon into or through the tunnel.
Any pathology noted in the spring ligament is addressed with repair or imbrication
with the sutures passed but not yet tied. Adjunctive osteotomies, fusions, or
arthroereisis procedures are then performed prior to tying the spring ligament sutures
and securing the FDL tendon transfer.
With appropriate tension applied to the FDL tendon via the passing suture, the
FDL tendon is secured within the navicular or medial cuneiform tunnel using suture
anchors,52,53 interference screws,54 57 or sewing the distal tendon back to itself. The
author finds that typically a plantar to dorsal 15-mm long and 5.5-mm diameter
bioabsorbable interference screw in the navicular provides secure fixation for an FDL
tendon transected just proximal to the master knot of Henry. The FDL tendon transfer
is then further secured by suturing it to the distal PTT and the periosteum at the
entrance of the navicular (or medial cuneiform) tunnel, and using a free needle to tie
the passing suture to the dorsal navicular (or medial cuneiform) periosteum. If there is
inadequate FDL tendon length to reach into the bone tunnel, the FDL tendon may be
secured to a plantar bone trough in the navicular or medial cuneiform using a suture
anchor and also sutured to the plantar periosteum and distal PTT.
The appropriate tension that the FDL tendon should be secured at is somewhat
controversial and principally anecdotal. If the tendon is secured under minimal tension
there is concern that the tendon transfer will be weak and nonfunctional. On the other
hand, if the tendon is secured in maximum tension there may be concern that there
would be limited excursion and decreased function. However, there is some animal
evidence that a tendon transfer has at least some ability to adapt to initial overtensioning.58 There is also the concern that immobilizing the foot in equinus after FDL
transfer may lead to residual equinus deformity or diminish the correction after
concurrently performed gastrocnemius recession or tendo-Achilles lengthening,
thereby increasing long-term biomechanical stress on the FDL tendon transfer.
Myerson59 warned against overtightening the FDL tendon because it might cause
subluxation out of the posterior tendon groove. He also recommended performing a
tighter repair for heavier patients and those with more extensive deformity and
believed that the foot should be positioned beyond the midline and in inversion
because it always straightens out and will assume a more neutral position over time.
Hansen60 placed the ankle and forefoot in slight flexion and inversion while pulling the
FDL 0.5 to 1.0 cm deep into the medial cuneiform bone, and suturing it over the
anterior tibial tendon or dorsal medial cuneiform capsular tissue. Mann44 firmly pulled
on the FDL tendon with the talonavicular joint in full adduction, the subtalar joint in
inversion, and the ankle in 20 plantarflexion. Haddad and Mann16 subsequently
recommended securing the FDL tendon transfer at a point halfway between where the
tendon sits in the navicular tunnel in its most relaxed state and where it rests when
pulled maximally while adducting the transverse tarsal joints and inverting the subtalar
joint. Shereff61 performed his side-to-side anastomosis of the posterior tibial and FDL
tendons while holding the foot in plantarflexion and inversion with the toes hyperextended. The author tends to position the foot in 30 to 40 of inversion and 10 to 20
of plantarflexion while tensioning and securing the FDL tendon but usually can still get
the ankle to neutral dorsiflexion when splinting at the end of the case.
Whereas there may be a role for the use of biologics with respect to tendon transfer
healing to bone and soft tissue as well as subsequent function with respect to gliding

Tendon Transfer for Adult Flexible Flatfoot

in the tendon sheath, the author is not aware of any published data on their use in the
surgical treatment of PTT dysfunction.
The posterior tibial and/or FDL tendon sheath(s) are closed distally over the
tendon(s) and the flexor retinaculum repaired if disrupted. Care is taken to avoid
constricting the tendons with too tight a closure. The incisions are then closed in
layers and the leg splinted with the exact position dependent on which adjunct
procedures were performed, but ideally in neutral to slight ankle plantarflexion and
neutral to slight supination of the hindfoot.
Postoperative rehabilitation is determined by the adjunctive procedures performed
and the tendon fixation strength. Isolated FDL transfers may be weight-bearing as
tolerated in a cast or boot with protected early motion allowed for reliable patients
with secure fixation.
FDL Transfer Results

Goldner and colleagues50 described using the FDL or FHL for talipes equinovalgus
secondary to acute laceration or progressive nontraumatic deformity. They noted
that plication of the PTT by itself was insufficient to eliminate pain completely and
restore the longitudinal arch and addition of the FDL or FHL tendon transfer
improved the result but still was insufficient to produce a pain-free foot with a
normal arch. They recommended adding a plication and reinforcement of the
medial plantar calcaneonavicular ligament, which they performed along with an
Achilles tendon lengthening.
Jahss62 performed a side-to-side anastomosis of the FDL to an elongated PTT in
6 symptomatic patients with uniformly good results. Inversion strength was restored
with an overall 20% to 25% increase in power.
When direct repair of a tear of a short portion of the PTT or reattachment of the
distal tendon to the navicular was not possible, Johnson63 recommended using a
proximal and distal side-to-side transfer of the FDL to bridge the gap in the
debrided PTT. At 24- to 32-month follow-up all 5 patients who underwent this
procedure could do a single-limb heel-rise test and noted considerable subjective
improvement with relief of pain in the posteromedial ankle and in the medial arch
when weight-bearing, although 3 had continued minor swelling in the region of the
medial malleolus.64
Mann and Thompson65 reported on 14 patients who underwent FDL transfer to the
navicular and 3 patients who underwent PTT advancement for symptomatic flatfoot
deformities. Results were 12 excellent, 1 good, 3 fair, and 1 poor at 2- to 5-year follow-up.
The flatfoot deformity was corrected in 4 patients, improved in 7, unchanged in 4, and
worse in 2. In a subsequent article Mann44 stated that the main indication for an FDL
transfer was a symptomatic flatfoot deformity secondary to tendinosis of the PTT
nonresponsive to conservative treatment with on physical examination at least 15
of subtalar motion, at least 10 of talonavicular adduction, and less than 10 to 12
of fixed forefoot varus. A medial calcaneal osteotomy was added when there was
increased calcaneal valgus, or about 80% of the time in his experience. He
resected diseased tendon and sutured the proximal PTT to the FDL only in young
individuals with isolated areas of tendinosis. Mann also stated that in over a
hundred cases in which an isolated FDL transfer was performed for the previously
mentioned indications, whereas long-term there seemed to be some minor
deterioration of the arch in some cases, fewer than 5% failed and required a
subsequent fusion at long-term follow-up. When he did not adhere to the
previously mentioned indications, Mann stated the need for revision to a fusion
approached 90%.

215

216

Aronow

Conti and colleagues66 noted that treatment for 6 of 20 patients undergoing


isolated PTT debridement and side-to-side anastomosis to the FDL tendon failed at
an average of 14.7 months and required subsequent triple arthrodesis.
Shereff61 reported the results of 17 patients with stage 2 PTT dysfunction who
underwent a partial or complete resection of the segment of discolored, attenuated,
and damaged tendon followed by end-to-end repair to shorten the PTT and
side-to-side tenodesis to the FDL, which was left intact distally. Four of the patients
also had a subtalar fusion. At average 27-month follow up all patients had a normal
single-limb heel rise and 5/5 inversion strength, all but one patient reported unlimited
ambulatory distance, and 10 of the 17 patients displayed an arch nonweightbearing
and 7 out of 17 while weightbearing. Of the patients who did not undergo concurrent
subtalar fusion, 9 had an excellent American Orthopedic Foot and Ankle society
(AOFAS) hindfoot ankle score between 90 and 100, 2 had a good score between 80
and 90, 1 had a fair score between 70 and 80, and 1 patient who was involved in
litigation for a motor vehicle accident had a poor score of less than 70.
Gazdag and Cracchiolo3 resected grossly abnormal areas of the PTT, performed
FDL transfers to the navicular, and addressed an associated spring ligament
pathologic condition in 17 of 18 patients with stage 2 PTT dysfunction. The 1 patient
with a relatively intact PTT underwent tendon shortening and advancement. In an
additional 4 patients with stage 2 PTT dysfunction and intact spring ligaments, 3
patients underwent FDL transfers and 1 had a longitudinal split tear repaired. At
average 32-month follow-up, 14 of the 18 patients with spring ligament pathology and
2 of the 4 without spring ligament pathology had excellent results with dramatic relief
of pain and ability to do a single-leg heel rise, although only 4 had associated active
varus angulation of the heel. Two patients with spring ligament pathology and 1
without spring ligament pathology had fair results with mild medial hindfoot pain on
exertion and inability to walk more than six blocks, but no limitation of daily activities
and the ability to perform a single leg heel rise. The other 3 patients had poor results;
1 who developed reflex sympathetic dystrophy postoperatively, 1 who had medial
hindfoot pain that limited daily activities and inability to perform a single-leg heel rise,
and 1 with moderate lateral hindfoot pain that limited recreational activities, inability
to perform a single-leg heel rise, and mild radiographic talonavicular and subtalar
osteoarthritis. The radiographic average lateral talo-first metatarsal angle improved
from 13 preoperatively to 9 postoperatively for the 19 patients with excellent or fair
results and decreased from 13 to 20 in the 3 patients with poor results, which was
because of an increase in 1 of these 3 patients from 12 preoperatively to 32
postoperatively.
Feldman and colleagues67 performed a similar side-to-side tenodesis of the FDL to
the PTT in 11 patients with stage 2 PTT dysfunction. The PTT was debrided, split tears
repaired, and in 2 cases of insertional pathologic condition advanced distally on the
navicular. At average 34-month follow-up, the mean AOFAS ankle hindfoot score
improved from 38.8 preoperatively to 78.1. Five patients had 100% relief of preoperative symptoms, 1 patient had between 90% and 99% relief of preoperative
symptoms, 2 patients had between 70% and 89% relief of preoperative symptoms,
and 2 patients had poor results with less than 50% pain relief. One of these 2 patients
was initially pain-free until her symptoms returned at 18 months postoperatively and
the other developed signs consistent with subtalar arthritis at 6 months postoperatively and underwent triple arthrodesis.
Lui68 described a technique in which PTT debridement and FDL transfer was
performed through limited incisions with endoscopic assistance. This procedure was
performed along with subtalar arthroereisis and an endoscopically assisted Cobb

Tendon Transfer for Adult Flexible Flatfoot

procedure. One case was illustrated with no residual medial heel or sinus tarsi
impingement pain and a well-corrected flatfoot deformity at 21-month follow-up.
FHL Transfer

As noted previously, Goldner and colleagues50 described using the FDL or FHL for
talipes equinovalgus secondary to acute laceration of progressive nontraumatic
deformity and added a plication and reinforcement of the medial plantar calcaneonavicular ligament and Achilles tendon lengthening. Comparing the two transfers, the
investigators noted that the FHL was more tendinous than the FDL and had greater
muscle mass. They also noted that if left in its normal course under the sustentaculum
tali prior to being tightened, sutured to the distal stump of the PTT, and then folded
back on itself, the FHL tendon had an excellent stabilizing anchor such that it elevated
the sustentaculum tali and aided in relocating the talus.
Sammarco and Hockenberry45 reported their results for FHL transfer in 17 patients
with stage 2 PTT dysfunction. PTT tears were debrided or repaired and a medial
displacement calcaneal osteotomy was also performed. The spring ligament was not
imbricated or repaired. The FHL tendon was transected at the master knot of Henry
and the tendon stump brought proximal to the ankle. The FHL tendon was then
passed deep to the tibial nerve and posterior tibial vessels, brought through the PTT
sheath, and then sutured to the distal PTT and navicular periosteum. At average
18-month follow-up, the average AOFAS hindfoot score had improved from 62.4 to
83.6. Ten patients were satisfied without reservations, 6 patients were satisfied with
minor reservations, and only 1 patient was dissatisfied. There was no statistically
significant radiographic or clinical improvement in the longitudinal arch, with only 3
patients having a normal arch and 6 patients with an arch symmetric to the
contralateral side.
Several studies have looked at morbidity after principally single incision FHL
transfers for Achilles tendinopathy. Richardson and colleagues69 noted decreased
hallux IP plantarflexion strength and decreased pedobarographic pressure under the
hallux distal phalanx but normal pressures underneath the first and second metatarsal
heads and high scores on the AOFAS Hallux MTP-IP Scale. Coull and colleagues70
did not notice any functional weakness of the hallux during activities of daily living.
Den Hartog71 also did not note any significant functional deficit or deformity of the
hallux. Mulier and colleagues47 harvested 24 cadaver FHL tendons through a
two-incision approach and noted 2 complete ruptures of the medial plantar nerve, 1
partial rupture of both the medial and lateral plantar nerves, and stretching of 3 medial
plantar nerves and 1 lateral plantar nerve.
FHL Transfer Surgical Technique

The FHL transfer procedure is done through a similar incision and using many of the
same steps as an FDL transfer. The FDL tendon is exposed through the medial
midfoot incision to where it crosses superficial to the FHL tendon at the master knot
of Henry. The FHL tendon is then identified and with the ankle plantarflexed, the
hindfoot supinated, and the hallux IP joint plantarflexed, carefully transected distally
(Fig. 2). If the patient could preoperatively plantarflex the lesser distal interphalangeal
joints without the hallux IP joints simultaneously plantarflexing, suggesting no
interconnecting slip from the FDL to the FHL tendon, then consideration could be
made for suturing the distal stump of the FHL to the adjacent FDL prior to transecting
the FHL tendon. However, if this step is not performed, the resultant usually mild
functional loss is typically well-tolerated. The FHL tendon is identified proximal to the
flexor retinaculum and the distal FHL tendon stump brought into the proximal wound.

217

218

Aronow

Fig. 2. FHL transfer. The PTT is exposed and debrided through the previously described
medial incision(s) (see Fig. 1). The FHL tendon is identified and transected at the master knot
of Henry and a passing suture placed. (A) The distal end of the FHL transfer is then brought
into the proximal wound at the level of the tibialis posterior myotendinous junction where it
is noted to be posteromedial to the tibial nerve and posterior tibial vessels (clamp). (B) The
FHL tendon is brought deep and then anterolateral to the neurovascular bundle (ribbon)
where it may be tenodesed to healthy tibialis posterior muscle and proximal tendon. The FHL
tendon is brought distally through PTT sheath under the flexor retinaculum (see Fig. 3B),
where it can be placed through a navicular tunnel and then secured to the distal PTT stump
as previously described for the flexor digitorum longus transfer already performed in this
specimen (see Fig. 1).

The tendon is then passed deep to the tibial nerve and posterior tibial vessels,
brought through the PTT sheath, and then transferred to the navicular, the medial
cuneiform, or the distal PTT similar to an FDL transfer. If the distal PTT is no longer
present and the more proximal muscle functional, the proximal tibialis posterior
muscle and tendon can be tenodesed to the proximal FHL muscle and tendon similar
to how it is performed for an FDL transfer.

Tendon Transfer for Adult Flexible Flatfoot

Peroneal Tendon Transfers

Song and Deland72 added a transfer of the peroneus brevis to an FDL transfer, medial
displacement calcaneal osteotomy, and if necessary, spring ligament repair for
patients with stage 2 PTT dysfunction. The rationale behind the procedure was to
provide additional inversion strength to replace the PTT and remove its principal
antagonist so that less strength would be required to maintain the arch and invert the
subtalar joint. The peroneus brevis transfer was used when the FDL tendon was smaller
than usual, which was in less than 10% of cases during the course of their study. The
investigators excised the diseased PTT after proximal tenodesis to the FDL. The distal
FDL was passed through a drill hole made in the navicular. The peroneus brevis was
exposed and released from its insertion from an incision made over the base of the fifth
metatarsal, and then brought into a second incision made over the peroneus brevis
tendon above the level of the ankle. The tendon was then passed behind the tibia into the
proximal medial wound, passed down the PTT sheath, and then placed into the same
navicular bony tunnel as the FDL transfer. With the foot held in moderate inversion and
plantar flexion both the FDL and peroneus tendons were sewn down to periosteum and
soft tissues with nonabsorbable sutures.
For 13 patients with average 20.6-month follow-up there was statistically significant weakness of both hindfoot inversion and eversion as compared with the
contralateral unaffected side, although only 1 patient had less than 4/5 hindfoot
inversion strength. When retrospectively compared with a matched group of patients
of similar age and follow-up that underwent the same procedure without the peroneus
brevis transfer, the peroneus brevis group had a nonstatistically significant higher
average AOFAS hindfoot score (75.8 vs 71.5) and nonstatistically significant lower
average grades of eversion and inversion strength, and eversion range of motion. The
investigators stated that subsequent to their study they now only use the peroneus
brevis transfer in revision surgery when the FDL is either damaged or no longer
available.
Mizel and colleagues73 reported that 10 patients with no peroneal tendon function
secondary to traumatic common peroneal palsy did not develop a hindfoot valgus
deformity after transfer of the entire PTT to the dorsal midfoot or one-half the PTT to
the peroneus longus at average 75-month follow-up. This article suggests that
procedures that decrease peroneus brevis mediated hindfoot eversion and midfoot
abduction force would theoretically limit further flatfoot progression and require the
repaired or substitute PTT to generate less force to perform its normal function.
Analogous to transferring the peroneus longus to the peroneus brevis in the
treatment of the cavus foot,74 transfer of the peroneus brevis to the peroneus longus
would be a logical choice in the treatment of the symptomatic flatfoot, converting
deforming abduction force on the fifth metatarsal to arch-correcting plantarflexion
force on the first metatarsal. Hansen75 described a technique that could be performed
through the same incision as a lateral column lengthening procedure in which the
peroneus brevis was transected at the level of the distal calcaneus, allowed to
retract1 to 2 cm, and after removal of its distal paratenon placed through a slit in the
peroneus longus with minimal tension and sutured. Hansen also described an
alternative technique whereby the peroneus brevis is transected proximal to the
superior peroneal retinaculum with the proximal stump placed through a slit in the
peroneus longus at physiologic tension. With the foot and ankle held in a neutral
plantigrade position, the distal stump of the peroneus brevis is either placed through
a more distal slit in the peroneus longus or tenodesed to the distal fibula. No clinical
results were given. Schweinberger and Roukis76 also described their technique for

219

220

Aronow

peroneus brevis to peroneus longus transfer along with triceps surae lengthening in
diabetic patients undergoing transmetatarsal amputation including the fifth metatarsal
base in order to reduce residual midfoot supination and the risk of plantar lateral
stump ulceration.
Proximal fractional or z-lengthening of the peroneus brevis would have theoretical
benefits in the treatment of stage 2 PTT dysfunction by weakening the PTTs principal
antagonist. However, there are also potential negative effects including discomfort
secondary to potential subsequent peroneal tendinopathy, functional loss due to
weakness of hindfoot eversion, and increased risk of lateral ankle sprain leading to
chronic lateral ankle functional instability. Whereas the author has no personal
experience with peroneus brevis lengthening or transfer to the peroneus longus in the
treatment of PTT dysfunction, the author has had to occasionally perform a complete
peroneus brevis tenotomy along with a triceps surae lengthening procedure in order
to realign a significant stage 3 PTT dysfunction deformity during triple arthrodesis.
Whereas the peroneus longus tendon is also an antagonist to the PTT in the sense
that it is a strong evertor of the hindfoot, it also helps maintain the longitudinal arch
of the foot by plantarflexing the first metatarsal and supporting the plantar ligaments
of the midfoot, particularly the first tarsometatarsal joint. Fried and Hendel77 described transferring the peroneus longus, as well as the FDL, FHL, and extensor
hallucis longus tendons, to the paralyzed PTTs of children with calcaneovalgus
deformities secondary to poliomyelitis. However, the author is not aware of any
published descriptions of transferring the peroneus longus to the PTT in the treatment
of adult PTT dysfunction, although it has been used clinically to reconstruct the spring
ligament.5
Peroneus Brevis Tendon Transfer Surgical Technique

The peroneus brevis tendon transfer procedure is also done using the same medial
incision(s) and many of the same steps as a PTT debridement and FDL transfer. In
addition, incisions are made over the course of the peroneus brevis tendon just proximal
to the fifth metatarsal base insertion and posterior to the fibula at the level of the tibialis
posterior myotendinous junction. The peroneus brevis is identified and transected
distally. With the aid of a passing suture and a long curved clamp, the peroneus brevis
tendon is then brought into the proximal lateral incision (Fig. 3). The peroneus brevis
tendon and some of its attached muscle is then passed deep to the peroneus longus,
FHL, tibial nerve, and posterior tibial vessels into the proximal medial incision. The
peroneus brevis may then be tenodesed to the proximal tibialis posterior muscle, brought
through the distal PTT sheath, and then transferred to the navicular, medial cuneiform, or
distal PTT similar to or in addition to an FDL transfer.
ROLE OF ISOLATED SOFT TISSUE PROCEDURES IN STAGE 2 PTT DYSFUNCTION

It is current orthopedic dogma that patients with stage 2 PTT dysfunction for whom
appropriate nonoperative treatment fails require more than just soft tissue procedures
addressing the PTT pathology with tendon debridement and repair, advancement, or
transfer; spring and/or deltoid ligament pathology; and significant triceps surae
contractures. The addition of concurrent arch-improving procedures has been
strongly recommended including arthroereisis, osteotomy, and fusion of joints not
considered critical for normal hindfoot function.
However, whereas these additional bone procedures seem to be associated in the
orthopedic literature with overall good short- and intermediate-term results including
improved radiographic correction as compared with isolated soft tissue procedures, they are not without added morbidity. Fusions and osteotomies may require

Tendon Transfer for Adult Flexible Flatfoot

Fig. 3. Peroneus brevis transfer. The PTT is exposed and debrided through the previously
described medial incision(s) (see Fig. 1). (A) The peroneus brevis tendon is identified through two
lateral incisions, one made just proximal to the fifth metatarsal base insertion and the other
posterior to the fibula at the level of the tibialis posterior myotendinous junction. The peroneus
brevis is transected distally and brought into the proximal lateral incision with the aid of a passing
suture. The peroneus brevis is identified deep to the more superficial peroneus longus tendon
(ribbon). (B) Using a clamp deep to the tibial nerve, posterior tibial vessels, and flexor hallucis
longus, the peroneus brevis is kept deep to the peroneus longus and passed from the proximal
lateral incision into the proximal medial incision where it may be tenodesed to the tibialis
posterior at the level of its myotendinous junction. The peroneus brevis is then brought distally
through the PTT sheath under the flexor retinaculum, similar to how the flexor hallucis longus
(arrow) was previously passed in this specimen for Fig. 2 The peroneus brevis can then be placed
through a navicular tunnel and secured to the distal PTT as previously performed for the flexor
digitorum longus tendon (arrowhead) in this specimen for Fig. 1.

221

222

Aronow

prolonged nonweightbearing immobilization, which may be particularly difficult for


elderly patients or those with a symptomatic contralateral lower extremity pathology. There may be symptomatic malunion, nonunion, undercorrection, overcorrection, or hardware that requires additional surgery. Although typically there is
increased arch improvement, not uncommonly there is still residual planovalgus
deformity, sometimes significant, particularly if only one osteotomy was performed for a severe preoperative deformity. Some patients who have had
significant flatfoot deformities all their life find suddenly having a normal arch,
particularly one that is unilateral, difficult to get used to and feel as if they are
walking on the lateral aspect of their foot. There are people with nonweightbearing
cavovarus feet who have developed symptomatic weightbearing planovalgus
deformities over time. The author has seen medial displacement calcaneal
osteotomy in conjunction with other procedures including spring ligament repair
and FDL transfer in these people lead to significant postoperative hindfoot varus
with subsequent symptomatic development of lateral foot pain and/or lateral ankle
stability. Evans calcaneal osteotomy may be associated with symptomatic bone
impingement in the sinus tarsi and increased forefoot supination. If not addressed,
this forefoot varus/supination may drive the hindfoot back into valgus analogous
to the cavovarus foot where forefoot valgus secondary to a plantarflexed first
metatarsal can cause compensatory hindfoot varus. Sural incisional neuromas
may develop after medial displacement calcaneal osteotomy. Arthroereisis components not uncommonly need to be removed secondary to lateral subtalar
irritation and the long-term effects of particulate debris from metal, polylactic acid,
and more historically silicone and polyethylene on the adjacent bone and articular
surfaces of the posterior facet are unknown.
Given the previously described morbidity and the fairly good short-term results for
most of the previously discussed reports describing isolated soft tissue procedures
including tendon transfer in stage 2 PTT dysfunction,3,44,50,61,62,64 67 it may be
reasonable to consider performing isolated soft tissue procedures in patients with
mild stage 2 deformity or elderly low demand patients with mild to moderate
deformity. Many of these patients will do quite well with PTT debridement, repair,
and/or transfer, spring ligament imbrication or repair, and if needed, gastrocnemius
recession. A minority of these patients will have noticeable improvement in their arch.
Many of the rest may tolerate residual planovalgus deformity, which in many cases
they have done for most of their life including their contralateral foot. Furthermore, the
medial soft tissue repair may be biomechanically protected by postoperative use of
an insole orthosis and once adequate soft tissue healing has occurred, an aggressive
therapy program similar to those often used successfully in the nonoperative
treatment of stage 2 PTT dysfunction.27 The patients may be early weightbearing as
tolerated in a cast or boot, which may avoid placement in an extended care facility for
an elderly patient with bilateral PTT dysfunction or symptomatic contralateral extremity arthritis.
In the study with the overall worst reported results for isolated soft tissue
procedures,66 30% of the 20 patients had a failure of their reconstruction and required
subsequent triple arthrodesis. However, 70% of the patients did not and may have
avoided an initial triple arthrodesis with less residual postoperative motion and
function. Also in that particular study, treatment for only 1 of 8 patients with grade I
tears of their PTT by preoperative MRI failed and required triple arthrodesis.
Furthermore, if an isolated soft tissue procedure is unsuccessful, minimal bridges
have burned and a more extensive bone procedure can subsequently be added. If the
medial soft tissue procedure has failed, a tendon transfer can be performed if not

Tendon Transfer for Adult Flexible Flatfoot

done originally; the previous tendon transfer may undergo debridement, advancement, or replacement transfer with the remaining FDL, FHL, or peroneus brevis
tendons, or biografts,39 or a hindfoot fusion performed. Therefore, whereas the author
would still recommend considering adjunctive bone procedures for most patients
undergoing surgery for stage 2 PTT dysfunction, particularly those patients who are
younger, more active, or who have more significant deformities, there are several
situations where with appropriate patient counseling about expectations the author
believes an isolated soft tissue procedure that may include tendon transfer is quite
reasonable.
REFERENCES

1. Johnson KA, Strom DE. Tibialis posterior tendon dysfunction. Clin Orthop Relat Res
1989;(239):196 206.
2. Myerson MS. Adult acquired flatfoot deformity: treatment of dysfunction of the
posterior tibial tendon. Instr Course Lect 1997;46:393 405.
3. Gazdag AR, Cracchiolo A 3rd. Rupture of the posterior tibial tendon. Evaluation of
injury of the spring ligament and clinical assessment of tendon transfer and ligament
repair. J Bone Joint Surg Am 1997;79(5):675 81.
4. Jeng CL, Bluman EM, Myerson MS. Minimally invasive deltoid ligament reconstruction
for stage IV flatfoot deformity. Foot Ankle Int 2011;32(1):2130.
5. Williams BR, Ellis SJ, Deyer TW, et al. Reconstruction of the spring ligament using a
peroneus longus autograft tendon transfer. Foot Ankle Int 2010;31(7):56777.
6. Aronow MS. Triceps surae contractures associated with posterior tibial tendon
dysfunction. Tech Orthop 2000;15:164 73.
7. Needleman RL. A surgical approach for flexible flatfeet in adults including a subtalar
arthroereisis with the MBA sinus tarsi implant. Foot Ankle Int 2006;27(1):9 18.
8. Guyton GP, Jeng C, Krieger LE, et al. Flexor digitorum longus transfer and medial
displacement calcaneal osteotomy for posterior tibial tendon dysfunction: a middleterm clinical follow-up. Foot Ankle Int 2001;22(8):62732.
9. Myerson MS, Badekas A, Schon LC. Treatment of stage II posterior tibial tendon
deficiency with flexor digitorum longus tendon transfer and calcaneal osteotomy. Foot
Ankle Int 2004;25(7):44550.
10. Thomas RL, Wells BC, Garrison RL, et al. Preliminary results comparing two methods
of lateral column lengthening. Foot Ankle Int 2001;22(2):10719.
11. Hirose CB, Johnson JE. Plantar flexion opening wedge medial cuneiform osteotomy
for correction of fixed forefoot varus associated with flatfoot deformity. Foot Ankle Int
2004;25(8):568 74.
12. Moseir-LaClair S, Pomeroy G, Manoli A 2nd. Intermediate follow-up on the double
osteotomy and tendon transfer procedure for stage II posterior tibial tendon insufficiency. Foot Ankle Int 2001;22(4):28391.
13. Toolan BC, Sangeorzan BJ, Hansen ST Jr. Complex reconstruction for the treatment
of dorsolateral peritalar subluxation of the foot. Early results after distraction arthrodesis of the calcaneocuboid joint in conjunction with stabilization of, and transfer of
the flexor digitorum longus tendon to, the midfoot to treat acquired pes planovalgus in
adults. J Bone Joint Surg Am 1999;81(11):1545 60.
14. Chi TD, Toolan BC, Sangeorzan BJ, et al. The lateral column lengthening and medial
column stabilization procedures. Clin Orthop Relat Res 1999;(365):8190.
15. Johnson JE, Cohen BE, DiGiovanni BF, et al. Subtalar arthrodesis with flexor digitorum longus transfer and spring ligament repair for treatment of posterior tibial tendon
insufficiency. Foot Ankle Int 2000;21(9):7229.

223

224

Aronow

16. Haddad SL, Mann RA. Flatfoot in adults. In: Coughlin MJ, Mann RA, Saltzman C,
editors. Surgery of the foot and ankle. 8th edition. St Louis (MO): Mosby; 2011. p.
1007 86.
17. Bluman EM, Dowd T. The basics and science of tendon transfers. Foot Ankle Clin
2011;16(3):38599.
18. Silver RL, de la Garza J, Rang M. The myth of muscle balance. A study of relative
strengths and excursions of normal muscles about the foot and ankle. J Bone Joint
Surg Br 1985;67(3):4327.
19. Kitaoka HB, Luo ZP, An KN. Subtalar arthrodesis versus flexor digitorum longus
tendon transfer for severe flatfoot deformity: an in vitro biomechanical analysis. Foot
Ankle Int 1997;18(11):710 5.
20. Kitaoka HB, Luo ZP, An KN. Reconstruction operations for acquired flatfoot: biomechanical evaluation. Foot Ankle Int 1998;19(4):2037.
21. Arangio GA, Salathe EP. A biomechanical analysis of posterior tibial tendon dysfunction, medial displacement calcaneal osteotomy and flexor digitorum longus transfer in
adult acquired flat foot. Clin Biomech (Bristol, Avon) 2009;24(4):38590.
22. Trevino S, Gould N, Korson R. Surgical treatment of stenosing tenosynovitis at the
ankle. Foot Ankle 1981;2(1):37 45.
23. Mosier SM, Lucas DR, Pomeroy G, et al. Pathology of the posterior tibial tendon in
posterior tibial tendon insufficiency. Foot Ankle Int 1998;19(8):520 4.
24. Mosier SM, Pomeroy G, Manoli A 2nd. Pathoanatomy and etiology of posterior tibial
tendon dysfunction. Clin Orthop Relat Res 1999;(365):1222.
25. Wacker J, Calder JD, Engstrom CM, et al. MR morphometry of posterior tibialis
muscle in adult acquired flat foot. Foot Ankle Int 2003;24(4):354 7.
26. Rosenfeld PF, Dick J, Saxby TS. The response of the flexor digitorum longus and
posterior tibial muscles to tendon transfer and calcaneal osteotomy for stage II
posterior tibial tendon dysfunction. Foot Ankle Int 2005;26(9):671 4.
27. Alvarez RG, Marini A, Schmitt C, et al. Stage I and II posterior tibial tendon dysfunction
treated by a structured nonoperative management protocol: an orthosis and exercise
program. Foot Ankle Int 2006;27(1):2 8.
28. Gould JS. Direct repair of the posterior tibial tendon. Foot Ankle Clin 1997;2(2):
275 80.
29. Bare AA, Haddad SL. Tenosynovitis of the posterior tibial tendon. Foot Ankle Clin
2001;6(1):37 66.
30. Teasdall RD, Johnson KA. Surgical treatment of stage I posterior tibial tendon
dysfunction. Foot Ankle Int 1994;15(12):646 8.
31. Malicky ES, Levine DS, Sangeorzan BJ. Modification of the Kidner procedure with
fusion of the primary and accessory navicular bones. Foot Ankle Int 1999;20(1):53 4.
32. Kidner FC. The pre-hallux (accessory scaphoid) and its relation to flatfoot. J Bone
Joint Surg 1929;11(4):8317.
33. Miller OL. A plastic flat foot operation. J Bone Joint Surg 1927;9(1):84 91.
34. Knupp M, Hintermann B. The Cobb procedure for treatment of acquired flatfoot
deformity associated with stage II insufficiency of the posterior tibial tendon. Foot
Ankle Int 2007;28(4):416 21.
35. Jacobs AM. Soft tissue procedures for the stabilization of medial arch pathology in the
management of flexible flatfoot deformity. Clin Podiatr Med Surg 2007;24(4):657 65,
viiviii.
36. Sekiya JK, Saltzman CL. Long term follow-up of medial column fusion and tibialis
anterior transposition for adolescent flatfoot deformity. Iowa Orthop J 1997;17:1219.
37. Young CS. Operative treatment of pes planus. Surg Gynecol Obstet 1939;99:1099
101.

Tendon Transfer for Adult Flexible Flatfoot

38. Kettlekamp BB, Alexander HH. Spontaneous rupture of the posterior tibial tendon. J
Bone Joint Surg Am 1969;51A(4):759 64.
39. Lee D. Effects of posterior tibial tendon augmented with biografts and calcaneal
osteotomy in stage II adult-acquired flatfoot deformity. Foot Ankle Spec 2009;2(1):
2731.
40. Didomenico L, Stein DY, Wargo-Dorsey M. Treatment of posterior tibial tendon
dysfunction without flexor digitorum tendon transfer: a retrospective study of 34
patients. J Foot Ankle Surg 2011;50(3):293 8.
41. Sullivan RJ, Aronow MS. Anatomical relationships of the medial plantar nerve within
the midfoot. Presented at the American Orthopaedic Foot and Ankle Society 17th
Annual Summer Meeting. San Diego (CA), July 19, 2001.
42. Panchbhavi VK, Yang J, Vallurupalli S. Minimally invasive method of harvesting the
flexor digitorum longus tendon: a cadaver study. Foot Ankle Int 2008;29(1):42 8.
43. Oddy MJ, Flowers MJ, Davies MB. Flexor digitorum longus tendon exposure for
flatfoot reconstruction: a comparison of two methods in a cadaveric model. Foot
Ankle Surg 2010;16(2):8790.
44. Mann RA. Posterior tibial tendon dysfunction. Treatment by flexor digitorum longus
transfer. Foot Ankle Clin 2001;6(1):77 87.
45. Sammarco GJ, Hockenbury RT. Treatment of stage II posterior tibial tendon dysfunction with flexor hallucis longus transfer and medial displacement calcaneal osteotomy.
Foot Ankle Int 2001;22(4):30512.
46. LaRue BG, Anctil EP. Distal anatomical relationship of the flexor hallucis longus and
flexor digitorum longus tendons. Foot Ankle Int 2006;27(7):528 32.
47. Mulier T, Rummens E, Dereymaeker G. Risk of neurovascular injuries in flexor hallucis
longus tendon transfers: an anatomic cadaver study. Foot Ankle Int 2007;28(8):
910 5.
48. OSullivan E, Carare-Nnadi R, Greenslade J, et al. Clinical significance of variations in
the interconnections between flexor digitorum longus and flexor hallucis longus in the
region of the knot of Henry. Clin Anat 2005;18(2):1215.
49. Wapner KL, Hecht PJ, Shea JR, et al. Anatomy of second muscular layer of the foot:
considerations for tendon selection in transfer for Achilles and posterior tibial tendon
reconstruction. Foot Ankle Int 1994;15(8):420 3.
50. Goldner JL, Keats PK, Bassett FH 3rd, et al. Progressive talipes equinovalgus due to
trauma or degeneration of the posterior tibial tendon and medial plantar ligaments.
Orthop Clin North Am 1974;5(1):39 51.
51. Hui HE, Beals TC, Brown NA. Influence of tendon transfer site on moment arms of the
flexor digitorum longus muscle. Foot Ankle Int 2007;28(4):4417.
52. Myerson MS, Cohen I, Uribe J. An easy way of tensioning and securing a tendon to
bone. Foot Ankle Int 2002;23(8):7535.
53. Sullivan RJ, Gladwell HA, Aronow MS, et al. An in vitro study comparing the use of
suture anchors and drill hole fixation for flexor digitorum longus transfer to the
navicular. Foot Ankle Int 2006;27(5):363 6.
54. Harris NJ, Ven A, Lavalette D. Flexor digitorum longus transfer using an interference
screw for stage 2 posterior tibial tendon dysfunction. Foot Ankle Int 2005;26(9):
7812.
55. Louden KW, Ambrose CG, Beaty SG, et al. Tendon transfer fixation in the foot and
ankle: a biomechanical study evaluating two sizes of pilot holes for bioabsorbable
screws. Foot Ankle Int 2003;24(1):6772.
56. Sabonghy EP, Wood RM, Ambrose CG, et al. Tendon transfer fixation: comparing a
tendon to tendon technique vs. bioabsorbable interference-fit screw fixation. Foot
Ankle Int 2003;24(3):260 2.

225

226

Aronow

57. Wukich DK, Rhim B, Lowery NJ, et al. Biotenodesis screw for fixation of FDL transfer
in the treatment of adult acquired flatfoot deformity. Foot Ankle Int 2008;29(7):730 4.
58. Takahashi M, Ward SR, Marchuk LL, et al. Asynchronous muscle and tendon
adaptation after surgical tensioning procedures. J Bone Joint Surg Am 2010;92(3):
664 74.
59. Myerson MS. Correction of flatfoot deformity in the adult. In: Myerson MS, Reconstructive foot and ankle surgery. Philadelphia: Elsevier Saunders; 2005. p. 189 215.
60. Hansen ST. Transfer of the flexor digitorum communis to the first cuneiform. In:
Hansen ST. Functional reconstruction of the foot and ankle. Philadelphia: Lippincott,
Williams & Wilkins; 2000. p. 430 2.
61. Shereff, MJ. Treatment of the ruptured posterior tibial tendon with direct repair and
tenodesis. Foot Ankle Clin 1997;2(2):28196.
62. Jahss MH. Spontaneous rupture of the tibialis posterior tendon: clinical findings,
tenographic studies, and a new technique of repair. Foot Ankle 1982;3(3):158 66.
63. Johnson KA. Tibialis posterior tendon rupture. Clin Orthop Relat Res 1983;(177):
140 7.
64. Funk DA, Cass JR, Johnson KA. Acquired adult flat foot secondary to posterior
tibial-tendon pathology. J Bone Joint Surg Am 1986;68(1):95102.
65. Mann RA, Thompson FM. Rupture of the posterior tibial tendon causing flat foot.
Surgical treatment. J Bone Joint Surg Am 1985;67(4):556 61.
66. Conti S, Michelson J, Jahss M. Clinical significance of magnetic resonance imaging in
preoperative planning for reconstruction of posterior tibial tendon ruptures. Foot Ankle
1992;13(4):208 14.
67. Feldman NJ, Oloff LM, Schulhofer SD. In situ tibialis posterior to flexor digitorum
longus tendon transfer for tibialis posterior tendon dysfunction: a simplified surgical
approach with outcome of 11 patients. J Foot Ankle Surg 2001;40(1):27.
68. Lui TH. Endoscopic assisted posterior tibial tendon reconstruction for stage 2
posterior tibial tendon insufficiency. Knee Surg Sports Traumatol Arthrosc 2007;
15(10):1228 34.
69. Richardson DR, Willers J, Cohen BE, et al. Evaluation of the hallux morbidity of singleincision flexor hallucis longus tendon transfer. Foot Ankle Int 2009;30(7):62730.
70. Coull R, Flavin R, Stephens MM. Flexor hallucis longus tendon transfer: evaluation of
postoperative morbidity. Foot Ankle Int 2003;24(12):931 4.
71. Den Hartog BD. Flexor hallucis longus transfer for chronic Achilles tendonosis. Foot
Ankle Int 2003;24(3):2337.
72. Song SJ, Deland JT. Outcome following addition of peroneus brevis tendon transfer
to treatment of acquired posterior tibial tendon insufficiency. Foot Ankle Int 2001;
22(4):301 4.
73. Mizel MS, Temple HT, Scranton PE Jr, et al. Role of the peroneal tendons in the
production of the deformed foot with posterior tibial tendon deficiency. Foot Ankle Int
1999;20(5):2859.
74. Ryssman DB, Myerson MS. Tendon transfers for the adult flexible cavovarus foot.
Foot Ankle Clin 2011;16(3):43550.
75. Hansen ST. Transfer of the peroneus brevis to the peroneus longus. In: Hansen ST.
Functional reconstruction of the foot and ankle. Philadelphia: Lippincott, Williams &
Wilkins; 2000. p. 439 41.
76. Schweinberger MH, Roukis TS. Balancing of the transmetatarsal amputation with peroneus
brevis to peroneus longus tendon transfer. J Foot Ankle Surg 2007;46(6):5104.
77. Fried A, Hendel C. Paralytic deformity of the ankle: replacement of the paralyzed
tibialis posterior by the peronaeus longus. J Bone Joint Surg Am1959;39A(4):92132.

Ten d o n Tr a n s f e r s in t h e Tre a t m e n t
of the Adult Flatfoot
Jonathon D. Backus,

MD

, Jeremy J. McCormick,

MD

b,

KEYWORDS
 Adult acquired flatfoot deformity  Posterior tibial tendon dysfunction
 Tendon transfer  Tibialis posterior  Flexor digitorum longus  Flexor hallicus longus
 Peroneus brevis  Peroneus longus
KEY POINTS
 Adult acquired flatfoot deformity (AAFD) describes a condition of progressive hindfoot
valgus, forefoot abduction, and forefoot varus. It is most commonly caused by posterior
tibial tendon dysfunction.
 Patients who have AAFD often complain of posteromedial hindfoot pain, a progressive
change in the shape of the foot, difficulty with weight bearing, and gait abnormalities.
 Conservative management consists of nonsteroidal antiinflammatory medications, selective corticosteroid injections, and physical therapy. Patients also benefit from temporary
immobilization in a cast or boot, custom foot orthosis, or a custom ankle-foot orthosis
or Arizona brace.
 In stage I and II posterior tibial tendon (PTT) dysfunction in which conservative management has failed, many clinicians think that the deformity can be corrected with PTT
debridement, tendon transfer, and a medial displacement calcaneal osteotomy. Spring
ligament repair may also be indicated in certain cases.
 In more significant stage II deformities, concomitant procedures such as an additional
lateral column-lengthening osteotomy and selective fusions may be required.
 In the case of a stage II flexible flatfoot deformity, our preference is to excise the diseased
PTT tendon and use a flexor digitorum longus (FDL) transfer through a tunnel in the navicular tuberosity. The FDL tendon is typically long enough to pass through the tunnel from
plantar to dorsal and be secured back to itself with nonabsorbable suture.
Continued

Financial Disclosures: There are no financial relationships or support to disclose for J.D. Backus.
Research funding was received from Wright Medical Technologies, Inc and Midwest Stone Institute, paid laboratory instructing for Synthes, Inc and Integra Life Sciences, Inc; there are no
conflicts with relation to this article for J.J. McCormick.
a
Department of Orthopaedic Surgery, Washington University School of Medicine in St Louis,
Campus Box 8233, 660 South Euclid Avenue, St Louis, MO 63110, USA; b Department of Orthopaedic Surgery, Washington University School of Medicine in St Louis, 14532 South Outer Forty
Drive, Suite 210, Chesterfield, MO 63017, USA
* Corresponding author.
E-mail address: mccormickj@wudosis.wustl.edu
Foot Ankle Clin N Am 19 (2014) 2948
http://dx.doi.org/10.1016/j.fcl.2013.11.002
1083-7515/14/$ see front matter 2014 Elsevier Inc. All rights reserved.

foot.theclinics.com

30

Backus & McCormick

Continued
 Patients with stage III and stage IV deformities typically require a hindfoot arthrodesis. A
plantarflexion osteotomy or fusion of the first ray may also be required in patients with
persistent forefoot varus. Furthermore, there may be a role for tendon transfer in patients
who undergo hindfoot arthrodesis.
 The decision to proceed with surgical correction of AAFD should be made after conservative treatment options have failed. Careful preoperative planning is necessary to choose
the appropriate reconstructive procedures given a patients clinical examination, deformity, and radiographic findings.

PATHOPHYSIOLOGY OF ADULT ACQUIRED FLATFOOT DEFORMITY

Adult acquired flatfoot deformity (AAFD) describes a condition of progressive hindfoot


valgus, forefoot abduction, and forefoot varus. Although there are several causes of
AAFD, posterior tibial tendon (PTT) dysfunction is the most common cause.1 The posterior tibialis (PT) muscle is a powerful supinator of the subtalar joint, adductor of the
midfoot, and serves to assist plantar flexion of the ankle. It has 9 attachments to the
medial midfoot that help this muscle statically support the arch of the foot along
with the spring ligament and the plantar calcaneonavicular ligament.2,3 Therefore,
the PT muscle has important functions throughout the stance phase of gait and in
the overall structure and function of the foot and ankle.
During the early part of stance phase, the PT muscle contracts eccentrically allowing the hindfoot to pronate in a controlled fashion and subsequently unlock the talonavicular and calcaneocuboid joints. This function allows the foot to become supple
and adapt to the undulations of the underlying surface that it contacts. In toe-off,
the muscle contracts concentrically, supinating the hindfoot and locking the talonavicular and calcaneocuboid joints, which provides a rigid fulcrum for the gastrocsoleus
muscle to propel the body forward and begin the swing phase of gait.
In the early stages of PTT disorder, the dysfunctional tendon leaves the spring and calcaneonavicular ligaments unsupported in counteracting the antagonistic hindfoot pronation and forefoot abduction forces of the peroneus brevis (PB). The arch is usually
maintained at first; however, with continued PTT dysfunction and the stress of ambulation, the spring ligament eventually fails and the arch gradually collapses.4 Persistent
hindfoot valgus and forefoot abduction eventually result in forefoot varus, which may
manifest as dorsiflexion of the first ray or global forefoot varus at the transverse tarsal joint
(Fig. 1).5 At first, these deformities are flexible and passively correctable; however, consequent synovitis and scarring lead to the foot becoming fixed in its malpositioned posture.
SYMPTOMS

Patients with AAFD typically complain of posteromedial hindfoot pain below the medial
malleolus, a progressive change in the shape of the foot, and difficulty with weight
bearing; however, patients rarely report a specific instance in which the foot collapses.
Some patients also experience lateral pain near the sinus tarsi if the deformity is such
that the lateral malleolus impinges with the calcaneus.6 Patients often note gait disturbances such as difficulty with push-off, long strides, and an inability to run.7
STAGING

Johnson and Strom8 described the first classification system for PTT dysfunction and
AAFD in 1998. This classification is helpful in that it is often used to determine

Tendon Transfers in the Adult Flatfoot

Fig. 1. (A) Preoperative assessment of hindfoot valgus. (B) Preoperative assessment of forefoot varus. (Courtesy of Jeffrey E. Johnson, MD, Chesterfield, MO.)

treatment. Stage I represents PT tendinitis and degeneration, with normal tendon


length and an intact medial longitudinal arch. In stage II, the PTT elongates and a flexible flatfoot deformity develops. Stage III represents a flatfoot with fixed deformity.
Myerson9 added a 4th stage to Johnson and Stroms8 classification that described
deltoid ligament attenuation and a fixed valgus ankle deformity as a result of prolonged planovalgus foot deformity (Table 1).
CONSERVATIVE MANAGEMENT

Nonoperative treatment can be performed initially for a deformity of any stage; however, it is most effective for stage I and mild stage II deformities.10 Initial treatment
consists of nonsteroidal antiinflammatory medications, physical therapy, and a
custom-molded orthosis. Clinicians can also prescribe a semirigid foot orthosis that
is posted at the medial heel and lateral forefoot, which helps support the arch and
Table 1
Classification system for PTT dysfunction
Stage

Clinical Finding

I8

Pain along PTT


Intact medial longitudinal arch
Able to perform single-limb heel rise

II8

Loss of medial longitudinal arch (flatfoot)


Unable to perform single-limb heel rise
Hindfoot can be passively reduced to neutral posture

III8

Loss of medial longitudinal arch (flatfoot)


Unable to perform single-limb heel rise
Hindfoot cannot be passively reduced to neutral posture

IV9

Stage II or III findings with valgus posture at the ankle caused by deltoid
insufficiency

31

32

Backus & McCormick

correct hindfoot valgus and pronation with the medial heel post and arch support, and
also corrects forefoot varus through lateral forefoot support. If the forefoot varus
deformity is rigid, an accommodative semirigid orthosis with a medial forefoot post
is recommended to support the forefoot in its fixed position. If an orthosis is unsuccessful or the patient has a more rigid deformity, a custom-molded plastic and leather
composite lace-up brace (ie, Arizona brace) or a rigid ankle-foot orthosis (AFO) with a
molded foot orthosis inside the brace is typically offered. Selective intra-articular corticosteroid injections are also considered in later stages of PTT dysfunction as hindfoot
joints become inflamed and arthritic.
OPERATIVE MANAGEMENT

Many combinations of bone and soft tissue procedures have been used in treating
AAFD without a clear consensus on an optimal management algorithm.2,5,1118 The
focus of operative treatment involves restoring PTT function and correcting the resulting deformity. Current practice supports the theory that isolated medial soft tissue procedures only provide temporary functional restoration; therefore, bony corrective
procedures of the underlying deformity are necessary to prevent attenuation or failure
of the soft tissue repair.19,20
Many clinicians think that stage I and II deformities refractory to conservative management can be corrected with PTT debridement, tendon transfer, and a medial
displacement calcaneal osteotomy (MDCO). This treatment allows sufficient hindfoot
valgus correction to protect the medial soft tissue repair and tendon transfer without
sacrificing significant motion.7,1923 More significant stage II deformities might require
an additional lateral column-lengthening procedure to correct forefoot abduction, or
limited joint arthrodeses to optimize correction.19,2326 Other clinicians have suggested that subtalar arthroereisis may be advantageous in decreasing the strain on
PTT repair and tendon transfer as well.16 Stage III and IV deformities often require a
more extensive arthrodesis procedure in addition to tendon transfers and repairing
or reconstructing any deltoid ligament disorder.27
Spring ligament and deltoid ligament disorders should be addressed, if present, in
the form of direct repair, imbrication, or reconstruction.4,28,29 It has recently been suggested that AAFD can be caused by an isolated spring ligament injury in the setting
of a normal PTT.30 In these cases, an isolated spring ligament repair may be the
only procedure necessary to correct the deformity.30 If significant forefoot varus remains after initial soft tissue reconstruction and bony procedures, all stages of PTT
dysfunction may also require a bony procedure involving the first ray to decrease
the risk of excessive loading on the lateral border of the foot and possible deltoid ligament strain,24 which is often achieved with a Cotton osteotomy,14,31 reverse Cotton
osteotomy,32 or 1st tarsometatarsal fusion. Triceps surae contractures should also be
addressed with a gastrocnemius recession or tendo-Achilles lengthening, because
this is a contributing factor in AAFD.33
There are many options for reconstruction of AAFD involving both bone and soft
tissue work. This article discusses the role of tendon transfers in medial soft tissue
restoration in AAFD. Common tendon transfers include flexor digitorum longus
(FDL), flexor hallucis longus (FHL), PB in addition to FDL or FHL transfer, and PB to
peroneus longus (PL) transfer.
TENDON TRANSFER PRINCIPLES

As recently outlined by Bluman and Dowd,34 the ideal tendon transfer must possess
5 basic properties to effectively correct deformity and improve motor function.

Tendon Transfers in the Adult Flatfoot

Thus, donor tendons should have appropriate strength, appropriate excursion, similar
direction and attachment as the replaced tendon, a single function, and ideally be in
phase with the tendon that it is replacing. Muscle strength (or its ability to generate
force) is directly proportional to the cross-sectional area of the muscle, and muscle
excursion is directly proportional to fiber length. A 1985 cadaveric study by Silver
and colleagues35 calculated the relative strength and excursion of the muscles about
the ankle using these principles. The relative strength of the PTT was 6.4; FHL, 3.6;
FDL, 1.8; PL, 5.5; and PB, 2.6. Likewise, the excursion of the PTT was 4.1 cm; FHL,
4.8 cm; FDL, 4.8 cm; PL, 4.7 cm; and PB, 4.1 cm (Table 2).
After considering these principles and data, the FHL and FDL seem to be appropriate tendons to replace the diseased PT because they have slightly greater excursions and relative strength that is similar to the antagonistic PB muscle. Both
muscles are also in phase with the PT on the medial aspect of the ankle.
DECISIONS INVOLVING THE DISEASED PTT

The first decision in planning for a tendon transfer in AAFD involves managing the
diseased PTT. The surgeon must decide whether to debride, excise, or keep the
damaged tendon. Aronow36 discussed the controversy surrounding this topic in a
recent review article. He presented theories regarding the diseased PTT: (1) in
AAFD, the PTT is a pain generator and pain cannot be alleviated until the tendon is
removed; and (2) the PTT is significantly stronger than the muscle tendon units used
to replace it, and therefore it should not be sacrificed unnecessarily.
Giving credence to the first theory is the histologic and gross comparisons of patients with stage II dysfunction and presumed normal cadaveric specimens.37,38 The
PTT in affected patients was hypertrophied with loss of normal appearance. Histologic
analysis found degenerative tendinosis characterized by increased mucin content,
fibroblast hypercellularity, chondroid metaplasia, and neovascularization, which
resulted in a markedly disorganized collagen structure. Furthermore, neither acute
nor chronic inflammation was identified. In contrast, another publication37 showed significant fibrosis within the tendon; however, it is unclear whether these changes preceded or postdated PTT dysfunction.
Aronow36 also indicated that the often-hypertrophied PTT combined with a tendon
transfer would be too bulky under a tight flexor retinaculum and possibly would prohibit tendon sheath closure. Therefore, the retained tendon could cause continued
discomfort secondary to its bulk or excessive scarring between the tendons and surrounding subcutaneous tissue.

Table 2
Comparison of tendon properties
Tendon

Relative Strength

Excursion (cm)

PTT

6.4

4.1

FDL

1.8

4.8

FHL

3.6

4.8

PB

2.6

4.1

PL

5.5

4.7

Data from Silver RL, de la Garza J, Rang M. The myth of muscle balance. A study of relative
strengths and excursions of normal muscles about the foot and ankle. J Bone Joint Surg Br
1985;67:4327.

33

34

Backus & McCormick

Additional evidence that supports excising the diseased PTT includes the resultant
hypertrophy and strength gains of the transferred tendon. A study by Rosenfeld and
colleagues39 found that patients who underwent FDL transfer and MDCO for stage
II dysfunction had a mean 23% atrophy of the PT muscle and mean 27% hypertrophy
of the FDL muscle, as determined by magnetic resonance imaging, compared with the
unaffected contralateral side. In patients in whom the PTT was retained, the FDL muscle hypertrophied 11% at 14 months; however, in patients in whom the PTT was
excised, the FDL hypertrophied 4 times more to 44% of the contralateral side. This
finding suggests that excising the PTT results in a more functional tendon transfer.
However, the American Orthopaedic Foot and Ankle Society (AOFAS) hindfoot scores
were not significantly different between the retained and excised groups.
Leaving a portion of the PTT intact is supported by the data presented earlier by
Silver and colleagues35 showing that the stronger PT tendon should be preserved if
possible to balance the foot against the PB and PL. Aronow36 recommends that if
the distal tendon appears healthy, a repair should be attempted or the tendon transfer
should be anastomosed with the distal stump.36 If the proximal tendon appears
healthy and a repair cannot be performed, then the tendon should be tenodesed to
the adjacent tendon being transferred.36
FDL TRANSFER
Rationale for FDL Transfer

Goldner and colleagues40 at Duke University were the first to describe FDL and FHL
transfer for talipes equinovalgus in 1974. They found that isolated plication of the
PTT was insufficient in alleviating pain and restoring the arch. They reported improved
results with FDL or FHL transfer in addition to imbrication of the spring ligament and
tendo-Achilles lengthening.
The FDL runs anteromedially to the proximal PTT and posterolaterally to the distal
PTT without any significant structures in between these two tendons. It can be
accessed easily through the same incision used to inspect and debride the PTT, if
necessary. Unlike the FHL, the FDL transfer does not require crossing the neurovascular bundle. In addition, as mentioned earlier, the muscle is in phase with the PTT and
has similar tendon excursion. Furthermore, if cut proximal to the master knot of Henry,
the FDL is partially preserved in many individuals because of FHL and quadratus plantae attachments on the most distal aspect of the tendon. The only disadvantage to an
FDL transfer is the weakness of this tendon compared with the antagonistic PB.35
FDL Transfer Surgical Technique

Palpation and routine marking of the bony landmarks are necessary in planning any
incision on the foot and ankle. This incision extends from just posterior to the medial
malleolus to a point just distal to the navicular tuberosity (Fig. 2). The flexor retinaculum is left in place and typically not divided. With tenotomy or mayo scissors, dissection is continued and the PTT sheath is identified just posterior to the medial malleolus.
The tendon sheath is then sharply divided and the tendon is inspected. Typical gross
pathologic findings of the PTT include tenosynovitis, yellowing, thickening, incontinuity or longitudinal tears, and at times a complete tendon rupture (Fig. 3).1,37,38 The
tendon is either debrided and repaired or excised based on the surgeons preference,
as discussed earlier. After this is achieved, the toes are flexed and extended to aid in
identifying the FDL posterolateral to the distal PTT tendon. The FDL sheath is also
divided sharply, exposing the tendon (Fig. 4). The ankle is plantarflexed, the hindfoot
is supinated, and the lesser toes are flexed to provide as much FDL length as possible

Tendon Transfers in the Adult Flatfoot

Fig. 2. Typical medial incision for FDL tendon transfer. (Courtesy of Jeffrey E. Johnson, MD,
Chesterfield, MO.)

Fig. 3. Ruptured PTT. Note that the flexor retinaculum is left intact posterior to the medial
malleolus. (Courtesy of Jeffrey E. Johnson, MD, Chesterfield, MO.)

Fig. 4. FDL tendon identified posterior to the medial malleolus after PTT debridement.
The tendon was then dissected distally toward the knot of Henry. (Courtesy of Jeffrey E.
Johnson, MD, Chesterfield, MO.)

35

36

Backus & McCormick

during harvest. The tendon is then dissected out distally, elevating the abductor
tendon from the plantar-medial aspect of the foot and carefully coagulating the venous
plexus that is frequently encountered. Care must also be taken to preserve the medial
plantar branch of the tibial nerve, because this structure often lies superficial to the
distal FDL. The FDL tendon is then divided sharply just proximal to the master knot
of Henry. We think that it is important to take as much tendon as possible to ensure
adequate length for transfer. Excess tendon can be trimmed after tensioning. The
distal end of the harvested tendon is then controlled with a grasping suture, which
aids in manipulating and tensioning the tendon in transfer (Fig. 5).
In the event of a ruptured and proximally retracted PTT, the initial incision can be
curved proximally up the lower leg, or a separate incision can be made at the myotendinous junction of the PTT and FDL. The proximal PTT stump is debrided and, if an
adequate amount appears healthy and the muscle is functional, it can be tenodesed
to the proximal FDL tendon, which can be achieved by applying physiologic tension
and using a side-to-side repair with braided nonabsorbable suture or a Pulvertaft
weave. This step can be performed before or after FDL tendon harvest.
Next, a decision must be made with the portion of FDL that remained in the foot at
the point of transection near the knot of Henry. Many clinicians think that losing lesser
toe distal interphalangeal strength is not functionally limiting in most patients, because
there is some compensation by the quadratus plantae insertion on the distal FDL.
Several studies have also shown that there are distal interconnections between the
FHL and FDL tendons near the master knot of Henry in most patients.4045 Nevertheless, the lesser toes have intact metatarsophalangeal and proximal interphalangeal
flexion from the flexor digitorum brevis and the intrinsic muscles of the foot. However,
other clinicians think that the distal FDL stump should be tenodesed to the intact FHL
(or distal FHL to FDL in an FHL transfer46,47) to preserve toe plantarflexion strength.48
At this point, the clinician needs to decide where to transfer the FDL tendon. It can
be transferred to the intact distal PTT, the ruptured distal PTT stump, or the navicular
tuberosity. If the distal PTT is intact, performing a Pulvertaft weave or side-to-side
repair recreates the normal biomechanical function of the PT by acting on all of its
insertion sites. If the distal PTT is too diseased, or the surgeon prefers to excise it
for the reasons mentioned earlier, the FDL can be transferred to the navicular or medial
cuneiform. The navicular tuberosity is often chosen, because it requires less FDL

Fig. 5. The black arrow points to the FDL tendon that has been harvested and has a grasping
suture in place to aid in transfer. The red arrow points to the drill preparing to create a tunnel through the navicular for FDL transfer. Note that the guide pin exits the plantar aspect
of the navicular. (Courtesy of Jeffrey E. Johnson, MD, Chesterfield, MO.)

Tendon Transfers in the Adult Flatfoot

tendon length and inserting the tendon here is thought to provide more powerful hindfoot inversion and forefoot adduction than when inserted into the medial cuneiform.
This concept was postulated by Mann,48 who thought that inserting the tendon on
the more medial navicular rather than the medial cuneiform provided an increased
moment arm on the subtalar joint axis. This concept was later corroborated by Hui
and colleagues49 in a cadaveric study that reported a 46% decreased moment arm
of the FDL when transferred to the navicular as opposed to a 56% decrease when
transferred to the medial cuneiform. However, other clinicians argue that transferring
the FDL to the medial cuneiform buttresses the plantar medial aspect of the naviculocuneiform joint and prevents the lateral sag often seen with patients who have AAFD.14
We prefer to pass the FDL through a bone tunnel in the navicular tuberosity. When
making the tunnel, care must be taken to avoid violating any articular surface as well
as to maintain an appropriate thickness to the medial wall to avoid breaking the tunnel
when the tendon is transferred. The tunnel can be made with a cannulated drill bit over
a guide pin to ensure that the trajectory of the tunnel is satisfactory by verifying the
location of the guide pin before drilling (see Fig. 5; Fig. 6). Once the tunnel is created,
the FDL is passed from plantar to dorsal. It is tensioned and secured with suture anchors,50,51 interference screws,5255 or by sewing it back on itself (Fig. 7). The tendon
is transferred and tensioned as the last step in the flatfoot reconstruction after all other
bone and soft tissue reconstruction has been completed.
No consensus exists on a technique for achieving ideal tension of the tendon transfer. Hansen56 placed the foot in slight flexion with hindfoot inversion while tightening
the FDL 5 to 10 mm into the medial cuneiform. He then sutured it over the tibialis anterior if possible or the dorsal medial cuneiform capsule. Myerson57 reported that his
team secured the tendon while positioning the foot beyond midline and in inversion.

Fig. 6. Intraoperative fluoroscopic image of guide pin (arrow) placed in the navicular tuberosity. The drill is used to create a bone tunnel over this guide pin.

37

38

Backus & McCormick

Fig. 7. The FDL tendon has been transferred through the navicular, passed from plantar to
dorsal through a bone tunnel (black arrow), and then sutured back to itself (red arrow). The
FDL can also be secured to the periosteum at the dorsal and plantar aspects of the navicular.
(Courtesy of Jeffrey E. Johnson, MD, Chesterfield, MO.)

He warned that overtightening the FDL and causing it to sublux out of its groove was
possible; however, heavy patients and those with a extensive deformity should have
tighter repairs. He also argued that the tendon would loosen over time and that the
foot would eventually achieve a more neutral position. Haddad and Mann58 more
recently recommend securing the FDL at a midpoint between where the tendon lies
in the navicular tunnel in its most relaxed state and at its most tensioned state while
adducting the forefoot and inverting the subtalar joint. In conclusion, most investigators recommend securing the tendon with the hindfoot in inversion along with some
plantarflexion of the foot. The ideal tension on the tendon seems to be a matter of
the personal preference and clinical experience of the surgeon.
FDL Transfer Results

In general, the reports for FDL transfer are favorable in treatment of AAFD; however, it
is difficult to isolate the success of the tendon transfer in the setting of so many other
concomitant procedures. In an effort to evaluate the effectiveness of FDL transfer, articles that primarily report the outcomes of soft tissue procedures are reviewed here.
These reports can generally be broken down into 2 groups: (1) FDL transfer to an intact
PTT tendon or distal stump, and (2) FDL transfer to the navicular bone.
FDL transfer to the native distal PTT

In 1982, Jahss59 described a series of 6 patients who underwent a side-to-side FDL


tenodesis proximally and distally to a ruptured PTT. He reported excellent results,
with restoration of 20% to 25% of inversion power. In 5 patients in whom a direct
repair of a PTT rupture could not be performed, Funk and colleagues1 described a
similar side-to-side proximal and distal FDL transfer to bridge the gap following
debridement. At 24 to 32 months follow-up, all 5 patients could perform a singlelimb heel rise and had improvements in subjective pain with weight bearing. Nevertheless, 3 patients continued to complain of mild swelling in the medial hindfoot. Feldman
and colleagues60 described a nearly identical procedure in 11 patients who had stage
2 PTT dysfunction at an average 34 months follow-up. The mean AOFAS hindfoot
score improved from 38.8 to 78.1, and 6 patients reported 90% or greater pain relief.
Shereff61 reported a series of 17 patients with stage II PTT dysfunction who had
debridement and end-to-end repair of the PTT tendon followed by a side-to-side
FDL tenodesis. The distal FDL was left intact on its insertion. At 27 months average

Tendon Transfers in the Adult Flatfoot

follow-up, all patients were able to perform a single-limb heel rise, had 5 out of 5 inversion strength, and all but 1 reported unlimited ambulation. Ten of 17 patients had an
arch without weight bearing and 7 had an arch with weight bearing. The AOFAS hindfoot score was improved from an average of 66 before surgery to an average of 96 after surgery and, of the 13 patients who did not have a concurrent subtalar fusion, 11
patients had scores between 80 and 100. The other 2 patients reported fair and poor
scores.
However, not all reports are supportive of FDL tenodesis to the native PTT tendon.
In 1992, Conti and colleagues62 noted that 6 of 20 patients who underwent isolated
PTT debridement and side-to-side anastomosis of the FDL failed at an average of
nearly 15 months. These patients required a triple arthrodesis to relieve their symptoms and restore function. To our knowledge, this is the most negative report of
FDL tendon transfers.
FDL transfer to the navicular

Mann and Thompson63 described a series of 14 patients who underwent an FDL transfer to the navicular and 3 patients with PTT advancement for AAFD. At 2 to 5 years
follow-up, 12 patients had complete pain relief and excellent strength, 1 had complete
pain relief with mild weakness (which they attributed to an ankle fracture sustained by
the patient after surgery), 3 had some mild persistent pain and weakness, and 1 patient had continued pain despite a functional transfer. The flatfoot deformity was
only corrected in 4 patients and 7 had improved arches. The other 6 patients had either
no change or worsening of their deformity. Mann48 later reported that after applying
more stringent requirements on deformity flexibility he had less than 5% of more
than 100 FDL transfer cases require subsequent fusion at long-term follow-up. His
newer criteria consisted of at least 15 of subtalar motion, 10 of talonavicular adduction, and less than 12 of fixed forefoot varus. In contrast, his prior FDL transfer indications resulted in a nearly 90% subsequent fusion rate.
Gazdag and Cracchiolo4 reported the outcomes of 22 patients who had surgical
treatment of stage II PTT dysfunction at an average of 32 months. Eighteen of these patients had spring ligament disorders that required surgery and 17 of these patients
required FDL transfer to the navicular as well. Of these 18 patients, 14 had an excellent
result, 2 patients had a fair result, and 2 patients had a poor result. Three of the remaining 4 patients without spring ligament disorder also underwent FDL transfer. Two of
these patients had excellent results, 1 had a fair result, and 1 had a poor result. The investigators defined an excellent result as those patients who were pain free, able to
walk more than 6 blocks, and able to do a single heel rise. Patients with a fair result either
did not have complete pain relief or were unable to walk more than 6 blocks, but they
were still able to perform a single-limb heel rise. Of the 3 patients with poor results, 1 patient developed complex regional pain syndrome, 1 had continued medial hindfoot pain
and was unable to perform a single-limb heel rise, and the last patient had lateral hindfoot pain and an inability to perform a single-limb heel rise in addition to mild talonavicular and subtalar arthritic changes. Radiographic improvement was noted in the 19
patients with excellent and fair results in which the average lateral talofirst metatarsal
angle improved from an average of 13 to 9 . The patients with poor results had a
decrease in lateral talofirst metatarsal angle from an average of 13 to 20 .
FHL TRANSFER
Rationale for FHL Transfer

As mentioned earlier, the FHL transfer was also described by Goldner and colleagues40 in their series on treatment of talipes equinovalgus. In their discussion,

39

40

Backus & McCormick

they noted that the FHL is readily available through a similar incision when assessing
the PTT, and that it had a greater muscle mass and more robust tendon compared
with the FDL. They also thought that, if the tendon was left underneath the sustentaculum tali, it helped to maintain elevation of the sustentaculum tali and aid in talar
reduction following transfer. Silver and colleagues35 study validates what Goldner
and colleagues40 found clinically in relation to the tendon size and proportional muscle
strength. The FHL tendon is nearly 56% as strong as the tibialis posterior and is almost
twice as strong as the FDL; therefore, it provides greater force to counteract the PB.
Nevertheless, many clinicians choose not to use the FHL because it has to be
rerouted around the neurovascular bundle and there is concern for functional deficits
with decreased hallux plantarflexion strength. Several studies have found that transfer
of the FHL for other indications does weaken hallux interphalangeal flexion, but
AOFAS Hallux MTP-IP (metatarsophalangeal-interphalangeal) Scale scores remain
high64 and no functional deficits were observed.65,66 A recent cadaveric study by
Spratley and colleagues47 noted that both FHL and FDL transfer significantly reduced
hallux and lesser toe flexion strength; however, when the distal FHL was tenodesed to
the FDL, hallux flexion strength was restored almost to pretransfer levels. The FDL
transfer with tenodesis also resulted in an increased medial midfoot and heel plantar
force compared with FHL transfer and tenodesis. This finding suggests that the FHL
may be a more biomechanically appropriate tendon transfer in AAFD treatment; however, further clinical studies are needed to prove this assertion.
FHL Transfer Technique

Using the same medial incision as described earlier, the PTT is exposed and prepared
in a similar manner to the FDL transfer procedure. The FHL is identified near the master
knot of Henry by flexing and extending the great toe. With the ankle plantarflexed, the
hindfoot supinated, and the hallux interphalangeal joint flexed, the FHL tendon is
transected just proximal to the master knot of Henry. As described earlier, the distal
FHL tendon left at the knot of Henry is either left alone or tenodesed to the FDL tendon,
depending on surgeon preference. The FHL tendon is then identified above the flexor
retinaculum through an additional proximal incision or the medial incision that is
extended proximally. After identification, the tendon is pulled into the proximal aspect
of the wound, out of its sheath. A grasping suture is then placed into the distal aspect
of the tendon to ease in managing it through transfer. The FHL is then passed deep to
the neurovascular bundle, into the PTT sheath, and toward the distal aspect of the
wound (Fig. 8). A Hewson suture passer or tendon-passing instrument can be helpful
in helping to pull the tendon on this course. The FHL tendon is then inserted into the
intact distal PTT or ruptured distal PTT stump if it appears healthy. If not, it is inserted
into the navicular tuberosity in a similar manner to that described for the FDL tendon.
Furthermore, if the proximal PTT is healthy, it can be tenodesed to the proximal FHL in
the same manner as the FDL transfer. As with the FDL transfer, the FHL is transferred
and fixed with the foot in maximal plantarflexion and inversion, after all other bone and
soft tissue procedures in the flatfoot reconstruction have been completed.
FHL Transfer Results

Sammarco and Hockenberry46 published the largest series of FHL transfers with an
MDCO for stage II PTT dysfunction. Seventeen patients were assessed at an average
of 18 months follow-up. The average AOFAS hindfoot score increased from 62.4 to
83.6. Ten patients were satisfied, 6 patients were satisfied with minor reservations,
and 1 patient was dissatisfied. No significant improvement was noted clinically or
radiographically in the medial longitudinal arch.

Tendon Transfers in the Adult Flatfoot

Fig. 8. The FHL tendon is identified and transected at the master knot of Henry. A passing
suture is placed into the free end of the tendon. (A) The tendon is then brought into the
proximal wound at the level of the tibialis posterior myotendinous junction. At this level,
the FHL is posteromedial to the neurovascular bundle (protected with a clamp in figure).
(B) The FHL tendon is passed deep and lateral to the neurovascular bundle. It is then
brought distally through the PTT sheath, beneath the flexor retinaculum (see Fig. 9B). At
this point the tendon is passed through a tunnel in the navicular and secured to itself as previously described for the FDL tendon (see Fig. 7). (From Aronow MS. Tendon transfer options
in managing the adult flexible flatfoot. Foot Ankle Clin 2012;17:20526.)

Mulier and colleagues43 studied FHL harvest technique in 24 cadavers through a 2incision approach and noted 2 complete medial plantar nerve disruptions. They also
noted 1 partial disruption of both medial and lateral plantar nerves. Significant nerve
stretch was noted in 3 medial plantar nerves and 1 lateral plantar nerve. Therefore,
they found a significant risk to the medial and lateral plantar nerves with their FHL harvest technique. The clinical implications of these finding are difficult to interpret in this
cadaveric study. However, there is a paucity of literature involving FHL transfer to validate their conclusions.
PB TRANSFER
Rationale for PB Transfer

In 2001, Song and Deland67 described a supplemental transfer of the PB to the FDL
transfer in addition to an MDCO in stage II PTT dysfunction. The senior author added
the PB transfer to the FDL tendon transfer in cases in which the FDL tendon was

41

42

Backus & McCormick

particularly small or had been used previously. They thought that, in addition to
removing the principal antagonist to the PTT, the PB tendon would provide additional
inversion strength to maintain the medial longitudinal arch. In this article, the PB was
transferred in less than 10% of all their patients with AAFD.
PB to FDL Transfer Technique

Using a similar incision to that described for FHL and FDL transfers, the diseased
distal PTT is excised after the FDL is tenodesed proximally. The FDL is harvested
as described earlier and is transferred into the navicular. An incision is then made at
the junction of the glabrous skin near the base of the 5th metatarsal to isolate the
PB. It is released from its insertion on the base of the 5th metatarsal and brought
out of the skin at a second, more proximal lateral incision above the ankle. A grasping
stitch is placed in the end of the tendon to help manipulate the tendon in transfer
(Fig. 9A). The PB tendon is then passed using a tendon-passing instrument behind
the tibia into the proximal medial wound where the proximal FDL to PTT tenodesis
was performed (see Fig. 9B). The PB tendon is then passed down the PTT sheath
and into the same bony navicular tunnel as the FDL. With the foot held in inversion
and plantar flexion, both tendons are secured in the bony tunnel using one of the
methods described earlier. As with the other transfers, this is performed as the last
step in the flatfoot reconstruction, after all other bone and soft tissue procedures
have been completed.
PB to FDL Transfer Results

In Song and Delands67 series of 13 patients with 20 months average follow-up, significant weakness was noted in both hindfoot inversion and eversion compared with
the contralateral limb; however, only 1 patient had less than 4 out of 5 inversion
strength. Compared retrospectively with matched patients who only underwent FDL
transfer and an MDCO, no statistical differences were shown in AOFAS hindfoot
scores, inversion strength, eversion strength, or range of motion. They concluded
that the additional PB tendon transfer is a good option to supplement small FDL tendons or in revision surgery without sacrificing hindfoot eversion strength.
PB to PL Transfer Rationale

Although the PL tendon is antagonistic to the PTT as a powerful hindfoot evertor, it


also maintains the medial longitudinal arch by plantarflexing the first ray and supporting the plantar ligaments. Hansen68 described a PB to PL transfer for young patients
with peroneal spastic flatfoot in an effort to exploit the arch-supporting functions of the
PL. In a recent review by Aronow,36 he cited this technique as a possible means to
convert a deforming abduction force of the PB to an arch-correcting plantarflexion
force on the first metatarsal in patients with AAFD. He also surmised that proximal
lengthening of the PB might be beneficial in AAFD by weakening the primary antagonist to the PTT. The risks of these procedures are peroneal tendinopathy, hindfoot
eversion weakness, and chronic lateral ankle functional instability; however, these
risks have not been proved to occur.67 Aronow36 stated that he had not performed
any PB-lengthening procedures, but he has performed a PB tenotomy and gastrocnemius recession for a patient with stage III PTT dysfunction undergoing a triple arthrodesis.36 We have limited experience with PB tenotomy, lengthening, or transfer for
patients with AAFD, using it only as a technique to decrease abduction deformity in
patients undergoing triple arthrodesis for AAFD. However, no clinical series have
been performed to analyze the results of any of these procedures.

Tendon Transfers in the Adult Flatfoot

Fig. 9. The PB tendon is harvested through 2 lateral incisions. One is made just proximal to
the 5th metatarsal base where the tendon is transected. The second is made just posterior to
the fibula, at the level of the PTT myotendinous junction. (A) The PB tendon is brought proximally through this second incision. A grasping suture is placed through the distal end of the
tendon to ease transfer. Using a clamp to aid in tendon transfer, the PB tendon is passed
through the proximal incisions from lateral to medial. The transfer is made posterior to
the tibia but the PB tendon is kept deep to the PL on the lateral side (A), and deep to the
medial neurovascular bundle and FHL on the medial side (B). The PB tendon is then passed
through the PTT sheath into the distal wound and prepared for transfer in a manner previously described for FDL transfer. (B) The arrow shows the FHL tendon that was previously
transferred in this cadaveric specimen (see Fig. 8). The arrowhead shows a completed FDL
transfer that was performed in this cadaveric specimen for demonstration. (From Aronow
MS. Tendon transfer options in managing the adult flexible flatfoot. Foot Ankle Clin
2012;17:20526.)

43

44

Backus & McCormick

PB to PL Technique

As described by Hansen,68 the PB is transected at the level of the distal calcaneus with
the same incision that is used for a lateral column-lengthening procedure. The tendon
is allowed to retract 1 to 2 cm and the peroneal sheath and paratenon are removed
from the tendon. The PB is tenodesed under minimal tension to the PL by passing it
from superficial to deep through a slit in the PL tendon, fastening the tendons to
one another with a nonabsorbable suture. As an alternative, the PB may be transected
proximal to the superior peroneal retinaculum and tenodesed to the PL under physiologic tension with the foot in a neutral plantigrade position.
SUMMARY

AAFD describes a condition of progressive hindfoot valgus, forefoot abduction, and


forefoot varus. It is most commonly caused by PTT dysfunction.1 Patients who have
AAFD often complain of posteromedial hindfoot pain, a progressive change in the
shape of the foot, difficulty with weight bearing, and gait abnormalities. Conservative
management consists of nonsteroidal antiinflammatory medications, selective corticosteroid injections, and physical therapy. Patients also benefit from temporary immobilization in a cast or boot, custom foot orthosis, or a custom AFO or Arizona brace.
In stage I and II PTT dysfunction, in which conservative management has failed,
many clinicians think that the deformity can be corrected with PTT debridement,
tendon transfer, and an MDCO.7,1923 Spring ligament repair may also be indicated
in certain cases.4,28,29 In more significant stage II deformities, subtalar arthroereisis,16
an additional lateral column-lengthening osteotomy, and selective fusions may be
required.19,2326 Few data exist to effectively compare transfers to a retained PTT or
distal PTT stump versus transfer into the navicular. The tendon harvested for transfer
is based on surgeon preference. FDL1,4,40,48,5963 and FHL40,43,46 transfers have been
shown to have good results in AAFD, particularly for stage II PTT dysfunction. There
are no comparative studies to differentiate these two procedures; however, the FDL
transfer is more commonly used. PB may also be beneficial in supplementing an
FDL or FHL transfer in revision cases or patients with small donor tendons.67 In theory,
transfer of the PB to the PL may be beneficial in deformity correction.36,68 No studies
have confirmed the efficacy of this procedure.
At present, most foot and ankle surgeons do not advocate isolated tendon transfers
in patients who need correction of AAFD. However, Aronow36 argued that lowdemand elderly patients and young patients with mild to moderate stage II deformities
often do well with PTT debridement, spring ligament imbrication or repair, tendon
transfer, and a gastrocnemius recession (if necessary) despite incomplete correction
of the medial longitudinal arch. He states that the soft tissue procedures can be protected by use of a custom-made supportive orthosis, and, if a limited soft tissue procedure fails, he argues that osteotomies and fusions can be performed at a later date.
Although this is reasonable, we caution against isolated soft tissue procedures for
correction of AAFD because of the risk of failure and need for further operations.19,20
In the case of a stage II flexible flatfoot deformity, our preference is to excise the
diseased PTT tendon and use an FDL transfer through a tunnel in the navicular tuberosity. The FDL tendon is typically long enough to pass through the tunnel from plantar
to dorsal and be secured back to itself with nonabsorbable suture. The transferred
tendon is also sutured to the periosteum of the navicular with nonabsorbable suture
at both the plantar and dorsal aspects of the bone tunnel to further secure the tendon.
An MDCO is performed in all cases to correct hindfoot valgus deformity. A lateral
column-lengthening osteotomy is added when the foot has a significant abduction

Tendon Transfers in the Adult Flatfoot

deformity. A Cotton osteotomy may be required if forefoot varus remains after correcting the hindfoot through calcaneal osteotomies.
Patients with stage III and stage IV deformities typically require a hindfoot arthrodesis.27 A plantarflexion osteotomy or fusion of the first ray may also be required in patients with persistent forefoot varus.14,31,32 Furthermore, there may be a role for
tendon transfer in patients who undergo hindfoot arthrodesis. Mann and colleagues69
advocated the addition of an FDL transfer to subtalar arthrodesis to improve the function and deformity in patients with rigid deformities. They thought that the additional
tendon transfer would balance the deforming forces created by the PB. It is thought
that using an FHL or FDL tendon transfer in the setting of a triple arthrodesis may
augment the deformity correction in addition to counteracting the PB.
The decision to proceed with surgical correction of AAFD should be made after conservative treatment options have failed. Careful preoperative planning is necessary to
choose the appropriate reconstructive procedures given a patients clinical examination, deformity, and radiographic findings. Tendon transfers are critical to successful
surgical correction of flexible flatfoot deformity and may offer benefit to correction
of a more rigid or deformed flatfoot as well.
REFERENCES

1. Funk DA, Cass JR, Johnson KA. Acquired adult flat foot secondary to posterior
tibial-tendon pathology. J Bone Joint Surg Am 1986;68:95102.
2. Pedowitz WJ, Kovatis P. Flatfoot in the adult. J Am Acad Orthop Surg 1995;3:
293302.
3. Davis WH, Sobel M, DiCarlo EF, et al. Gross, histological, and microvascular
anatomy and biomechanical testing of the spring ligament complex. Foot Ankle
Int 1996;17:95102.
4. Gazdag AR, Cracchiolo A 3rd. Rupture of the posterior tibial tendon. Evaluation
of injury of the spring ligament and clinical assessment of tendon transfer and
ligament repair. J Bone Joint Surg Am 1997;79:67581.
5. Cohen BE, Ogden F. Medial column procedures in the acquired flatfoot deformity. Foot Ankle Clin 2007;12:28799, vi.
6. Malicky ES, Crary JL, Houghton MJ, et al. Talocalcaneal and subfibular impingement in symptomatic flatfoot in adults. J Bone Joint Surg Am 2002;84:20059.
7. Pinney SJ, Lin SS. Current concept review: acquired adult flatfoot deformity.
Foot Ankle Int 2006;27:6675.
8. Johnson KA, Strom DE. Tibialis posterior tendon dysfunction. Clin Orthop Relat
Res 1989;239:196206.
9. Myerson MS. Adult acquired flatfoot deformity: treatment of dysfunction of the
posterior tibial tendon. Instr Course Lect 1997;46:393405.
10. Alvarez RG, Marini A, Schmitt C, et al. Stage I and II posterior tibial tendon
dysfunction treated by a structured nonoperative management protocol: an
orthosis and exercise program. Foot Ankle Int 2006;27:28.
11. Myerson MS, Corrigan J, Thompson F, et al. Tendon transfer combined with
calcaneal osteotomy for treatment of posterior tibial tendon insufficiency: a
radiological investigation. Foot Ankle Int 1995;16:7128.
12. Sands A, Early J, Harrington RM, et al. Effect of variations in calcaneocuboid
fusion technique on kinematics of the normal hindfoot. Foot Ankle Int 1998;19:
1925.
13. Toolan BC, Sangeorzan BJ, Hansen ST Jr. Complex reconstruction for the treatment of dorsolateral peritalar subluxation of the foot. Early results after

45

46

Backus & McCormick

14.

15.

16.
17.

18.
19.
20.
21.
22.

23.

24.
25.
26.
27.
28.
29.
30.
31.
32.

33.
34.

distraction arthrodesis of the calcaneocuboid joint in conjunction with stabilization of, and transfer of the flexor digitorum longus tendon to, the midfoot to treat
acquired pes planovalgus in adults. J Bone Joint Surg Am 1999;81:154560.
Hirose CB, Johnson JE. Plantarflexion opening wedge medial cuneiform osteotomy for correction of fixed forefoot varus associated with flatfoot deformity. Foot
Ankle Int 2004;25:56874.
Hiller L, Pinney SJ. Surgical treatment of acquired flatfoot deformity: what is the
state of practice among academic foot and ankle surgeons in 2002? Foot Ankle
Int 2003;24:7015.
Stephens HM, Walling AK, Solmen JD, et al. Subtalar repositional arthrodesis for
adult acquired flatfoot. Clin Orthop Relat Res 1999;365:6973.
Habbu R, Holthusen SM, Anderson JG, et al. Operative correction of arch
collapse with forefoot deformity: a retrospective analysis of outcomes. Foot
Ankle Int 2011;32:76473.
Needleman RL. A surgical approach for flexible flatfeet in adults including a subtalar arthroereisis with the MBA sinus tarsi implant. Foot Ankle Int 2006;27:918.
Mosier-LaClair S, Pomeroy G, Manoli A 2nd. Operative treatment of the difficult
stage 2 adult acquired flatfoot deformity. Foot Ankle Clin 2001;6:95119.
Den Hartog BD. Flexor digitorum longus transfer with medial displacement
calcaneal osteotomy. Biomechanical rationale. Foot Ankle Clin 2001;6:6776, vi.
Deland JT. Adult-acquired flatfoot deformity. J Am Acad Orthop Surg 2008;16:
399406.
Myerson MS, Badekas A, Schon LC. Treatment of stage II posterior tibial tendon
deficiency with flexor digitorum longus tendon transfer and calcaneal osteotomy. Foot Ankle Int 2004;25:44550.
Guyton GP, Jeng C, Krieger LE, et al. Flexor digitorum longus transfer and
medial displacement calcaneal osteotomy for posterior tibial tendon dysfunction: a middle-term clinical follow-up. Foot Ankle Int 2001;22:62732.
Neufeld SK, Myerson MS. Complications of surgical treatments for adult flatfoot
deformities. Foot Ankle Clin 2001;6:17991.
Fortin PT. Posterior tibial tendon insufficiency. Isolated fusion of the talonavicular
joint. Foot Ankle Clin 2001;6:13751, viiviii.
Taylor R, Sammarco VJ. Minimizing the role of fusion in the rigid flatfoot. Foot
Ankle Clin 2012;17:33749.
Bennett GL, Graham CE, Mauldin DM. Triple arthrodesis in adults. Foot Ankle
1991;12:13843.
Jeng CL, Bluman EM, Myerson MS. Minimally invasive deltoid ligament reconstruction for stage IV flatfoot deformity. Foot Ankle Int 2011;32:2130.
Williams BR, Ellis SJ, Deyer TW, et al. Reconstruction of the spring ligament using a peroneus longus autograft tendon transfer. Foot Ankle Int 2010;31:56777.
Orr JD, Nunley JA 2nd. Isolated spring ligament failure as a cause of adultacquired flatfoot deformity. Foot Ankle Int 2013;34:81823.
Cotton F. Foot statics and surgery. N Engl J Med 1936;214:35362.
Ling JS, Ross KA, Hannon CP, et al. A plantar closing wedge osteotomy of the
medial cuneiform for residual forefoot supination in flatfoot reconstruction. Foot
Ankle Int 2013;34(9):12216.
Van Boerum DH, Sangeorzan BJ. Biomechanics and pathophysiology of flat
foot. Foot Ankle Clin 2003;8:41930.
Bluman EM, Dowd T. The basics and science of tendon transfers. Foot Ankle
Clin 2011;16:38599.

Tendon Transfers in the Adult Flatfoot

35. Silver RL, de la Garza J, Rang M. The myth of muscle balance. A study of relative strengths and excursions of normal muscles about the foot and ankle.
J Bone Joint Surg Br 1985;67:4327.
36. Aronow MS. Tendon transfer options in managing the adult flexible flatfoot. Foot
Ankle Clin 2012;17:20526, vii.
37. Trevino S, Gould N, Korson R. Surgical treatment of stenosing tenosynovitis at
the ankle. Foot Ankle 1981;2:3745.
38. Mosier SM, Lucas DR, Pomeroy G, et al. Pathology of the posterior tibial tendon
in posterior tibial tendon insufficiency. Foot Ankle Int 1998;19:5204.
39. Rosenfeld PF, Dick J, Saxby TS. The response of the flexor digitorum longus and
posterior tibial muscles to tendon transfer and calcaneal osteotomy for stage II
posterior tibial tendon dysfunction. Foot Ankle Int 2005;26:6714.
40. Goldner JL, Keats PK, Bassett FH 3rd, et al. Progressive talipes equinovalgus
due to trauma or degeneration of the posterior tibial tendon and medial plantar
ligaments. Orthop Clin North Am 1974;5:3951.
41. Oddy MJ, Flowers MJ, Davies MB. Flexor digitorum longus tendon exposure for
flatfoot reconstruction: a comparison of two methods in a cadaveric model. Foot
Ankle Surg 2010;16:8790.
42. LaRue BG, Anctil EP. Distal anatomical relationship of the flexor hallucis longus
and flexor digitorum longus tendons. Foot Ankle Int 2006;27:52832.
43. Mulier T, Rummens E, Dereymaeker G. Risk of neurovascular injuries in flexor
hallucis longus tendon transfers: an anatomic cadaver study. Foot Ankle Int
2007;28:9105.
44. OSullivan E, Carare-Nnadi R, Greenslade J, et al. Clinical significance of variations in the interconnections between flexor digitorum longus and flexor hallucis
longus in the region of the knot of Henry. Clin Anat 2005;18:1215.
45. Wapner KL, Hecht PJ, Shea JR, et al. Anatomy of second muscular layer of the
foot: considerations for tendon selection in transfer for Achilles and posterior
tibial tendon reconstruction. Foot Ankle Int 1994;15:4203.
46. Sammarco GJ, Hockenbury RT. Treatment of stage II posterior tibial tendon
dysfunction with flexor hallucis longus transfer and medial displacement calcaneal osteotomy. Foot Ankle Int 2001;22:30512.
47. Spratley EM, Arnold JM, Owen JR, et al. Plantar forces in flexor hallucis longus
versus flexor digitorum longus transfer in adult acquired flatfoot deformity. Foot
Ankle Int 2013;34(9):128693.
48. Mann RA. Posterior tibial tendon dysfunction. Treatment by flexor digitorum longus transfer. Foot Ankle Clin 2001;6:7787, vi.
49. Hui HE, Beals TC, Brown NA. Influence of tendon transfer site on moment arms
of the flexor digitorum longus muscle. Foot Ankle Int 2007;28:4417.
50. Myerson MS, Cohen I, Uribe J. An easy way of tensioning and securing a tendon
to bone. Foot Ankle Int 2002;23:7535.
51. Sullivan RJ, Gladwell HA, Aronow MS, et al. An in vitro study comparing the use
of suture anchors and drill hole fixation for flexor digitorum longus transfer to the
navicular. Foot Ankle Int 2006;27:3636.
52. Harris NJ, Ven A, Lavalette D. Flexor digitorum longus transfer using an interference screw for stage 2 posterior tibial tendon dysfunction. Foot Ankle Int 2005;
26:7812.
53. Louden KW, Ambrose CG, Beaty SG, et al. Tendon transfer fixation in the foot
and ankle: a biomechanical study evaluating two sizes of pilot holes for bioabsorbable screws. Foot Ankle Int 2003;24:6772.

47

48

Backus & McCormick

54. Sabonghy EP, Wood RM, Ambrose CG, et al. Tendon transfer fixation: comparing a
tendon to tendon technique vs. bioabsorbable interference-fit screw fixation. Foot
Ankle Int 2003;24:2602.
55. Wukich DK, Rhim B, Lowery NJ, et al. Biotenodesis screw for fixation of FDL
transfer in the treatment of adult acquired flatfoot deformity. Foot Ankle Int
2008;29:7304.
56. Hansen ST. Transfer of the flexor digitorum communis to the first cuneiform. In:
Hansen ST, editor. Functional reconstruction of the foot and ankle. Philadelphia:
Lippincott Williams & Wilkins; 2000. p. 4303.
57. Myerson MS. Correction of flatfoot deformity in the adult. In: Myerson MS, editor.
Reconstructive foot and ankle surgery. Philadelphia: Elsevier Saunders; 2005.
p. 189215.
58. Haddad S, Mann RA. Flatfoot in adults. In: Coughlin M, Mann RA, Saltzman C,
editors. Surgery of the foot and ankle. St Louis (MO): Mosby; 2011. p. 100786.
59. Jahss MH. Spontaneous rupture of the tibialis posterior tendon: clinical findings,
tenographic studies, and a new technique of repair. Foot Ankle 1982;3:15866.
60. Feldman NJ, Oloff LM, Schulhofer SD. In situ tibialis posterior to flexor digitorum
longus tendon transfer for tibialis posterior tendon dysfunction: a simplified surgical approach with outcome of 11 patients. J Foot Ankle Surg 2001;40:27.
61. Shereff M. Adult flatfoot: posterior tibial tendon dysfunction. Foot Ankle Clin
1997;2:217361.
62. Conti S, Michelson J, Jahss M. Clinical significance of magnetic resonance imaging in preoperative planning for reconstruction of posterior tibial tendon ruptures. Foot Ankle 1992;13:20814.
63. Mann RA, Thompson FM. Rupture of the posterior tibial tendon causing flat foot.
Surgical treatment. J Bone Joint Surg Am 1985;67:55661.
64. Richardson DR, Willers J, Cohen BE, et al. Evaluation of the hallux morbidity of
single-incision flexor hallucis longus tendon transfer. Foot Ankle Int 2009;30:
62730.
65. Coull R, Flavin R, Stephens MM. Flexor hallucis longus tendon transfer: evaluation of postoperative morbidity. Foot Ankle Int 2003;24:9314.
66. Den Hartog BD. Flexor hallucis longus transfer for chronic Achilles tendonosis.
Foot Ankle Int 2003;24:2337.
67. Song SJ, Deland JT. Outcome following addition of peroneus brevis tendon
transfer to treatment of acquired posterior tibial tendon insufficiency. Foot Ankle
Int 2001;22:3014.
68. Hansen ST. Transfer of the peroneus brevis to the peroneus longus. In:
Hansen ST, editor. Functional reconstruction of the foot and ankle. Philadelphia:
Lippincott Williams & Wilkins; 2000. p. 43941.
69. Mann RA, Beaman DN, Horton GA. Isolated subtalar arthrodesis. Foot Ankle Int
1998;19:5119.

Foot Ankle Clin N Am


12 (2007) 251271

The Flexible Flatfoot in the Adult


Eric Giza, MDa,*, Gerard Cush, MDb,
Lew C. Schon, MDc
a

Santa Monica Orthopaedic Group, 1313 20th Street, Suite 150, Santa Monica,
CA 90404, USA
b
Department of Orthopaedics, Geisinger Medical Center, 100 N. Academy Boulevard,
Danville, PA 17821, USA
c
Department of Orthopaedics, Union Memorial Hospital, 3333 N. Calvert Street #400,
Baltimore, MD 21218, USA

The adult acquired atfoot deformity is characterized by attening of


the medial longitudinal arch with insuciency of the supporting posteromedial soft tissue structures of the ankle and hindfoot [1]. Although the
etiology of this deformity can be arthritic or traumatic in nature [15],
it is most commonly associated with posterior tibial tendon dysfunction
(PTTD) [2]. By one estimate, PTTD aects approximately 5 million people
in the United States [6]. The clinical presentation of adult atfoot can
range from a exible deformity with normal joint integrity to a rigid,
arthritic foot.
Johnson and Strom [7] described three stages of PTTD, with an additional stage added by Myerson [8]. Stage I consists of painful synovitis of
the tendon. Nevertheless, tendon length and function are maintained so
there is no deformity. With stage II disease, there is progressive tendon dysfunction and a exible atfoot deformity develops. Stage III involves a rigid
deformity with stiness and often arthritis of the midfoot and hindfoot. Finally, stage IV consists of tibiotalar valgus, usually with associated arthritic
change [1].
The prevalence and increased awareness of PTTD has sparked a recent
trend toward surgical interventions that involve joint preservation techniques [6,9]. This article focuses on the anatomy, diagnosis, and current
treatments of the exible atfoot deformity.

One author has a financial interest in the Prostop arthroereisis device (Arthrex Inc.,
Naples, FL).
* Corresponding author.
E-mail address: egiza@smog-ortho.net (E. Giza).
1083-7515/07/$ - see front matter 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.fcl.2007.03.008
foot.theclinics.com

252

GIZA

et al

Anatomy and biomechanics


The relevant anatomy of the exible atfoot deformity not only includes
the posterior tibial tendon (PTT), but also the spring ligament, deltoid ligament complex, and the articular relationship of the talonavicular and subtalar
joints. The PTT has its origin on the posterior tibia, bula, and interosseous
membrane (Fig. 1). It courses immediately posterior to the medial malleolus
and inserts on the medial midfoot [1,10]. In an 11-specimen cadaver dissection, Bloome and colleagues [11] demonstrated that the PTT has three consistent bands: anterior, middle, and posterior. The anterior band represents
65% of the tendon width and inserts on the navicular tuberosity, inferior naviculocunieform joint capsule, and the inferior medial cuneiform. The posterior band comprises 20% of the tendon width, originates 23 mm proximal to
the anterior band, and helps form the acetabulum pedis. The middle band
represents approximately 15% of the PTT and is the tarsometatarsal extension with insertions to the middle/lateral cuneiforms, cuboid, and second to
fourth metatarsals. The investigators also found variations of insertions to
the spring ligament (4 of 11), fth metatarsal base (7 of 11), exor hallucis brevis (9 of 11), peroneus longus (4 of 11), and abductor hallucis (5 of 11) [11].
The vascular supply of the PTT consists of branches of the posterior tibial artery. Proximal to the medial malleolus, the PTT has vessels in the synovial sheath, which arise from muscle branches. Distally, the insertion has
a periosteal blood supply. In between, there is a zone of relative hypovascularity, which often corresponds to the site of attritional rupture and extends
approximately 1415 mm distal to the medial malleolus [12].
The spring ligament complex cradles the plantar medial aspect of the talar
head extending from the anterior margin of the sustentaculum tali to the plantar medial aspect of the navicular [1]. It is comprised of a superomedial

Fig. 1. Anatomy of the PTT. Arrow points to the posterior tibial tendon. (From Sarraan S.
Anatomy of the foot and ankle: descriptive, topographical. 2nd edition. Philadelphia: JB Lippincott; 1993. p. 217; with permission.)

THE FLEXIBLE FLATFOOT IN THE ADULT

253

calcaneonavicular portion that lies medial to the PTT and blends with the deltoid, which is wider and stronger than the inferior portion. During surgery,
this ligament is often found to be torn [13]. The inferior calcaneonavicular ligament is narrower and lies at the plantar most aspect of the acetabulum pedis
[14]. A third component, or articular portion, is located between the two bands
and articulates with the inferior aspect of the talar head.
The deltoid ligament complex has been described as having six separate
ligaments and deep supercial layers [15]. The deltoid prevents external rotation of the talus in the mortise and prohibits abduction of the ankle
[16,17]. The anterior portion of the deltoid has a conuence of bers with
the spring ligament complex [10,13,14].
Brodsky [18] stressed the importance of understanding the biomechanical
functions of the foot to comprehend the complex progressive deformity associated with adult acquired atfoot. In normal gait, the tibialis posterior
muscle produces inversion of the hindfoot. It is also an adductor of the forefoot at the midtarsal joint opposing the action of the peroneus brevis. The
posterior tibial muscle functions during the stance phase of gait. After
heel contact, the muscle acts as a shock absorber for the subtalar joint,
which limits hindfoot eversion by eccentric contraction [18].
During mid-stance, the posterior tibial muscle contraction causes subtalar
inversion, thus locking the transverse tarsal joints. This results in a rigid lever
for forward propulsion by allowing the powerful gastrocnemiussoleus complex to act through the foot at the metatarsal heads. During the propulsive
phase, the tibialis posterior muscle functions to accelerate subtalar joint supination and assist in heel lift. The shift of the body weight center from lateral to
medial through the longitudinal axis of the foot during the propulsive phase of
gait is achieved by the balanced activity of the tibialis posterior and peroneal
muscles. The posterior tibialis muscle then becomes inactive shortly after heel
lift, with the residual eect of acceleration during swing phase [18].
With PTTD, this balance is altered and weighted toward the peroneals
with greater hindfoot eversion and ligamentous tension and stretching during stance. In mid-stance there is decreased inversion of the subtalar complex, such that the hindfoot cannot function as a rigid lever [18]. The
gastrocnemiussoleus then acts at the talonavicular joint creating excessive
midfoot stress, which allows increased midfoot abduction and produces a levering action at the tarsometatarsal joints. In the propulsive phase, there is
less lateral column loading and a delayed heel lift that reduces the magnitude of propulsive activity [18]. The repetitive biomechanical alteration of
gait results in progressive midfoot collapse, forefoot abduction, and excessive hindfoot valgus [18].
Pathology
An acquired exible atfoot deformity is most often associated with
PTTD. Biomechanic overloading as described above can lead to chronic

254

GIZA

et al

microtrauma in the tendon. Microtears occur after 1500 to 2000 cycles per
hour and trigger an inammatory response [3]. With advancing age, the tendons elastic compliance decreases because of changes in collagen structure
[3], thus creating a pathologic sequence where tendon weakening results in
failure of the static stabilizers of the arch. Poor blood supply may initiate
this process or may prevent an adequate healing response, resulting in
chronic inammation, tenosynovitis, and tendinosis. Both Thordarson
and colleagues [19] and Deland and colleagues [20] were able to demonstrate
the relationship of medial soft tissue failure in cadaver models with sequential sectioning of the spring ligament complex and posterior medial
structures.
Deland and colleagues [14] reviewed MRI of patients with PTTD and
aged-matched controls. They evaluated the PTT, spring ligament complex,
talocalcaneal interosseous ligament, long and short plantar ligaments, plantar fascia, deltoid complex, plantar naviculocunieform ligament,and tarsometatarsal ligaments. The investigators found that ligament involvement
is extensive with PTTD, with the most signicant pathology in the superior and inferior medial calcaneonavicular ligaments and the interosseous
ligament [14].
Developmental etiologies also may be responsible for a exible atfoot
deformity. These include conditions associated with soft tissue laxity (Ehlers-Danlos and Marfan syndromes), accessory navicular, and neuromuscular diseases [11,21]. Extrinsic factors are less common but can result from
trauma involving the medial structures in an eversion type injury. Brodsky
and colleagues [4] reported two cases of traumatic rupture of the PTT in athletes, and Park and colleagues [5] described PTTD secondary to os subtibiale impingement.
Two potential mechanical causes of an acquired atfoot deformity include medial column instability and a contracture of the Achilles tendon
or gastrocnemius fascia. With the former, medial column instability results
in forefoot varus and a compensatory hindfoot valgus [1]. With the latter,
a tight Achilles tendon or gastrocnemius fascia results in transmission of
dorsiexion forces from the ankle to the transverse tarsal joint and midfoot.
This leads to midfoot collapse and hindfoot valgus with lateral peritalar subluxation of the navicular and subbular impingement [22,23].

Clinical examination
The diagnosis of the exible atfoot is relatively straightforward. However, the subtleties of the associated pathology require a systematic approach to examination and work-up. Patients usually complain of medial
ankle and hindfoot pain that radiates to the arch of the foot or proximally
to the leg. As the deformity progresses, there may be a complaint of lateral
or sinus tarsi pain caused by subbular impingement. Although some

THE FLEXIBLE FLATFOOT IN THE ADULT

255

patients will attribute a nonspecic traumatic event to the pain, most patients will relate a gradual onset of the pain with loss of the medial plantar
arch over recent months or years. A careful history is important and should
include previous trauma, steroid use, orthotic use, diabetes, smoking, and
family history of inammatory arthropathy. Body mass index may helpful
with patient counseling and surgical decision-making.
On physical examination, it is helpful to evaluate the patient in short
pants with both shoes o. This allows the clinician to note the alignment
of not only the foot and ankle, but also the knee. With genu valgus, an individuals center of gravity may be altered and more load may be placed on
the medial ankle and PTT. Comparison of tread wear on the shoes may reveal more posteromedial wear than the opposite side. On examination of the
standing patient from behind, the presence of hindfoot valgus can be noted
and measured, and the too many toes sign can be identied (Fig. 2). The
patient should be asked to perform a double leg heel rise so that the presence
or absence of hindfoot inversion can be identied. Next, the patient is asked
to perform a single leg heel rise on the aected side noting that inability to
do so is consistent with PTTD.
Examination sitting should include assessment of ankle and subtalar
range of motion. Ankle motion should be measured with the knee extended
and exed with the transverse tarsal joint locked and unlocked. This will allow the examiner to assess for Achilles tendon and gastrocnemius contractures. Palpation of the posteromedial ankle and hindfoot may reveal
tenderness, swelling, or fullness. The sinus tarsi, talar dome, and navicular
tuberosity should be palpated. Callus formation over the subluxated talar
head may be noted. For patients who have a exible atfoot, reduction of
the talonavicular joint and correction of the hindfoot valgus/forefoot abduction is possible. Lastly, the PTT strength is tested with resistance against the
inverted and plantarexed foot.

Fig. 2. Clinical photograph of a patient who exhibits the too many toes sign.

256

GIZA

et al

Diagnostic imaging
Radiographs should be obtained on the rst oce visit and should include weight-bearing views of the foot and ankle. Nonweight-bearing views
fail to provide valid information regarding the alignment of the foot [1,24].
The presence of joint space narrowing or varus/valgus deformity of the ankle should be noted. Selected measurements of the lateral and anteroposterior radiographs to measure the degree of the deformity have been shown in
one study to be inaccurate [25]; however, Saltzman and colleagues [26] demonstrated that with careful measurement techniques, reliability is possible.
Recently, Younger and colleagues [27] studied inter- and intraobserver
reliability in the evaluation of plain radiographs of patients with PTTD.
On lateral radiographs they measured talar-rst metatarsal angle (Mearys
angle), calcaneal pitch, talocalcaneal angle, medial column height, calcaneal-fth metatarsal height, and lateral column height. On anteroposterior
radiographs they measured talar-rst metatarsal angle, talonavicular uncoverage angle, and calcaneal-fth metatarsal angle. The axis of the talar-rst
metatarsal angle on the lateral radiograph was the most discriminating and
statistically signicant radiographic parameter in patients who had symptomatic atfoot. On the anteroposterior radiograph, the talar head uncoverage was the most statistically signicant value, despite low inter- and
intraobserver reliability [27].
Clinical examination and radiographs are usually sucient to establish
the diagnosis of PTTD. In certain instances, however, the use of MRI can
be helpful to conrm the diagnosis, evaluate the amount of pathology in
the PTT and spring ligament complex, and detect bone edema (Fig. 3)
[1,14,28]. Meanwhile, CT can be helpful in the evaluation of advanced lateral impingement and the detection of arthritic change [23]. When a exible
atfoot is associated with minor degenerative change (especially of the

Fig. 3. Axial (proton density weighted) MRI of an attenuated PTT. Arrow points to the posterior tibial tendon.

THE FLEXIBLE FLATFOOT IN THE ADULT

257

tarsometatarsal joints), selective injection with local anesthetic under uoroscopy can help dierentiate the amount of the patients discomfort that
is attributable to the degenerative change.
In some patients, the discomfort that is caused by the chronic inammation and tendinosis associated with PTTD cannot always be elucidated
clearly on examination or imaging. In these instances, a diagnostic injection
of local anesthetic into the PTT sheath can be helpful. The authors prefer to
use a 1:1 mixture of 1% lidocaine and 0.25% marcaine. After the skin is
sterilized, a 22-gauge needle is inserted parallel to the PTT approximately
12 mm above the medial malleolus and 23 mL of the anesthetic is injected.
Care is taken to ensure that the needle is not deep within the tendon. The
patient is then asked to ambulate for a few minutes after which they are
re-examined.
Cooper and Mizel [29] recently reviewed 17 ankles with stage I PTTD
that had a diagnostic injection and MRI of the PTT at dierent times.
They found that although 15 of 17 ankles (88%) had positive MRI ndings,
all 17 ankles (100%) had relief of pain from the diagnostic injection. The investigators concluded that diagnostic injection is an accurate, sensitive, and
safe diagnostic tool [29].
Nonoperative treatment
The stage and progression of the atfoot deformity will generally determine the degree and duration of the conservative treatment [30]. The initial
treatment of the adult exible atfoot deformity (stage II PTTD) focuses on
improving symptoms by decreasing the forces transmitted through the posteromedial hindfoot [1]. The patient should be encouraged to lose weight,
modify repetitive loading activities, and use supportive shoes [1,2]. Cast immobilization or a removable walking boot with an insert that supports the
longitudinal arch can be helpful with acute inammation. A stirrup brace
can be used and has been shown to assist in unloading the tendon by transferring the plantarexion component solely to the Achilles tendon [3].
Some patients may tolerate footwar modications, but extra deep shoes
that can house adequate orthosis are often necessary. Key elements include
long, rigid medial counters, soft leather uppers, high toe boxes, and shock
absorbing soles to decrease the ground reaction forces [30]. A moderate
rocker bottom will assist in toe o [30].
Many patients who have a exible atfoot deformity also may benet
from custom prescribed orthotics. Either a full-length or 34-length device
may be used. Most often a medial post and deep heel is used to control
the hindfoot. The more rigid University of California at Berkeley Laboratory (UCBL) orthosis is designed to correct the exible hindfoot and hold
it in a neutral position. It limits motion in the subtalar joint and helps prevent subbular impingement by exerting pressure on the lateral wall of the
calcaneus and the medial longitudinal arch [1,30]. Havenhill and colleagues

258

GIZA

et al

[31] studied contact pressures in atfoot cadaver specimens and then compared change with a UCBL versus a medializing calcaneal osteotomy
(MCO). The UCBL and the osteotomy signicantly reduced the tibiotalar
contact pressure and peak pressure. The investigators concluded that both
options can be used for clinical management of the exible atfoot. Although the UCBL can be quite helpful in correcting the exible atfoot deformity, some patients do not tolerate the device because of pressure over
the navicular and/or diculty tting it in shoes.
An ankle foot orthosis (AFO) can correct deformity by stabilizing the
medial to lateral movement of the hindfoot and limiting excursion of the
PTT by preventing plantar exion and pronation in the push-o stage of
gait [30]. A rigid, molded AFO can be useful during the initial stage of exible atfoot treatment, but it can lead to compliance issues for long-term
use.
Recently, the Airlift PTTD Brace (djOrtho, Inc., Carlsbad, California)
has been introduced for patients who have a exible atfoot deformity
[32]. Like other athletic ankle braces, the Airiftbrace has a low prole,
semirigid vertical shell that helps to control hindfoot valgus. In addition,
the Airlift PTTD Brace incorporates an inatable bladder underneath the
hindfoot and midfoot (Fig. 4). When inated, this bladder supports and elevates the medial longitudinal arch. Early anecdotal experience with this device has been encouraging with regard to compliance and symptomatic
relief.
Physical therapy also should be considered, because it serves to strengthen
the PTT and surrounding tendons, thereby providing a stronger structural
support for the foot and lessening the stress on the PTT [3]. Any physical

Fig. 4. The Airlift PTTD Brace. (Courtesy of djOrtho, Carlsbad, CA; with permission.)

THE FLEXIBLE FLATFOOT IN THE ADULT

259

therapy regimen also should address the presence of a gastrocnemius contracture. Cryotherapy may be considered also after activity, because it decreases the release of inammatory mediators. Iontophoresis with topical
steroidal anti-inammatory cream can be added to the treatment and has
not been shown to cause an increased risk in tendon rupture [3].
Alvarez and colleagues [33] recently assessed the ecacy of a focused
nonoperative treatment protocol for stage I and stage II PTTD. The investigators found only four reports [3437] of nonoperative management of
PTTD and designed a program that included the use of a short articulated
AFO or foot orthosis, high-repetition exercises, aggressive plantarexion activities, and an aggressive high-repetition home exercise program that included gastroc-soleus tendon lengthening. They studied a group of 47
patients (37 women and 10 men) over a three-year period. Patients who
were unable to perform a single leg heel raise or had pain for greater than
3 months were given an AFO, and the remainder of patients were given
a UCBL type orthosis. Patients underwent a four-phase intensive treatment
program for 4 months with a minimum of 10 physical therapy visits. Investigators found that all patients had signicant weakness of all ankle muscle
groups before therapy. After treatment, 83% of patients had successful subjective and functional outcomes, and 89% of patients were satised with
their results [33].

Operative treatment
Overview
Once nonoperative treatment measures have failed, operative intervention focuses on soft tissue balancing with tendon transfer in combination
with one or more osteotomies to return the foot to a more normal anatomic
position and restore normal foot biomechanics [38]. Before 1980, the adult
atfoot was routinely treated with a triple arthrodesis; thereafter, the operative focus has been on augmentation or substitution of the PTT [1]. Initial
results with isolated soft tissue procedures generally were not as successful
and often lead to failure caused by lack of bony realignment [1]. Currently,
some type of bony realignment is recommended to augment and protect the
soft tissue reconstruction.
Hiller and Pinney [9] recently conducted a large survey of foot and ankle
surgeons to assess the state of practice in the approach to surgical reconstruction of stage II PTTD. In this study, a sample operative was presented
to each surgeon. The investigators found that 97% of respondents reported
they would employ some type of bony procedure in their surgical treatment.
Eighty-eight percent described techniques that would preserve the subtalar
and talonavicular joints. These techniques included an MCO in 73% of patients, a lateral column lengthening in 41% of patients, and medial column
stabilization (rst tarsometatarsal and/or navicular cuneiform arthrodesis)

260

GIZA

et al

in 15% of patients. Twelve percent of respondents reported that they would


perform an arthrodesis on one or more of the hindfoot joints. Ninety-eight
percent of respondents reported that they would employ some type of soft
tissue procedure. Ninety-four percent of respondents would augment the
PTT, 53% would formally repair the spring ligament, and 70% would address a presumed equinus contracture. The investigators concluded that
most surgeons employed a combination of bony and soft tissue procedures
that preserved the subtalar and talonavicular joints [9].
Soft tissue augmentation is performed to restore or recreate the function
of the PTT. Whereas many surgeons prefer to transect the PTT to remove
the degenerative tissue, this may not always be necessary. Valderrabano
and colleagues [39] performed MRI analysis of the PTT and muscle in patients who underwent successful atfoot reconstruction. The investigators
found that even though fatty degeneration of the posterior tibial muscle
was present in all patients preoperatively, there was a decrease in degeneration with increasing strength of the posterior tibial muscle and muscular size
postoperatively. They also found that the recovery potential of the posterior
tibial muscle was signicant even after delayed repair of a diseased tendon.
They concluded that the PTT should not be transected because it precludes
recovery potential of the posterior tibial muscle [39].
Wacker and colleagues [28] studied the MRIs of 12 patients who had
PTTD and found that all patients had atrophy of the PTT muscle and compensatory hypertrophy of the exor digitorum longus (FDL) muscle. In the
three patients who had complete rupture of the PTT, fatty inltration occurred throughout the muscle. They suggested that in patients who have
an intact but diseased PTT, augmentation with preservation or tenodesis
of the PTT may benet the reconstruction [28].
The spring ligament (calcaneonavicular) complex often is attenuated and
contributes to the pathology of the adult exible atfoot. As such, reconstruction of the spring ligament is performed frequently to augment the medial soft tissue reconstruction. The attenuated complex can be imbricated
with suture or advanced and secured with small suture anchors in the navicular. Choi and colleagues [40] reported using peroneus longus tendon in a cadaver model to reconstruct the spring ligament complex. They tested three
dierent congurations and found that a superomedial/plantar passage of
the tendon through the calcaneus and navicular best corrected the talonavicular deformity [40].
Although it is commonly assumed that transfer of the FDL tendon has
no morbidity secondary to the intertendonous connections with the exor
hallucis longus (FHL) tendon, a recent cadaver study examined the distal
anatomic relationship of these structures [41]. The following three dierent
congurations were found: a tendon slip from FHL to FDL in 10 feet
(42%); a tendon slip from FHL to FDL and another slip from FDL to
FHL in 10 feet (42%); and no attachment between FHL and FDL in
four feet (16%) [41].

THE FLEXIBLE FLATFOOT IN THE ADULT

261

Flexor digitorum longus transfer


In 2001, Guyton and colleagues [42] performed a retrospective review of
26 patients who had a combined FDL transfer and an MCO. Two patients
experienced early failure of xation, and the remaining 24 patients were able
to perform a single leg heel rise within one year. The study found that patients continued to improve for up to 10 months after the procedure. It
also noted that while surgery improved pain and function, it did not appreciably improve the medial arch [42].
Wacker and colleagues [28] reported 35 year results on 44 patients who
underwent FDL transfer and an MCO. They reported persistent improvement in the American Orthopaedic Foot and Ankle Society (AOFAS) hindfoot score (from 48.8 to 88.5). Guyton and colleagues [42] reported similar
results; 43 of 44 patients rated their outcome with regard to pain and function as good or excellent. However, only 36 of 44 patients rated their alignment as good or excellent.
Myerson and colleagues [43] reported on a large series of 129 patients
with a mean of 5.2 years follow-up. These investigators found that the
procedure yielded excellent results with minimal complications; the mean
AOFAS hindfoot score was 79. They reported that patients experienced
improvement in symptoms for up to 14 months postoperatively, and that
patients who had lower preoperative levels of daily function had the greatest
benet from the procedure. The investigators surmised that the ideal patient
for a combined FDL transfer with an MCO is a patient who has a exible
atfoot, insignicant forefoot supination, and less than 30% uncovering of
the talar head on anteroposterior radiographs [43].
Brodsky [18] performed a prospective pre- and postoperative gait analysis
study on 12 patients with PTT reconstruction. Six of the patients had concomitant midfoot fusion that resulted in a signicant improvement in radiologic parameters. Although thesample size was limited, Brodsky found small
but statistically signicant increases in cadence, velocity, and maximum sagittal ankle joint power.
Sullivan and colleagues [44] recently described an alternative method for
xation of the FDL transfer. Using a cadaver model, they placed an anchor
in the proximal/plantar portion of the navicular tunnel and tensioned the
FDL by a passing suture through the distal/dorsal portion of the tunnel.
This technique was believed to entail a shorter incision and less dissection
in addition to obviating the need to disrupt the FDL distal to the knot of
Henry. The investigators found identical pullout strength compared with
the standard tendon to tendon repair with a navicular tunnel [44].
Surgical technique: exor digitorum longus transfer
After completion of the bony correction (MCO or LCL), a curvilinear incision is made from the posterior aspect of the medial malleolus to the medial

262

GIZA

et al

navicular tuberosity. The PTT sheath is opened to expose the tendon. Quite
often the tendon has longitudinal tears with areas of tendinosis or detachment
from its insertion on the navicular. If the tendon has minimal or no excursion,
it is resected at the most proximal aspect of the incision. In cases where PTT
has 1 cm or more of excursion, the tendon can be saved for tenodesis to the
FDL transfer.
A capsular incision between the talar head and navicular can be made at
this time to remove any redundant tissue. Then the FDL is harvested proximal to the knot of Henry. The FDL is brought proximally into the wound
and a grasping stitch placed in the tendon. A bone tunnel is made from the
anatomic insertion of the PTT dorsally and distally through the navicular.
This is performed with a 5.5-mm cannulated drill over a guide wire if necessary. A sucient bone bridge must be maintained to avoid fracture. Then the
FDL tendon is passed from plantar to dorsal through the bone tunnel. The
foot is plantar exed and inverted while the FDL is tensioned. A 4.5-mm
interference screw is placed to secure the FDL within the tunnel. Another
form of xation includes the use of heavy suture passed through the tendon
and into the periosteum on either side the tunnel.
The wound is closed in layers with the foot splinted in plantarexion and
inversion. At 2 weeks postoperatively, sutures are removed and the patient is
placed in a short leg cast. At 56 weeks postoperatively, a removable boot
brace is prescribed, and weight-bearing and motion are encouraged.

Medializing calcaneal osteotomy


An MCO is performed to correct the hindfoot valgus deformity and also
redirect the coronal vector of the Achilles tendon. Using pressure sensitive
lm, Steensmeier and colleagues [45] studied the eects of medial and lateral displacement on tibiotalar joint contact. They loaded cadaver specimens
in three positions: neutral, 1 cm of lateral displacement of the posterior fragment, and 1 cm of medial displacement of the fragment. For an applied load
equaling two times body weight, a 1-cm lateral displacement shifted the center of pressure an average of 1.06 mm laterally, whereas a 1-cm medial displacement shifted the center of pressure an average of 1.58 mm medially
[45]. Regional contact parameters changed in a reproducible and statistically
signicant manner. Among four equal sized, para-sagittally bounded cartilage zones, lateral displacements consistently unloaded the most medial zone
and increased loading of the most lateral zone; medial calcaneal displacements had the converse eect. The investigators concluded that translational
calcaneal osteotomies may be used clinically to partially ooad focal areas
of cartilage along the medial and lateral borders of the tibiotalar joint [45].
It was initially hypothesized that plantar fascia tightening occurs with
these procedures, helping to restore a more normal longitudinal arch. In
1998, Horton and colleagues [46] studied nine cadaver specimens using

THE FLEXIBLE FLATFOOT IN THE ADULT

263

a atfoot model. They measured fractional length changes in the plantar fascia after a medial displacement calcaneal osteotomy and after lateral column
lengthening through the calcaneocuboid joint. They reported that tightening
of the plantar fascia did not occur with either medial calcaneal displacement
or lateral column lengthening; moreover, the plantar fascia became signicantly less taut with both procedures. The investigators also found that lateral column lengthening produced signicantly looser plantar fascia than
did medial calcaneal displacement [46].
Hadeld and colleagues [6] performed a 1-cm MCO on 14 cadaver specimens and measured the eects on Achilles tendon lengthening and changes
in plantar foot pressures. They found a statistically signicant decrease in
average pressure over the rst and second metatarsal regions and a concomitant increase in pressure over the lateral aspect of the calcaneus and the lateral forefoot.
Greene and colleagues [47] studied the anatomic relationship of the medial neurovascular structures in relation to an MCO in 22 cadaver specimens. They performed an MCO in each specimen with subsequent
dissection medially to identify the medial plantar nerve (MPN), the lateral
plantar nerve (LPN), the posterior tibial artery (PTA), and their respective
branches. The investigators found that an average of four neurovascular
structures crossed each osteotomy site (range 26), most of which were
branches of the LPN or the PTA. The MPN did not cross in any of the specimens studied, the LPN crossed in one specimen, and the PTA crossed in
two specimens. The calcaneal branch of the LPN was identied and crossed
in 86% of the cadavers. A more distal second branch of the LPN (Baxters
nerve) was identied and crossed in 95% of the specimens. Each PTA distributed from zero to three branches that variably crossed the osteotomy
at a point from 2% to 100% along its length. The investigators recommended that completion of the osteotomy through the medial calcaneal cortex
be carefully performed in a controlled manner to reduce the risk of complications [47].
Surgical technique: medial calcaneal osteotomy
The patient is positioned lateral after a thigh tourniquet is placed. An oblique incision is made one nger breadth below the tip of the lateral malleolus. Sharp dissection can be performed down to the lateral calcaneal wall.
Still, care should be taken with the skin edges and with the sural nerve if encountered. The direction and level of the MCO is conrmed with uoroscopy. The osteotomy is performed with a sagittal saw and can be
completed with a thin osteotome. Then the calcaneal tuberosity is displaced
medially by approximately 1 cm. Distal translation also can be performed to
increase calcaneal pitch. The osteotomy is secured with a 6.5-mm or 7.3-mm
cannulated screw place from posterior to anterior (Fig. 5). Then the wound
is closed in layers.

264

GIZA

et al

Fig. 5. (A, B) Lateral and calcaneal radiographs of an MCO.

Lateral column lengthening


For more severe exible atfoot deformity with uncovering of the head of
the talus, the lateral column of the foot becomes relatively shortened with
respect to the medial column. The Evans procedure helps restore the medial
longitudinal arch with an opening wedge osteotomy of the anterior calcaneus centered 10 mm proximal to the calcaneocuboid joint [38]. Lateral column lengthening results in loosening of the plantar fascia and may tighten
the peroneus longus to increase plantar exion of the rst ray [2,38]. Lengthening of 812 mm is advised using a tricortical autograft or allograft wedge
[38]. Some reports have shown increased calcaneo-cuboid joint pressures associated with the osteotomy and an increased risk of arthritis in the joint
[38]. Distraction arthrodesis of the calcaneo-cuboid joint is another surgical
option that eliminates the concern for arthritis; however, this could lead to
future midfoot arthritis [38].
Surgical technique: lateral column lengthening (arthrodesis)
The patient is positioned in the lateral position and a thigh tourniquet
used. A longitudinal incision is made dorsally over the calcaneo-cuboid
joint. The sural nerve is protected, and dissection is carried down to the extensor digitorum brevis muscle belly. The extensor digitorum brevis muscle
is elevated o its proximal origin to expose the anterior process of the calcaneus and its articulation with the cuboid. A sagittal saw is used to remove
the articular surface of calcaneo-cuboid joint. One pin from a cervical distractor device is placed in the anterior process of the calcaneus and the other
in the cuboid. Then the distractor is used to correct the abduction under image intensication. Once the talar-rst metarsal axis is realigned, measurements are taken for the graft to be fashioned on the back table. A height
and depth of 18 mm allows for maximal bone ll without medial or dorsal
impingement. The only variable is the length of the graft. Tricortical allograft from the anterior process of the calcaneus or the femoral neck is the
made and impacted into place. With removal of the distractor, the

THE FLEXIBLE FLATFOOT IN THE ADULT

265

correction is checked and conrmed using uoroscopy (Fig. 6). A compression staple is used for solid xation of the lateral column. After completion,
soft tissue reconstruction can be performed.

Subtalar arthroereisis implant


For patients who have more severe exible atfoot deformities, lateral
column support of an FDL transfer/medial displacement calcaneal osteotomy may be necessary. There has been a recent increase in interest for the
subtalar arthroereisis procedure, because decreased mobility, lateral column
overload, and more extensive surgical exposure can be associated with a double calcaneal osteotomy. Arthroereisis involves the placement of an implant
(Fig. 7) into the sinus tarsi to correct forefoot abduction and prevents lateral
column shortening. By blocking nonphysiologic eversion, the implant decreases the tendency of the talus to rotate medially and plantarly [48]. The
implant prevents strain on the medial tissues during the healing phase after
a atfoot reconstruction. Once healing is completed, the role of the implant
is marginal, and it may be removed if it becomes symptomatic [48]. The implant is not recommended for use as an isolated procedure [49].
While Grice [50] initially described an extra-articular arthrodesis of the
subtalar joint with a sinus tarsi bone graft, Subotnick [51] rst described
the arthroereisis procedure using a silicone block in the sinus tarsi. The arthroereisis procedure was initially more common in the pediatric population, and it continues to be used in children who have symptomatic
atfeet. Giannini and colleagues [52] recently reported the results of the procedure in 21 children. These investigators found that in combination with
other indicated procedures (heel cord lengthening, modied Kidner), 95%

Fig. 6. (A) Intraoperative lateral radiograph of the distraction during an LCL procedure. (B)
Postoperative anteroposterior radiograph after an FDL transfer, MCO, and LCL. Arrow depicts bone tunnel in the navicular.

266

GIZA

et al

Fig. 7. The Prostop arthroereisis device. (Courtesy of Arthrex, Inc., Naples, FL; with
permission.)

of patients were asymptomatic at 4 years. Viladot and colleagues [53] performed arthroereisis in 21 adults in conjunction with PTT synovectomy or
FDL transfer. In this study, the average AOFAS hindfoot scores improved
from 47 to 82 at an average of 27 months.
Needleman [49] recently reported on 28 arthroereisis procedures combined with FDL transfer and MCO with an average 44-month follow-up.
He found signicant improvement of the AOFAS hindfoot score (52 preoperatively to 87 postoperatively). On a 10-point scale, the average patient satisfaction was 8.3 points, and 78% of patients said they would have the
surgery again. The implant was surgically removed in 11 feet (39%) because
of pain. Needleman [49] concluded that reconstructive foot and ankle surgery that included a subtalar arthroereisis resulted in favorable clinical outcomes despite the high incidence of temporary sinus tarsi pain until the
implant was removed.
Surgical technique: subtalar arthroereisis
VanAman and Schon [48] recently described the technique of subtalar arthroereisis. The surgical decision-making guidelines are described in Table 1.
The implant can be used to supplement atfoot deformities in patients who
are obese, have diabetes, use tobacco, and have noncompliance issues. Heel
cord lengthening is performed when necessary.
The investigators perform the surgery as three separate procedures in the
following order: MCO, subtalar arthroereisis, and FDL transfer. The overall foot deformity is assessed under anesthesia with attention given to the degree of hindfoot valgus and forefoot supination.
The goal is to correct the heel to a neutral position without causing forefoot supination. The sinus tarsi is palpated just inferior and distal to the tip
of bula and a 1 cm incision is created dorsal to the peroneal tendons over

THE FLEXIBLE FLATFOOT IN THE ADULT

267

Table 1
Indications for use of an arthroereisis implant
Preoperative guidelines for subtalar
arthroereisis implant
Isolated FDL transfer
with subtalar arthroereisis

FDL transfer, MCO, and


subtalar arthroereisis

Criteria
Able to invert past midline
Clinical heel valgus (!5 )
Minimal abduction deformity (20%30%
talonavicular uncoverage, talo-rst
metatarsal angle !20 on
anteroposterior radiograph)
Minimal arch collapse (talo-rst
metatarsal angle !10 on lateral
radiograph)
Unable to invert past midline
Moderate heel valgus (!15 )
Moderate abduction deformity
(30%40% talonavicular uncoverage,
talo-rst metatarsal angle 21 40
on anteroposterior radiograph)
Moderate arch collapse (talo-rst
metatarsal angle 11 20 on lateral
radiograph)

Adapted from VanAman S, Schon L. Subtalar arthroereisis as adjunct treatment for type II
posterior tibial tendon deciency. Techniques in Foot & Ankle Surgery 2006;5(2):11725; with
permission.

the soft tissue sulcus. Blunt dissection with a hemostat exposes the tarsal canal (Fig. 8), and routine release of the interosseous ligament is not recommended. Then a blunt guide wire is inserted from lateral to medial,
recognizing that the sinus tarsi is oriented obliquely from anterolateral to
posteromedial. Resistance may be felt as the guide wire penetrates the interosseus ligament. The guide wire is advanced until it tents the medial skin just
dorsal to the sustentaculum tali (Fig. 9). A relaxing puncture wound is

Fig. 8. Intraoperative photo depicting the incision and dissection for the arthroereisis implant.

268

GIZA

et al

Fig. 9. Intraoperative photo depicting correct placement of the guidewire for the arthroereisis
implant.

created to allow the wire to pass medially, where it is secured with a hemostat. The medial penetration point should fall within the planned FDL incision site [48].
With the guide wire in place, cannulated blunt trial sizers are advanced
from lateral to medial. Under uoroscopy, the trial implant typically crosses
half of the mediolateral diameter of the talar neck and should not be proud
of the lateral border. It is recommended that motion and position of the
hindfoot be assessed to ensure the implant is not too small (allowing too
much eversion) or too large (hindfoot overcorrected into varus). Then the
implant is inserted over the guide wire until it is ush with the lateral border
of the talar neck (Fig. 10) [48].

Fig. 10. Postoperative anteroposterior radiograph showing the arthroereisis implant (arrow)
with optimal placement.

THE FLEXIBLE FLATFOOT IN THE ADULT

269

Summary
The adult exible atfoot is commonly the result of PTTD, which ultimately leads to attenuation of the static stabilizers of the medial longitudinal
arch. Recent advances in physical therapy regimens and lightweight bracing
have improved nonoperative treatment. Past surgical approaches included
selected hindfoot and midfoot fusions; however, more recent surgical procedures focus on transfer of the FDL combined with an MCO. The subtalar
arthroereisis implant has emerged as a potentially benecial treatment option to aide the correction of hindfoot valgus.

References
[1] Pinney SJ, Lin SS. Current concept review: acquired adult atfoot deformity. Foot Ankle Int
2006;27(1):6675.
[2] Beals TC, Pomeroy GC, Manoli A 2nd. Posterior tendon insuciency: diagnosis and treatment. J Am Acad Orthop Surg 1999;7(2):1128.
[3] Bare AA, Haddad SL. Tenosynovitis of the posterior tibial tendon. Foot Ankle Clin 2001;
6(1):3766.
[4] Brodsky JW, Baum BS, Pollo FE, et al. Surgical reconstruction of posterior tibial tendon
tear in adolescents: report of two cases and review of the literature. Foot Ankle Int 2005;
26(3):21823.
[5] Park HG, Sim JA, Koh YH. Posterior tibial tendon dysfunction secondary to os subtibiale
impingement: a case report. Foot Ankle Int 2005;26(2):1846.
[6] Hadeld MH, Snyder JW, Liacouras PC, et al. Eects of medializing calcaneal osteotomy on
Achilles tendon lengthening and plantar foot pressures. Foot Ankle Int 2003;24(7):5239.
[7] Johnson KA, Strom DE. Tibialis posterior tendon dysfunction. Clin Orthop Relat Res 1989;
239:196206.
[8] Myerson MS. Adult acquired atfoot deformity: treatment of dysfunction of the posterior
tibial tendon. Instr Course Lect 1997;46:393405.
[9] Hiller L, Pinney SJ. Surgical treatment of acquired atfoot deformity: what is the state of
practice among academic foot and ankle surgeons in 2002? Foot Ankle Int 2003;24(9):7015.
[10] Sarraan S. Anatomy of the foot and ankle: descriptive, topographical. 2nd edition. Philadelphia: JB Lippincott; 1993.
[11] Bloome DM, Marymont JV, Varner KE. Variations on the insertion of the posterior tibialis
tendon: a cadaveric study. Foot Ankle Int 2003;24(10):7803.
[12] Otis JC, Gage T. Function of the posterior tibial tendon muscle. Foot Ankle Clin 2001;6(1):
114.
[13] Deland JT. The adult acquired atfoot and spring ligament complex. Pathology and implications for treatment. Foot Ankle Clin 2001;6(1):12935.
[14] Deland JT, de Asla RJ, Sung IH, et al. Posterior tibial tendon insuciency: which ligaments
are involved? Foot Ankle Int 2005;26(6):42735.
[15] Sarraan S. Syndesmology. In: Sarraan S, editor. Anatomy of the foot and ankle, vol. 1.
Philadelphia: JP Lippencott; 1983. p. 14389.
[16] Hintermann B. Biomechanics of the unstable ankle joint and clinical implications. Med Sci
Sports Exerc 1999;31(Suppl 7):S45969.
[17] Clanton T. Athletic injuries to the soft tissues of the foot and ankle. In: Coughlin M, Mann R,
editors. Surgery of the foot & ankle, vol. 2. 7th edition. St. Louis (MO): Mosby; 1999.
p. 1090209.
[18] Brodsky JW. Preliminary gait analysis results after posterior tibial tendon reconstruction:
a prospective study. Foot Ankle Int 2004;25(2):96100.

270

GIZA

et al

[19] Thordarson DB, Schmotzer H, Chon J. Reconstruction with tenodesis in an adult atfoot
model. A biomechanical evaluation of four methods. J Bone Joint Surg Am 1995;77(10):
155764.
[20] Deland JT, Arnoczky SP, Thompson FM. Adult acquired atfoot deformity at the talonavicular joint: reconstruction of the spring ligament in an in vitro model. Foot Ankle 1992;
13(6):32732.
[21] McCormack AP, Ching RP, Sangeorzan BJ. Biomechanics of procedures used in adult atfoot deformity. Foot Ankle Clin 2001;6(1):1523.
[22] Ananthakrisnan D, Ching R, Tencer A, et-al. Subluxation of the talocalcaneal joint in adults
who have symptomatic atfoot. J Bone Joint Surg Am 1999;81(8):114754.
[23] Malicky ES, Crary JL, Houghton MJ, et al. Talocalcaneal and subbular impingement in
symptomatic atfoot in adults. J Bone Joint Surg Am 2002;84-A(11):20059.
[24] Chadha H, Pomeroy G, Manoli A 2nd. Radiologic signs of unilateral pes planus. Foot Ankle
Int 1997;18(9):6034.
[25] Perry MD, Mont MA, Einhorn TA, et al. The validity of measurements made on standard
foot orthoroentgenograms [see comment]. Foot Ankle 1992;13(9):5027.
[26] Saltzman CL, Brandser EA, Berbaum KS, et al. Reliability of standard foot radiographic
measurements. Foot Ankle Int 1994;15(12):6615.
[27] Younger AS, Sawatzky B, Dryden P. Radiographic assessment of adult atfoot. Foot Ankle
Int 2005;26(10):8205.
[28] Wacker J, Calder JD, Engstrom CM, et al. MR morphometry of posterior tibialis muscle in
adult acquired at foot. Foot Ankle Int 2003;24(4):3547.
[29] Cooper J, Mizel M. The diagnosis of posterior tibial tendon tenosynovitis: comparison of
MRI vs local anesthetic tendon sheath injection. Paper Presented at the American Orthopaedic Foot and Ankle Society Annual Meeting. La Jolla (CA), July 1315, 2006.
[30] Noll KH. The use of orthotic devices in adult acquired atfoot deformity. Foot Ankle Clin
2001;6(1):2536.
[31] Havenhill TG, Toolan BC, Draganich LF. Eects of a UCBL orthosis and a calcaneal osteotomy on tibiotalar contact characteristics in a cadaver atfoot model. Foot Ankle Int
2005;26(8):60713.
[32] Schon L, Chiodo C. Current techology in posterior tibial tendonitis. Paper Presented
at the American Orthopaedic Foot & Ankle Soceity, Annual Meeting. Lo Jolla (CA),
July 1315, 2006.
[33] Alvarez RG, Marini A, Schmitt C, et al. Stage I and II posterior tibial tendon dysfunction
treated by a structured nonoperative management protocol: an orthosis and exercise program. Foot Ankle Int 2006;27(1):28.
[34] Augustin JF, Lin SS, Berberian WS, et al. Nonoperative treatment of adult acquired at foot
with the Arizona brace. Foot Ankle Clin 2003;8(3):491502.
[35] Chao W, Wapner KL, Lee TH, et al. Nonoperative management of posterior tibial tendon
dysfunction [see comment]. Foot Ankle Int 1996;17(12):73641.
[36] Frey CC, Shere MJ. Tendon injuries about the ankle in athletes. Clin Sports Med 1988;7(1):
10318.
[37] Williams R. Chronic non-specic tendovaginitis of tibialis posterior. J Bone Joint Surg Br
1963;45:5425.
[38] Mosier-LaClair S, Pomeroy G, Manoli A 2nd. Operative treatment of the dicult stage 2
adult acquired atfoot deformity. Foot Ankle Clin 2001;6(1):95119.
[39] Valderrabano V, Hintermann B, Wischer T, et al. Recovery of the posterior tibial
muscle after late reconstruction following tendon rupture. Foot Ankle Int 2004;25(2):
8595.
[40] Choi K, Lee S, Otis JC, et al. Anatomical reconstruction of the spring ligament using peroneus longus tendon graft. Foot Ankle Int 2003;24(5):4306.
[41] LaRue B, Anctil E. Distal anatomical relationship of the exor hallucis longus and exor
digitorum longus tendons. Foot Ankle Int 2006;27(7):52832.

THE FLEXIBLE FLATFOOT IN THE ADULT

271

[42] Guyton G, Jeng C, Krieger L, et al. Flexor digitorum longus transfer and medial displacement calcaneal osteotomy for posterior tibial tendon dysfunction: a middle-term clinical follow-up. Foot Ankle Int 2001;22(8):62732.
[43] Myerson MS, Badekas A, Schon LC. Treatment of stage II posterior tibial tendon deciency
with exor digitorum longus tendon transfer and calcaneal osteotomy. Foot Ankle Int 2004;
25(7):44550.
[44] Sullivan R, Gladwell H, Aronow M, et al. An in vitro study comparing the use of suture anchors and drill hole xation for exor digitorum longus transfer to the navicular. Foot Ankle
Int 2006;27(5):3636.
[45] Steensmeier SJ, Saltzman CL, Berbaum KS, et al. Eects of medial and lateral displacement calcaneal osteotomies on tibiotalar joint contact stresses. J Orthop Res 1996;14(6):
9805.
[46] Horton GA, Myerson MS, Parks BG, et al. Eect of calcaneal osteotomy and lateral column
lengthening on the plantar fascia: a biomechanical investigation. Foot Ankle Int 1998;19(6):
3703.
[47] Greene DL, Thompson MC, Gesink DS, et al. Anatomic study of the medial neurovascular
structures in relation to calcaneal osteotomy. Foot Ankle Int 2001;22(7):56971.
[48] VanAman S, Schon L. Subtalar arthroereisis as adjunct treatment for type II posterior tibial
tendon deciency. Techniques in Foot & Ankle Surgery 2006;5(2):11725.
[49] Needleman RL. A surgical approach for exible atfeet in adults including a subtalar arthroereisis with the MBA sinus tarsi implant. Foot Ankle Int 2006;27(1):918.
[50] Grice DS. An extra-articular arthrodesis of the subastragalar joint for correction of paralytic
at feet in children. J Bone Joint Surg Am 1952;34(4):92740.
[51] Subotnick SI. The subtalar joint lateral extra-articular arthroereisis: a follow-up report.
J Am Podiatry Assoc 1977;67(3):15771.
[52] Giannini BS, Ceccarelli F, Benedetti MG, et al. Surgical treatment of exible atfoot in children a four-year follow-up study. J Bone Joint Surg Am 2001;2:739.
[53] Viladot R, Pons M, Alvarez F, et al. Subtalar arthroereisis for posterior tibial tendon dysfunction: a preliminary report. Foot Ankle Int 2003;24(8):6006.

Foot Ankle Clin N Am


12 (2007) 363379

The Role of Isolated Gastrocnemius


and Combined Achilles Contractures
in the Flatfoot
Christopher W. DiGiovanni, MDa,*,
Phillip Langer, MDb
a

Division of Foot and Ankle, Department of Orthopedic Surgery, Brown Medical School,
Rhode Island Hospital, 593 Eddy Street, Providence, RI 02903, USA
b
Department of Orthopedic Surgery, Brown Medical School, Rhode Island Hospital, 593 Eddy
Street, Providence, RI 02903, USA

In the absence of bony deformity, ankle equinus is generally the result of


shortening within the gastrocnemiussoleus complex. Typical of other extremity contractures but contrary to popular belief, this functional tightness
presents itself more in the muscular portion of the musculotendinous unit
than the tendinous portion [1]. To date, equinus deformity of the ankle
has been observed to occur in two forms: one in which the gastrocnemius
muscle alone is aected, and another existing as a combined form of contracture aecting both gastrocnemius and soleus (the so-called Achilles
contracture) [15]. Isolated soleal contracture occurring in the absence of
gastrocnemius pathology has not been documented specically.
Restriction of ankle dorsiexion as a proxy for equinus contracture has
been linked to increased mechanical strains and resultant foot/ankle
pathology for a long time [1,3,629]. This entity may reect a congenital
or atavistic structural anomaly, an acquired decit from prolonged non
weight-bearing, or a developmental muscular response to chronic poor posture, as documented in joggers who run on the balls of their feet or women
who wear high-heeled shoes [30,31]. Other well-described causes of triceps
surae tightness in the literature include paralytic conditions of the lower
extremity (such as poliomyelitis) or neurologic impairments resulting in
muscle stiening or spasticity (such as diabetes and cerebral palsy) [15
20,29,32,33]. However, while such systemic abnormalities (particularly the

* Corresponding author.
E-mail address: yodigi@aol.com (C.W. DiGiovanni).
1083-7515/07/$ - see front matter 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.fcl.2007.03.005
foot.theclinics.com

364

DIGIOVANNI & LANGER

neurological derangements) undoubtedly comprise the preponderant etiologies responsible for equinus contracture, recent research eorts by Hill [3]
and DiGiovanni and colleagues [1] have suggested that contracture isolated
to the gastrocnemius can be found in healthy individuals who are devoid of
any known causative predisposition. Although not proven yet, it is suspected that the problem, as it exists in this latter group, is attributable to
the sheer structural advantage of this muscle, based on its 2/3 size dominance within the posterior calf [34], and its mechanical inuence relative
to how many joints it crosses. Evolution has probably played a role too,
considering that human transcendence into bipedal organisms has necessitated a stretching of the gastrocnemius for heels to meet the ground [35,36].
The relative degree to which isolated versus combined muscle tightness
contribute to development of equinus disease in the ankle remains unclear. Data continue to mount, however, indicating that any such stiness
left unchecked can have cumulative and perhaps profound detrimental effects on foot function [1,3,629]. Numerous disorders of the foot and ankle
have been linked with contracture of the Achilles tendon or gastrocnemius
muscle, although several of these associations remain loose and somewhat
controversial [1,3,622]. This potentially adverse relationship has been implicated, however, in a variety of pathologic processes suered by the
foot: posterior tibial tendon dysfunction (PTTD) [1,612,3747], Achilles
tendinopathy [3,42], ankle sprains and fractures [13], plantar fasciitis
[1,3,12,14,15,48], diabetic foot ulcers [1620], Charcot neuroarthropathy
[1922], amputation [17], metatarsalgia [1,3,7,11,12,28], metatarsophalangeal (MTP) joint synovitis [4,12], hallux valgus [3,7,11], claw toes [11,40],
and generalized atfoot deformity [3,6,10,26,30,39,49]. Among these pathologic conditions, the literature has established the strongest causal relationship between gastroc-soleal contracture and eventual development of
metatarsalgia, neuropathic ulceration, plantar fasciitis, and perhaps Charcot
midfoot breakdown [1,3,7,11,12,1422,48].
A reasonably convincing relationship between equinus contracture and
the development of atfoot deformity also exists, although whether this association should be characterized as more cause or eect remains unclear
and controversial [1,612,27,41,50,51]. Part of this confusion stems from
the mere lack of understanding as to what a atfoot really is. The adult
atfoot, as currently dened, occurs either as an acquired deformity in later
years or as the sequela of a developmental abnormality in childhood. Although there are many dierent manifestations of what determines a atfoot (and equally as many causative factors that can lead to this
deformity), a supercial posterior muscle compartment that is too tight,
too short, or too overpowering remains one of the most commonly observed
associated abnormalities [27,3840,42,49].
Unfortunately, the level of evidence (IV and V) used for most of the literature answering this question has been poor. The recent impetus toward
better scientic research, however, promises improvement in data and study

ROLE OF ISOLATED GASTROCNEMIUS AND COMBINED ACHILLES

365

design, which will help dene the strength and nature of this curious relationship in years to come. Regardless, there is currently enough plausible
information linking equinus contracture and atfoot deformity to warrant
in-depth discussion. This discussion requires a review of the known anatomic, kinematic, pathomechanic, and surgical relationships between the
Achilles complex and the foot.

Applied anatomy
Both the gastrocnemiussoleus complex and Achilles tendon are located
within the supercial posterior compartment of the calf. The gastrocnemius
arises from the posterior femoral condyles. Thus, it is one of few muscles in
the body to cross three major joints: knee, ankle, and subtalar. The soleus
arises from the posterior surface of the tibia, bula, and interosseous membrane, and crosses only two major joints: ankle and subtalar. As a result, the
gastrocnemius, in contradistinction to the soleus, is considered a exor of
the knee and ankle. This anatomic dierence makes it possible for the clinician to dierentiate the respective equinus contributions of these two muscles by performing the Silfverskiold [33] manuever. The examiner should be
aware, however, that the dynamic actions of both gastrocnemius and soleus
during gait might not correlate with this passive test in patients who have
muscle spasticity.
The triceps surae consists of the two heads of the gastrocnemius and a singular soleus muscle belly. These muscles converge to become an aponeurosis
roughly 1215 cm from their common insertion at the Achilles tendon. The
soleus can remain independent until insertion, but it usually does not. The
length of the Achilles tendon varies (between 48 cm depending upon patient size), and the tendon spirals counterclockwise in the axial plane before
its insertion, and most of the rotation occurs in the last 5 cm [52,53]. Hence,
bers that are medial proximally become posterior distally. The recent anatomic work presented by Baudet [23] supports the idea that the medial head
of the gastrocnemius is typically tighter than the lateral head and is responsible for the majority of dorsiexion correction upon surgical release. A priori, the same data reasonably suggest the medial head of the gastrocnemius
is most responsible for equinus pathology. Inman and colleagues have
shown that the exion moment crossing the ankle increases by a third
when the knee is in extension, because of the concomitant increased
mechanical advantage of the gastrocnemius [34].
Although the majority of available literature on Achilles tendon tightness
broaches its adverse eects by way of limited ankle dorsiexion, one should
remember that contracture of these muscles also must aect version and rotation of the subtalar joint. Whereas this might appear clinically evident
during advanced atfoot deformity, convincing scientic data specically
quantifying this nding are lacking. Available studies addressing this issue

366

DIGIOVANNI & LANGER

suggest that the resultant hindfoot valgus occurring as a part of progressive


atfoot deformity is exacerbated thereafter by Achilles contracture, as opposed to the other way around [30,49,51,54]. However, the ankle and subtalar joints work conjointly to control sagittal and transverse/coronal plane
motion respectively, as a universal joint. Thus, one cannot discuss equinus pathology as a singular plane deformity (ie, as a dorsiexion stiness
of the ankle). Rather, given the subtalar joints oblique axis and the coupled
motion of the foot in the form of dorsiexioneversion and plantarexion
inversion [55], it is perhaps more accurate to describe the anatomic impact
of a tight triceps surae in three dimensions. Hence, as a foot is forced downward against an equal and oppositely (upwardly) directed ground reactive
force, the presence of restricted ankle dorsiexion during stance phase
must be compensated by one of the following: (1) external rotation of the
leg, (2) lumbar lordosis, hip exion, or knee recurvatum, or (3) excess hindfoot pronatory malrotation through the oblique axis subtalar joint (a potential precursor to exible and eventually rigid atfoot deformity) [40,56].
Conceptually, mechanical medial column breakdown as part of the natural history of repetitive equinus stress is also supported by limited histopathologic data. It is well accepted that one of the leading causes of
atfoot deformity is PTTD. Mosier and colleagues [47] have performed histochemical analyses on patients who have this disorder and found that the
diseased tendons examined exhibited a non-specic reparative response
to tissue injury, as opposed to showing any histologic changes suggestive
of acute or chronic tendonitis. These tendons had evidence of neovascularization, mucoid degeneration, chondroid metaplasia, and broblast hypercelluarity, which were changes considered far more suggestive of
a degenerative response rather than an inammatory response. Such information could be interpreted as being consistent with the notion that equinus
contracture predates most atfoot deformities as the catalyst of a gradual
wear and tear process, rather than the other way around. This remains
debatable, however, and in fact the authors of this study point out that
they are unclear as to whether these changes pre- or post-date the PTTD.
When weighing the merits of any potential relationship between Achilles
or isolated gastrocnemius contractures and atfoot deformity, one must
consider that the Achilles tendon is subjected to the highest loads in the
body. It routinely supports tensile loads up to 10 times the body weight during running, jumping, hopping, and skipping [57]. The tendon lacks a true
synovial sheath, but has a paratenon surrounding it. Under normal circumstances, this tendon is able to stretch 23 cm in response to movement, allowing maximal gliding action. Its blood supply is provided by
longitudinal arteries, which descend the length of the muscle complex and
ascend from its calcaneal insertion. The so-called watershed area of the
tendon, where the most tenuous vascularity can be found, is located approximately 26 cm above the Achilles tendon insertion [58]. With age or alternatively pathologic conditions such as diabetes, this blood supply diminishes

ROLE OF ISOLATED GASTROCNEMIUS AND COMBINED ACHILLES

367

and the tendon is predisposed to chronic inammation, decreased cell and


collagen bril density, increased collagen cross linking, and repetitive damage that often results in developmental contracture or even rupture [59].
Consequent to both the increased size and stiness that occur as a natural
part of the aging process in this tendon, it is doubtful that the Achilles tendon remains easily stretched over time [60]. Under such circumstances, most
patients are told to stretch their Achilles to treat plantar fasciitis or equinus. One might speculate that the leveraged plantar capsular and ligamentous structures in the midfoot and forefoot are being stretched instead of
the gastrocnemiussoleus complex. This question has yet to be answered,
but it is an interesting one in light of the discussion on the natural history
of atfoot deformity.

Gait kinematics and pathomechanics


During the normal gait cycle, the foot transitions itself from an adaptive, exible construct at heel strike to a rigid lever at push o [40]. The
morphology of the longitudinal arch is critical to this mechanism and
can be thought of as a conduit between the three points of a tripod. In traversing our entire foot, this bridge must rst absorb all force descending
through the ankle, and then distribute the force in rapid succession from
hindfoot (tuber heel strike) to forefoot (medial and lateral column metatarsal head toe o) during weight-bearing progression. Because the tibiotalar
joint is situated posteriorly to the central axis of this arch and the Achilles
tendon exerts its inuence through an insertion located even farther posteriorly, any imbalance favoring relative overpull of the triceps surae will
result in excess forces being dispersed by the longitudinal arch during
load transfer [49,61,62]. In fact, Thordarson and colleagues [61] demonstrated (in a cadaver model) that the gastroc-soleal complex was the greatest contributor to arch attening in the sagittal plane and a major
contributor to forefoot abduction in the transverse plane of the human
foot. Hence, a competent and stable longitudinal arch and a proper
balance of forces generated between the posterior and antagonist compartments in the leg are vital to maintaining normal gait propulsion and preventing atfoot deformity.
Under normal circumstances, the anterior compartment of the leg eccentrically contracts at heel strike to decelerate the foot as it rolls into foot at,
while the muscles in the posterior compartment relax. From heel strike to
mid-stance, the gastrocnemiussoleus complex remains quiescent; when it
is not tight, progressive eversion of the subtalar joint is allowed. As the hindfoot assumes a valgus position, the axes of the talonavicular and calcaneocuboid joints become parallel, unlock, and permit abduction through the
transverse tarsal joints [56]. When working properly, this chain of events increases the amount of accommodative motion that can occur through the

368

DIGIOVANNI & LANGER

midfoot for shock absorption during mid-stance. As the body progresses


forward over the foot during mid-stance, the knee extends and the resultant
eccentric muscle contraction of the posterior compartment of the leg acts to
decelerate the tibia and stabilize the midfoot and ankle. To create a rigid lever for push o before the heel rise phase of gait, the normal PTT begins
contracting to initiate subtalar inversion, which progressively occurs
through the coupled motions of tibial external rotation, ankle dorsiexion,
and tightening of the plantar fascia [63,64]. The windlass mechanism elevates and inverts the arch, which thereafter locks the midfoot joints, and
shifts the insertion of the Achilles tendon medial to the axis of rotation of
the subtalar joint. Such positioning transforms the entire length of the
foot into a rigid lever, which, together with the contraction of the Achilles
complex, becomes capable of propelling the body forward through push
o from the metatarsal heads. After swing phase is completed, this normal
cycle repeats itself.
However, medial column incompetence through either ligamentous or
tendinous insuciency can disrupt this sequence. If the subtalar joint is
able to or forced to undergo excessive pronation, the unlocking of the Chopart joints permits increased dorsiexion and eventual attenuation of the
plantar soft tissue stabilizers [40,56,59]. If the structures comprising the medial architecture become stretched, the resultant decient arch stability and
poor inversion impede ecient load transfer to the forefoot as a means of
eective propulsion. This causes repetitive loading, which cannot eectively
bypass the midfoot and alternatively attens the arch further [41,47]. Because the mechanical axis of the ankle joint rests anterior to the weight-bearing surface of the calcaneal tuber, the medial arch has been described as an
upside down leaf spring that dampers the impact of loads during gait [49].
If one espouses the theory that a tight gastroc-soleal complex is primarily
a causative agent in this process, then an argument could be made that
the repetitive mechanical stress that occurs as a result of equinus overload
eventuates into insidious deterioration of plantar supporting structures, particularly the peroneus longus tendon, the plantar fascia, and the plantar
(spring) ligaments and capsules. Persistent strain on the PTT leads to eventual tendonitis, elongation (tendinosis), or, nally, rupture [50]. It is these
vital structures on which the integrity of the medial longitudinal arch depends [7,8,38,39,42,49,61,62]. As the structures progressively become
stretched out, the atfoot is developed. Alternatively, if one believes it is
the deterioration of these supporting structures that occurs primarily, then
the sequence of events would be interpreted dierently. Under these circumstances, such collapse initially would result in abduction deformity and a relatively plantarexed posture to the hindfoot (in the face of an initially
normal Achilles complex). Calcaneal eversion and declination, accompanied
by an inevitable talar plantarexion caused by lack of plantar support,
would predispose to adaptive shortening of the gastroc-soleal complex.
The Achilles tendon would become unable to stretch during gait and would

ROLE OF ISOLATED GASTROCNEMIUS AND COMBINED ACHILLES

369

tighten further. Regardless of whether this relationship is primarily one of


cause or eect, the likely reality is that the process represents a self-perpetuating combination of both issues: (1) a tightened Achilles tendon leads to
increased force transmittal through the midfoot structures, further altering
their integrity and thereby hastening midfoot collapse, and (2) hindfoot collapse permits greater midfoot attening and hindfoot eversion, culminating
in greater accommodative triceps tightness. The exact reverse can be argued
reasonably as well, with a similar net eect. Unfortunately, of the fair
amount of data that have identied the probability of this relationship,
much have been collected during cadaveric or bench top research, and
very little have been collected during natural history researchdwhich is
mandatory before any conclusion can be drawn regarding this chicken
and egg debate [27,37,43,44,50,54].
Various investigators have implied that the primary mechanism by
which equinus contracture probably leads to foot and ankle pathology
is limited ankle dorisiexion during gait [4]. The reported amount of ankle dorsiexion required for normal gait ranges from 1022 ; the most
commonly quoted value is 10 [24,25,32,65,66]. During the mid-stance
phase of the normal gait cycle, maximal ankle dorsiexion is required
just before heel rise [40]. At this point, the knee nears full extension,
which stretches the gastrocnemius. In the event an individual already
has a tight gastrocnemius or combined gastroc-soleal (Achilles) contracture, this additional plantarexion tension may force one to adopt an
early heel rise gait pattern, which is known to compromise normal gait
biomechanics [6769]. In its extreme, some investigators suggest an early
heel rise gait pattern could lead to a toe walking gait or the generation of
chronic breakdown or pain within the midfoot and forefoot, since neither
the ground nor the Achilles tendon will ever be the loser of this battle
(Fig. 1) [1,3,4,70]. Several investigators have supported this concept with
retrospective observations, but their information was poorly documented
and likely would not withstand the rigors of todays elevated research
standards. Hill [3], Downey and Banks [7], and Subotnick [28] have called
attention to notable improvement of gait mechanics and foot function
following correction of gastrocnemius or soleus tightness in nonspastic individuals. Whereas these investigators, like others, have been unable to
cite signicant scientic evidence to support their observations, Subotnick
has taken literary license to poignantly describe gastroc-soleal equinus as
the greatest symptom producer in the human foot [28]. Recent investigative eorts by others also have lent support to these words. Aronow
and colleagues [4] demonstrated that abnormally increased forces from either the triceps surae or isolated gastrocnemius muscle shift plantar
weight-bearing pressures from hindfoot to midfoot and forefoot. Interestingly, similar pathologic load shifts were observed in this study regardless
if Achilles tendon force was applied through the gastrocnemius, soleus, or
both muscles.

370

DIGIOVANNI & LANGER

Fig. 1. Graphically depicted load patterns during gait. This illustration demonstrates the anticipated variation in forefoot and midfoot loading during late stance phase gait between subjects who
have normal gastrocnemius length and patients who have an abnormal (ie, contracted) gastrocnemius length. (From DiGiovanni CW, Kuo R, Tejwani N, et al. Isolated gastrocnemius tightness. J
Bone Joint Surg 2002;84:968; with permission from the Journal of Bone and Joint Surgery, Inc.)

Adult atfoot: natural history and related ndings


In spite of our growing understanding of atfoot pathomechanics, it is
unclear still as to whether a tight supercial posterior leg compartment is
considered most aptly as an etiologic factor or one of the many anatomic
sequelae of this progressive deformity. Most of the available literature can
be interpreted to support either side of this argument. What is clear, but remains to be quantied, is that, historically, acquired atfeet in adults and
developmental atfeet in children have been identied in frequent association with a short or contracted Achilles tendon [3,8,26]. In the 1940s, Harris
and Beath [8] made these observations independently, and Broderson and
colleagues [26] followed in the 1990s; together, they developed some of
the few studies broaching the natural history of atfoot deformity. Additional studies investigating treatment of atfeet have noted a relationship
between Achilles contracture and atfoot. In a retrospective investigation
by Fraser and colleagues [71], 84% of symptomatic mobile atfeet associated with an isolated break at the naviculocuneiform joint had a satisfactory

ROLE OF ISOLATED GASTROCNEMIUS AND COMBINED ACHILLES

371

clinical result when treated with the Miller procedure, and 92% of the operations required an Achilles tendon lengthening for xed equinus deformity.
PTTD has been related to and probably is the most representative and studied of all causative factors in atfoot [1,612]. PTTD commonly is considered a clinical continuum that is capable of progressing in stages, from an
isolated tenosynovitis into a exible and eventually rigid atfoot malalignment [27,50]. Apart from the before or after debate, the exact magnitude
that isolated or combined gastroc-soleal contracture plays in this course has
yet to be quantied. Whereas some investigators continue to question the
very existence of isolated gastrocnemius contracture, others remain staunch
advocates that this entity is part of the natural history of developmental atfoot [1,35,37,40,42,49,72].
The precipitating events aecting the natural history of atfoot deformity
are centered on the position of the talus. The talus is normally in contact
with the anterolateral edge of the proximal navicular bone, the most superior bone of the medial longitudinal arch. Anteriorly, the talus articulates
with and is supported by the medial aspects of the calcaneus. Any loss of
calcaneal support through forced eversion/external rotation produces plantarmedial talar subluxation and a dorsally subluxating force on the talonavicular joint; this causes excessive pressure on the spring ligament [51] and
subsequent collapse of the medial longitudinal arch. Everything begins to
move around a predominantly stationary talus. Gradual stretching or rupture of the PTT eliminates the primary inverter of the foot and renders the
antagonist everters (the peroneus brevis and longus) relatively unopposed.
Over time, this over pull of the peroneal musculature exacerbates medial
longitudinal arch attening, forefoot abduction, and hindfoot valgus. The
moment arm of a tight Achilles tendon under these circumstances transforms the gastrocnemiussoleus complex from a heel inverter into a heel
everter and creates a plantar-driven force, which further exacerbates calcaneal eversion and hindfoot pronation [73]. This subsequently places extra
stress on the medial column and increases weight-bearing on the medial
side of the heel. Once such a process is set in motion, it is questionable if
it can be halted successfully, regardless of empiric attempts at triceps surae
correction. The progress of many such deformities, if left uncorrected surgically, is arguably dependent on genetics and eventually may lead to deltoid
ligament insuciency and valgus instability of the ankle (stage IV PTTD)
[27,51]. This level of disease is associated often with severe Achilles and peroneal contracture, subbular impingement, lateral foot pain, and eventually ankylosis and degenerative arthrosis.
In an attempt to overcome either progressive or pre-existent triceps surae
contracture in atfoot deformity, the antagonistic anterior compartment extensors (extensor digitorum longus and extensor hallucis longus) are recruited
to help the mechanically disadvantaged tibialis anterior tendon dorsiex the
ankle. This often leads to a mild hyperextension deformity of the MTP joints
commonly seen as a component of atfeet, a process termed extensor

372

DIGIOVANNI & LANGER

recruitment [42]. As the hyperextended toes strike the top of a shoe repetitively,
hammertoe deformities may develop at the interphalangeal joints as the shoe
pushes the tips of the toes back down. Distal plantar fat pad migration also
accompanies this extensor recruitment, and when exacerbated by the foreand midfoot weight redistribution commonly caused by the associated equinus contracture, many other maladies have been reported to occur with
chronic atfoot deformity. These include metatarsalgia [1,3,7,11,12,28],
MTP joint synovitis [4,12], metatarsal stress syndrome or fracture, plantar ulceration (in the event of superimposed neuropathy) [19,21,22], and complaints
of calf tightness or symptomatic insertional Achilles tendinopathy [3,42].
However, it should be noted that these problems have been observed (often anecdotally) in combination with and separate from atfoot deformity when isolated gastrocnemius or triceps surae contracture is present. Further study is
warranted to examine the true nature of these purported relationships.
The radiographic eects of atfoot deformity are easily recognizable on
standing, weight-bearing anteroposterior and lateral views of the foot
(Fig. 2A, B). Although atfoot is known to have a varied clinical presentation
and set of causative factors, there are several reliable radiographic indicators
one can use to identify it. Because the other bones move around the talus in
patients who have atfoot deformity, the talus typically exhibits little change
in position except in the more extreme deformities (late stage 2 and beyond)
when it can be seen to plantarex and increasingly overlap the calcaneus.
Termed dorsolateral peritalar subluxation, the foot on weight-bearing plain
lms swings out from underneath the talus. This can be observed as a combination of any number of anatomic ndings: talonavicular uncovering, medial
column sag (the tarsometatarsal, naviculocuneiform, and/or talonavicular
joints), plantar medial column gapping, increased talocalcaneal overlap, or increased talar declination and/or decreased calcaneal inclination. The hindfoot
often can be observed in a normal unaltered position, but it is rarely noted to
dorsiex during atfoot deformity. Hence, when the atfoot deformity aects
the hindfoot, the calcaneus typically assumes a more equinus position, presumably resulting from the pull of a tight Achilles tendon.

Fig. 2. (A, B) Preoperative anteroposterior and lateral weight-bearing radiographs of a patient


who have a exible atfoot deformity.

ROLE OF ISOLATED GASTROCNEMIUS AND COMBINED ACHILLES

373

Examples of some classic medial column changes that occur radiographically as a result of chronic mechanical collapse of the medial longitudinal
arch can be seen clearly in the preoperative radiographs of a patient who
has long-standing stage II PTTD in Fig. 2B. Note that the postoperative radiographs in Fig. 3 demonstrate medial column correction at the site of deformity,
rather than somewhere else in the foot, so as not to trade a deformity for a deformity. This realignment was accomplished through a combination of midfoot naviculocuneiform and tarsometatarsal fusion, exor digitorum longus
to posterior tibial augmentation, and gastrocnemius recession, which allowed
maximal restoration of PTT strength, calcaneal inclination, hindfoot/midfoot
realignment, and return of functional ankle dorsiexion. Patients who have
chronic but exible atfoot deformity frequently exhibit isolated gastrocnemius or combined triceps contracture when carefully examined. However,
exible atfoot deformity can be identied only by reducing the talonavicular
joint and maintaining a neutrally aligned foot before any assessment of equinus. If this realignment is not performed rst, a false identication of normal
dorsiexion ability (which is coming from medial column sag and dorsal subluxation during application of upwardly directed plantar pressure) is erroneously interpreted as lack of equinus. Identication of this entity also is
underestimated if patients have a rigid (irreducible) deformity that disallows
hindfoot reduction. Under these circumstances, true appreciation of gastroc-soleal tightness is impossible through the Silfverskiold [33] maneuver.
It has been the authors continued experience and opinion that the vast majority of patients who have atfoot deformity, if tested properly, will demonstrate
isolated or combined supercial posterior compartment tightness. Thus, the
authors strongly advocate its assessment as part of the routine workup and
treatment of atfoot deformity.
Conservative and surgical management
The authors believe one of the most convincing arguments, which inexorably links Achilles contracture with atfoot deformity, is made by simply

Fig. 3. (A, B) Postoperative anteroposterior and lateral radiographs demonstrating surgical


correction through medial column fusion.

374

DIGIOVANNI & LANGER

noting that it is very dicult, if not impossible, to enact surgical, anatomic


correction of stage II or stage III PTTD without concomitant Achilles tendon or isolated gastrocnemius lengthening [5,7,11,12,27,37,41,47,73,74]. In
the authors experience, trying to do so can result in either persistent malalignment despite concomitant arthrodesis/osteotomy or fracture/nonunion
in the event of lateral column lengthening. Without a concomitant release
of the supercial posterior compartment, it is very dicult to restore calcaneal inclination, reposition the calcaneus inferiorly to again support the talus, or lengthen the lateral column to swing the foot around during the
correction of atfoot deformity. Similar diculties can be encountered during posttraumatic reconstruction of the hindfoot when trying to re-establish
height, width, and length after calcaneal fracture. When left untreated, such
individuals often develop superior and medial malunion of the heel caused
by the pathologic pull of the Achilles similarly seen in atfoot deformity.
Most if not all of these patients develop gastroc-soleal contracture, either
in the acute setting if this is not prevented with splinting while waiting for
the soft tissues to subside, or in the chronic setting if calcaneal malunion
is permitted [45]. Whereas some of these anecdotal observations are liable
to be contested, the authors believe they serve to strengthen the belief that
a tight gastroc-soleal complex plays a vital and important pathologic role
in malalignments such as atfoot.
Empiric stretching of the gastroc-soleal complex with physical therapy,
night splints, or serial casting is advised constantly in the literature as prophylaxis against or even treatment of many aforementioned ailments associated with equinus contracture. However, there are few data to support
this recommendation [46,48,7577]. Occasionally, reported outcome scores
support that a management regimen may be symptomatically helpful, but
there is little science documenting that any stretching of this complex actually occurs. In fact, some data imply that this complex cannot be stretched
[60]. The authors propose that the act of triceps surae stretching does not
actually lengthen anything in the supercial posterior compartment, but instead it eectively prevents further gastroc-soleal contracture and may result
in the stretching of alternatively weaker, less resistant structures, such as the
plantar tendons, ligaments, fascia, and capsules. Recent studies conrm that
stretching of these important stabilizing structures of the medial column
rapidly can lead to atfoot deformity [61,78,79]. However, more carefully
controlled outcome studies need to be performed to denitively answer
this question.
There is little doubt that percutaneous, open, and endoscopic surgical release of the gastrocnemius or Achilles tendon can result in signicant immediate equinus correction and restoration of functional dorsiexion (Fig. 4)
[4,5,18,67,74,80]. Data also conrm that such corrections can be maintained
over time, but this information is more limited [5]. Nonetheless, despite the
objective return of motion, there is currently a dearth of sucient science
linking prophylactic Achilles tendon or gastrocnemius release to any

ROLE OF ISOLATED GASTROCNEMIUS AND COMBINED ACHILLES

375

Fig. 4. Intraoperative photograph showing the gastrocnemius muscle-tendinous junction


exposed in preparation for open release.

slowing of atfoot deformity or any deterrence of its natural history. Although research clearly needs to be performed to evaluate this more closely,
this idea seems plausible based on increasingly mounting evidence [74] to advocate such treatment, because many surgeons empirically do it now. This
support is predicated on the relatively low risk, minimal morbidity, and
rapid recovery of such surgery for the patient, particularly for isolated gastrocnemius release. Those who embrace this technique also point to the considerable need to address this problem surgically when performing atfoot
correction [74,8183]. As much as this concept may make sense, though,
one must accept that it remains, as yet, a predominantly theoretic relationship. No unequivocal evidence exists that equinus contracture single-handedly can cause a atfoot deformity or that its removal can prevent one from
occurring. As interest in this pathologic entity continues to escalate, it is
hoped that these questions will have better answers and explanations in
the years forthcoming.

Summary
Considerable controversy and confusion persist regarding the role of gastroc-soleal and isolated gastrocnemius contracture in atfoot deformity.
The limited data currently available can be interpreted to support equinus
contracture as being either a cause or eect in this process. There is at least
general agreement by most authors in the orthopedic literature that some relationship exists between equinus contracture and atfoot deformity. Furthermore, there is reasonable consensus that the primary mode of failure
in progressive atfoot deformity is loss of dynamic arch support, followed
closely by tension failure of the surrounding static restraints. Certainly,
the interplay between atfoot development and Achilles tendon or isolated
gastrocnemius tightness remains complicated and incompletely understood.

376

DIGIOVANNI & LANGER

Regardless of ones current belief regarding the exact biomechanical sequence of events surrounding these abnormalities, the important point to
recognize is that they frequently coexist. Appreciating this relationship as
a very real component of such pathology is imperative in being able to
eectively diagnose and manage the problems associated with atfoot and
equinus deformity, which typically occurs in several planes. Future well-designed, prospective, randomized, controlled clinical trials will be the only
way to characterize this association denitively. These criteria are undoubtedly necessary when considering the complex and pervasive nature of such
problems in daily foot and ankle care.
References
[1] DiGiovanni CW, Kuo R, Tejwani N, et al. Isolated gastrocnemius tightness. J Bone Joint
Surg Am 2002;84:96270.
[2] Hillstrom HJ, Perlberg G, Siegler S, et al. Objective identication of ankle equinus deformity
and resulting contracture. J Am Podiatr Med Assoc 1991;81:51924.
[3] Hill RS. Ankle equinus. Prevalence and linkage to common foot pathology. J Am Podiatr
Med Assoc 1995;85:295300.
[4] Aronow MS, Diaz-Doran V, Sullivan RJ, et al. The eect of triceps surae contracture force
on plantar foot distribution. Foot Ankle Int 2006;27(1):4352.
[5] Pinney SJ, Hansen ST, Sangeorzan BJ. The eect on ankle dorsiexion of gastrocnemius
recession. Foot Ankle Int 2002;23(1):269.
[6] Aronow MS. Triceps surae contractures associated with posterior tibial tendon dysfunction.
Tech Orthop 2000;15:16473.
[7] Downey MS, Banks AS. Gastrocnemius recession in the treatment of nonspastic ankle equinus: a retrospective study. J Am Podiatr Med Assoc 1989;79:15974.
[8] Harris RI, Beath T. Hypermobile at-foot with short tendo Achilles. J Bone Joint Surg Am
1948;30:11641.
[9] Hibbs RA. Muscle bound feet. New York Medical Journal 1914;17C:7979.
[10] Hoke M. An operation for the correction of extremely relaxed at feet. J Bone Joint Surg Am
1931;13:77384.
[11] Sgarlato TE, Morgan J, Shane HS, et al. Tendo Achilles lengthening and its eect on foot
disorders. J Am Podiatr Med Assoc 1975;65:84971.
[12] Stotler WM, Van Bergeyk A, Manoli A. Preliminary results of gastrocnemius recession in
adults with nonspastic equinus contracture. Presented at the American Orthopaedic Foot
and Ankle Society 18th Annual Summer Meeting. Traverse City (MI), July 13, 2002.
[13] Tabrizi P, McIntyre WM, Quesnel MB, et al. Limited dorsiexion predisposes to injuries of
the ankle in children. J Bone Joint Surg Br 2000;82:11036.
[14] Kibler WB, Goldberg C, Chandler TJ. Functional biomechanical decits in running athletes
with plantar fasciitis. Am J Sports Med 1991;19:6671.
[15] Lavery LA, Armstrong DG, Boulton AJ. Ankle equinus deformity and it is relationship to
high plantar pressure in a large population with diabetes mellitus. J Am Podiatr Med Assoc
2002;92:47984.
[16] Armstrong DG, Stacpoole-Shea S, Nguyen H, et al. Lengthening of the Achilles tendon in
diabetic patients who are at high risk for ulceration of the foot. J Bone Joint Surg Am
1999;81:5358.
[17] Barry DC, Sabacinski KA, Havershaw GM, et al. Tendon Achilles procedures for
chronic ulcerations with transmetatarsal amputations. J Am Podiatr Med Assoc 1993;
83:96100.

ROLE OF ISOLATED GASTROCNEMIUS AND COMBINED ACHILLES

377

[18] Lin SS, Lee TH, Wapner KL. Plantar forefoot ulceration with equinus deformity of the ankle
in diabetic patients. The eect of tendo-Achilles lengthening and total contact casting.
Orthopedics 1996;19:46575.
[19] Mueller MJ, Sinacore DR, Hastings MK, et al. Eect of Achilles tendon lengthening on neuropathic ulcers. A randomized clinical trial. J Bone Joint Surg Am 2003;85:143645.
[20] Nishimoto GS, Attinger CE, Cooper PS. Lengthening the Achilles tendon for the treatment
of diabetic plantar forefoot ulceration. Surg Clin North Am 2003;83:70726.
[21] Early JS, Hansen ST. Surgical reconstruction for the diabetic foot: a salvage approach for
midfoot collapse. Foot Ankle Int 1996;17:32530.
[22] Schon LC, Easley ME, Weinfeld SB. Charcot neuroarthropathy for the foot and ankle. Clin
Orthop 1998;349:11631.
[23] Baudet B. The tennis leg. Presented at the International Spring Meeting of the French Foot
Socities. Toulouse (France), June 9, 2006.
[24] Morton DJ. Human locomtion and body form: a study of gravity and man. Baltimore (MD):
The Williams & Wilkins Company; 1952.
[25] Root ML, Orien WP, Weed JH. Forces acting upon the foot during locomotion. In: Root
ML, Orien WP, Weed JH, editors. Normal and abnormal function of the foot, clinical biomechanics (Clinical Biomechanics Corporation, vol. 2). Los Angeles (CA): 1977. p. 16579.
[26] Brodersen A, Pedersen B, Reimers J. Foot deformities and relation to the length of leg muscles in Danish children aged 3-17 years. Ugeskr Laeger 1993;155:29146.
[27] Myerson MS. Adult acquired atfoot deformity. Treatment of dysfunction of the posterior
tibial tendon. J Bone Joint Surg Am 1996;78:78092.
[28] Subotnick SI. Equinus deformity as it aects the forefoot. J Am Podiatry Assoc 1971;61:
4237.
[29] Beckerman H, Becher J, Lankhorst GJ, et al. Walking ability of stroke patients: ecacy of
tibial nerve blocking and a polypropylene ankle-foot orthosis. Arch Phys Med Rehabil 1996;
77:114451.
[30] Stovitz SD, Coetzee JC. Hyperpronation and foot pain. Phys Sportsmed 2004;32(8).
[31] Snow RE, Williams KR. High heeled shoes: their eect on center of mass position, posture,
three-dimensional kinematics, rearfoot motion, and ground reaction forces. Arch Phys Med
Rehabil 1994;75(5):56876.
[32] Banks HH. Equinus and cerebral palsydits management. Foot Ankle Int 1983;4:19159.
[33] Silfverskiold N. Reduction of the uncrossed two-joints muscles of the leg to one-joint muscles in spastic conditions. Acta Chir Scand 1924;56:14959.
[34] Inman VT, Ralston HJ, Todd F. Human walking. Baltimore (MD): Williams and Wilkins;
1981.
[35] Ackerman J. The downside of upright. National Geogr Mag 2006;210(1):12645.
[36] Lemonick MD, Dorfman A. Up from the apes: remarkable new evidence is lling in the story
of how we became human. Time 1999;154(8):4453.
[37] Beals T, Pomeroy G, Manoli A. Posterior tibial tendon insuciency: diagnosis and treatment. J Am Acad Orthop Surg 1999;7(2):1128.
[38] Johnson KA. Tibialis posterior tendon rupture. Clin Orthop Relat Res 1983;177:1407.
[39] Meehan RE, Brage M. Adult acquired at foot deformity: clinical and radiographic examination. Foot Ankle Clin 2003;8(3):43152.
[40] Mann RA. Flatfoot in adults. In: Mann RA, Coughlin M, editors. Surgery of the foot and
ankle. 7th edition. St. Louis (MO): Mosby; 1999. p. 73367.
[41] Funk DA, Cass JR, Johnson KA. Acquired adult at foot secondary to posterior tibialtendon pathology. J Bone Joint Surg Am 1986;68:95102.
[42] Hansen ST. Progressive symptomatic atfoot (lateral peritalar subluxation). In: Hansen ST,
editor. Functional reconstruction of the foot and ankle. Philadelphia: Lippincott Williams &
Wilkins; 2000. p. 195207.
[43] Niki H, Ching RP, Kiser P, et al. The eect of posterior tibial tendon dysfunction on hindfoot kinematics. Foot Ankle Int 2001;22(4):292300.

378

DIGIOVANNI & LANGER

[44] Imhauser CW, Siegler S, Abidi NA, et al. The eect of posterior tibialis tendon dysfunction
on the plantar pressure characteristics and the kinematics of the arch and the hindfoot. Clin
Biomech 2004;19(2):1619.
[45] DiGiovanni CW, Benirschke SK, Hansen ST, et al. Foot injuries. In: Browner BD, Jupiter
JB, Levine AM, editors. Skeletal trauma. 3rd edition , vol. 2. Philadelphia: WB Saunders Co;
2003. p. 240634.
[46] Chao W, Wapner KL, Lee TH, et al. Nonoperative management of posterior tibial tendon
dysfunction. Foot Ankle Int 1996;17(12):73641.
[47] Mosier-LaClair S, Pomeroy G, Manoli A. Operative treatment of the dicult stage 2 adult
acquired atfoot deformity. Foot Ankle Clin 2001;6(1):95119.
[48] Powell M, Post WR, Keener J, et al. Eective treatment of chronic plantar fasciitis with dorsiexion night splints: a crossover prospective randomized outcome study. Foot Ankle Int
1998;19(1):108.
[49] Van Boerum DH, Sangeorzan BJ. Biomechanics and pathophysiology of at foot. Foot
Ankle Clin 2003;8(3):41930.
[50] Johnson KA, Strom DE. Tibialis posterior tendon dysfunction. Clin Orthop 1989;239:
196206.
[51] Goldner JL, Keats PK, Bassett FH, et al. Progressive talipes equinovalgus due to trauma or
degeneration of the posterior tibial tendon and medial plantar ligaments. Orthop Clin North
Am 1974;5:3951.
[52] Schepsis AA, Jones H, Haas AL. Achilles tendon disorders in athletes. Am J Sports Med
2002;30(2):287305.
[53] Palalastanga N. Muscles plantarexing the ankle joint. In: Anatomy and human movement:
structure and function. Oxford (UK): Heinemann Medical Books; 1989. p. 3603.
[54] Erdemier A, Hamel AJ, Fauth AR, et al. Dynamic loading of the plantar aponeurosis in
walking. J Bone Joint Surg Am 2004;86:54652.
[55] Lewis OJ. The joints of the evolving foot. Part II. The intrinsic joints. J Anat 1980;130(4):
83357.
[56] Perry J. Ankle foot complex. Gait analysis: normal and pathology function. Thorofare (NJ):
SLACK, Inc.; 1992. p. 5188.
[57] OBrien M. The anatomy of the Achilles tendon. Foot Ankle Clin 2005;10(2):22538.
[58] Langergren C, Lindholm A. Vascular distribution in the Achilles tendon. An angiographic
and microangiographic study. Acta Chir Scand 1958;116:4915.
[59] Saltzman CL, Tearse D. Achilles tendon injuries. J Am Acad Orthop Surg 1998;6:31625.
[60] Grady JF, Saxena A. Eects of stretching the gastrocnemius muscle. J Foot Surg 1991;30:
4659.
[61] Thordarson DB, Schmotzer H, Chon J, et al. Dynamic support of the human longitudinal
arch. A biomechanical evaluation. Clin Orthop Relat Res 1995;316:16572.
[62] Pedowitz WJ, Kovatis P. Flat foot in the adult. J Am Acad Orthop Surg 1995;3:293302.
[63] Mann R, Inman VT. Phasic activity of intrinsic muscles of the foot. J Bone Joint Surg Am
1964;46:46981.
[64] Sutherland DH. An electromyographic study of the plantar exors of the ankle in normal
walking on the level. J. Bone Joint Surg Am 1966;48:6671.
[65] Murray M, Drought A, Kory RC. Walking patterns of normal men. J Bone Joint Surg Am
1964;46:33560.
[66] Rome K. Ankle joint dorsiexion measurement studies. A review of the literature. J Am
Podiatr Med Assoc 1996;86:20511.
[67] Tashjian RZ, Appel AJ, Banerjee R, et al. Anatomic study of the gastrocnemius-soleus junction and its relationship to the sural nerve. Foot Ankle Int 2003;24(6):4736.
[68] Matjacic Z, Olensek A, Bajd T. Biomechanical characterization and clinical implications of
articially induced toe-walking: dierences between pure soleus, pure gastrocnemius, and
combination of soleus and gastrocnemius contractures. J Biomech 2006;39(2):25566.

ROLE OF ISOLATED GASTROCNEMIUS AND COMBINED ACHILLES

379

[69] Wright DG, Desai SM, Henderson WH. Action of the subtalar and ankle-joint complex
during stance phase of walking. J Bone Joint Surg Am 1964;46:36182.
[70] Neptune RR, Burneld JM, Mulroy SJ. The neuromuscular demands of toe walking: a
forward dynamics simulation analysis. J Biomech 2006;40(6):1293300.
[71] Fraser RK, Menelaus MB, Williams PF, et al. The miller procedure for mobile at feet.
J Bone Joint Surg Br 1995;77:3969.
[72] Mann RA. RE: the eect on ankle dorsiexion of gastrocnemius recession, Pinney SJ, et al.,
Foot Ankle Int. 2002;23(1):269. Foot Ankle Int 2003;24(9):7267.
[73] Anderson RB, Davis WH. Management of adult atfoot deformity. In: Myerson MS, editor.
Foot and ankle disorders. Philadelphia: WB Saunders Co; 2000. p. 101739.
[74] Saxena A, Widtfeldt A. Endoscopic gastrocnemius recession: preliminary report on 18 cases.
J Foot Ankle Surg 2004;45(5):3026.
[75] Batt ME, Tanji JL, Skattum N. Plantar fasciitis: a prospective randomized clinical trial of the
tension night splint. Clin J Sport Med 1996;6(3):15862.
[76] Dananberg HJ, Shearstone J, Guillano M. Manipulation method for the treatment of ankle
equinus. J Am Podiatr Med Assoc 2000;90(8):3859.
[77] Elftman WW. Nonsurgical treatment of adult acquired atfoot deformity. Foot Ankle Clin
2003;8(3):47389.
[78] Huang C, Kitaoka HB, An KN, et al. Biomechanical evaluation of longitudinal arch stability. Foot Ankle 1993;14(6):3537.
[79] Crary JL, Hollis JM, Manoli A. The eect of plantar fascia release on strain in the spring and
long plantar ligaments. Foot Ankle Int 2003;24(3):24550.
[80] Pinney SJ, Sangeorzan BJ, Hansen ST. Surgical anatomy of the gastrocnemius recession
(Strayer procedure). Foot Ankle Int 2004;25(4):24750.
[81] Sammarco GJ, Bagwe MR, Sammarco VJ, et al. The eects of unilateral gastrocsoleus recession. Foot Ankle Int 2006;27(7):50811.
[82] Trevino S, Gibbs M, Panchbhavi V. Evaluation of results of endoscopic gastrocnemius
recession. Foot Ankle Int 2005;26(5):35964.
[83] Rush SM, Ford LA, Hamilton GA. Morbidity associated with high gastrocnemius recession:
retrospective review of 126 cases. J Foot Ankle Surg 2006;45(3):15660.

10837515/01 $15.00  .00

ADULT ACQUIRED FLATFOOT DEFORMITY

THE ADULT ACQUIRED


FLATFOOT AND SPRING
LIGAMENT COMPLEX
Pathology and Implications for Treatment
Jonathan T. Deland, MD

Treatment for adult acquired atfoot secondary to posterior tibial


tendon (PTT) insufciency rst centered on the tendon. It soon was
noted that reconstruction of the tendon did not correct the deformity,
and in the laboratory, loss of the function of the tendon by itself did not
cause deformity.2 Because the deformity was not secondary to arthritic
change or bone deformation, a reasonable conclusion was that the deformity resulted from ligament failure.
The ligament most commonly implicated and one that was noted
to be attenuated or torn is the spring ligament. It is not clear, however,
that direct repair of the spring ligament adds to correction of deformity;
Reasons for this include poor tissue, excessive stress on the repair, and
involvement of other ligaments. To treat patients better at present and
to develop better treatment in the future, it is important to understand
the ligamentous involvement in adult acquired atfoot. This article
reviews the anatomy of the spring ligament complex, the extent of
ligament involvement in addition to the spring ligament, and implications for clinical reconstruction.
ANATOMY OF THE SPRING LIGAMENT COMPLEX
Authors have noted gross tears or attenuation in the spring ligament
when performing surgery on patients with adult acquired atfoot secFrom the Hospital for Special Surgery, New York, New York

FOOT AND ANKLE CLINICS


VOLUME 6 NUMBER 1 MARCH 2001

129

130

DELAND

ondary to PTT insufciency.5 Before dening the frequency of involvement of this ligament in adult acquired atfoot, it is important to have
a clear understanding of its anatomy and proper nomenclature. The
term spring ligament has been used by authors to include different
ligaments at the medial and inferior portion of the talonavicular joint.1, 5
These 2 ligaments span from the calcaneus to the navicular bone (Fig.
1). The superomedial calcaneonavicular ligament lies directly medial to
the PTT and blends in with the deltoid ligament. This ligament is
wider and stronger than the inferior calcaneonavicular ligament. The
superomedial ligament has direct attachments to the PTT, and it is this
ligament that is seen most easily at surgery and that has been noted to
have gross tears. When describing tears of the spring ligament, clinicians
are referring most often to the superomedial calcaneonavicular ligament.
The term spring ligament can be confusing because various authors,
usually clinicians, have used it to refer to the superomedial ligament,
whereas others, particularly in descriptions of anatomy, have used it to
refer to the other portion, the inferior calcaneonavicular ligament. To
avoid confusion, it is best to refer specically to the superomedial
calcaneonavicular ligament (sometimes called the medial talonavicular
capsule, which is incorporated in the ligament) or the inferior/plantar
calcaneonavicular ligament.1 When including both structures, as both of

Figure 1. A spring ligament complex from a dorsal view, talar head removed. Two parts of
the complex include the superomedial calcaneonavicular ligament and the inferior calcaneonavicular ligament.

THE ADULT ACQUIRED FLATFOOT AND SPRING LIGAMENT COMPLEX

131

these structures do provide support to the plantar/medial aspect of the


talonavicular joint, it is best to use the term spring ligament complex.
The superomedial calcaneonavicular ligament has several distinguishing features. It is larger than the plantar calcaneonavicular ligament. It lies directly medial to the PTT, and in contrast to the plantar
calcaneonavicular ligament, there is brocartilage in the ligament. The
presence of brocartilage probably indicates that this ligament undergoes not only tension, but also compression from the nearby PTT and
talar head. This ligament is part of the acetabulum pedis, the bone
and ligament support of the talar head. The plantar calcaneonavicular
ligament, although also part of the acetabulum pedis, is an entirely
brous structure, lying directly underneath the talonavicular joint. Although not seen easily at surgery, tears can occur in this ligament, and
from biomechanical models, the plantar calcaneonavicular ligament and
the superomedial ligament are released to produce the deformity seen
in PTT insufciency.
EXTENT OF LIGAMENTOUS INVOLVEMENT IN ADULT
ACQUIRED FLATFOOT
Gazdag and Cracchiolo5 reported a high incidence of involvement
of the spring ligament in patients with PTT insufciency. When this
spring ligament complex was examined at surgery (they were apparently
documenting pathology in the superomedial calcaneonavicular ligament), most patients (18 of 22) had tears or attenuation. In patients with
deformity at the talonavicular joint, this is not surprising. The spring
ligament complex is a direct support to the medial/plantar aspect of the
talar head. Various biomechanical models have been used to simulate
the atfoot deformity seen in adult acquired atfoot. Deland et al2
simulated the atfoot by releasing the spring ligament complex as well
as the medial subtalar joint capsule, the plantar ligaments, and the
plantar fascia. This release produces a large amount of deformity with a
lateral rst talometatarsal angle of 100. As documented in a more recent
MR imaging study, however, the plantar fascia and long and short
plantar ligaments rarely are involved in adult acquired atfoot.4 Releasing the spring complex alone did not produce signicant deformity.
Release of this ligamentous complex and cyclically loading the foot does
produce deformity, however. Secondary restraints are therefore likely to
be involved.
With many ligaments supporting the arch and different degrees of
deformity, there is likely to be more than one secondary restraint. From
the same MR imaging study, structures next most frequently involved
after the spring ligament complex were the talocalcaneal interosseous
ligament and the anterior deltoid.4 The understanding and treatment of
PTT insufciency should take into account all these structures, all of
which are medially situated. The spring ligament complex followed by
cyclically loading the foot produces deformity and is the best biomechan-

132

DELAND

ical model for the deformity at present. The spring ligament complex
has been documented to have a high frequency of involvement in adult
acquired atfoot secondary to PTT insufciency, being involved in most
patients. The spring ligament complex is the key ligament complex to
try to reconstruct.
REPAIR AND RECONSTRUCTION OF THE SPRING
LIGAMENT COMPLEX
Surgeons have incorporated direct repair of the spring ligament in
treatment of patients with PTT insufciency. Myerson et al7 described
this repair as part of their procedure along with calcaneal osteotomy. No
one has shown that direct repair of the spring ligament changed the
outcome of results of treatment, however. This lack of change in outcome
could be owing to degenerated ligament tissue, which could not hold
the repair, and stress on the repair. Because not only tear, but also
degeneration of the ligament is noted in most patients and because other
ligaments are likely to be involved, it is not surprising that direct repair
alone of the spring ligament is inadequate. Reconstruction using good
tissue and resolving deformity (which could place too much stress on
the reconstruction or be part of failure of other ligaments) would be
necessary for a successful result. Deland et al2 described the reconstruction of the spring ligament using the anterior deltoid. This reconstruction
was attempted but was not successful clinically. Reconstruction using a
drill hole in the calcaneus and tendon graft was described, but the
clinical results were not reported. Thorardson et al11 used a similar drill
hole in a biomechanical study but took the peroneus longus tendon left
attached distally as the graft. This procedure was successful in the
laboratory but not reported on clinically. A successful reconstruction of
the spring ligament complex shown to be important in clinical results
has not been documented.
The author has done reconstructions of the spring ligament complex
with correction of the arch using tendon graft (unpublished data). To
show that this technique is a possibility, a patient is described. It is
premature to recommend this technique. The patient presented with a
severe atfoot deformity (Fig. 2) with rupture of the spring ligament
complex noted on preoperative MR imaging and at surgery. The most
severely involved part of the complex was the superomedial calcaneonavicular ligament. This patient underwent a posterior calcaneal osteotomy and reconstruction using the peroneus longus and a calcaneal drill
hole. The postoperative radiographs are presented in Figure 3. At 2-year
follow-up, the correction gained has remained excellent as well as the
clinical results. Although the correction achieved was more than expected from a medial slide calcaneal osteotomy, at present it is not
possible to ascertain how much correction was owing to the spring
ligament reconstruction and how much was owing to the osteotomy
itself. This case shows that use of the peroneus longus could be per-

THE ADULT ACQUIRED FLATFOOT AND SPRING LIGAMENT COMPLEX

133

Figure 2. Preoperative radiograph of a patient with deformity associated with spring


ligament complex rupture.

formed with little functional decit and it is possible to add good tissue
to the spring ligament complex with good correction without a lateral
column lengthening at or near the calcaneocuboid joint. A more anatomic reconstruction of considerable deformity secondary to failure of
the spring ligament complex may be possible without the lateral column
lengthening at or near the calcaneocuboid joint.

Figure 3. Postoperative radiograph following calcaneal osteotomy and spring ligament


reconstruction using peroneus longus. The calcaneal drill hole is no longer visible, a staple
is used to x the peroneus longus to the lateral calcaneus.

134

DELAND

PRESENT TREATMENT AND CONCLUSION


With the knowledge gained from studying the spring ligament
complex and the ligamentous failure in adult acquired atfoot, how
should clinicians treat patients at present? Patients with progressive
deformity should be treated early before progression to later stages if
possible. Patients with late stage II adult acquired atfoot (patients with
exible but severe deformity, including considerable abduction of the
talonavicular joint) are likely to have extensive ligamentous involvement
without correction of deformity by a simple medial slide calcaneal osteotomy. Because of the limited ability to correct considerable deformity
without lateral column lengthening or fusion, it would be best to treat
such patients before they advance to this stage.3 If they could be treated
when a medial slide osteotomy as well as reconstruction of the tendon
would give a good clinical result, a foot with good motion and adequate
function is likely.
It is logical that the medial slide calcaneal osteotomy would relieve
strain on the medial ligament support, particularly the spring ligament
complex. There is laboratory evidence to support this.8 Use of the medial
slide calcaneal osteotomy to unload the spring ligament complex and
other medial ligaments to give the patient the best clinical results (because reconstruction of these ligaments is still problematic) is an attractive treatment. It is also apparent that clinicians must be open to the
development of new treatments in PTT insufciency. Various authors
describe different procedures when treating patients in late stage II
disease. These treatments can range from a lateral column lengthening
near the calcaneocuboid joint, a lengthening fusion of that joint, or
fusion of the subtalar joint. Although each treatment may have its
advantages, none of them is anatomic, and surgeons are not unanimous
as to which is the best approach.9, 10 The author prefers lateral column
lengthening near the calcaneocuboid joint combined with a medial slide
to minimize the amount of lengthening needed. Better still would be an
anatomic stable ligament reconstruction if it could be shown to be
consistently successful. Authors have not yet reached anatomic reconstruction of these patients,9 and the clinician should be aware of this
fact. The deformity in adult acquired atfoot is progressive, from ligament failure, with the present treatment indirect and best done early.

References
1. Davis WH, Sobel M, DiCarlo EF, et al: Gross, histological and microvascular anatomy
and biomechanical testing of the spring ligament complex. Foot Ankle Int 17:95102,
1996
2. Deland JT, Annoczky S, Thompson FM: Adult acquired atfoot deformity at the
talonavicular joint: Reconstruction of the spring ligament in an in vitro model. Foot
Ankle Int 13:327332, 1992
3. Deland JT, Sung IH, Page A: Functional results of posterior tibial tendon insufciency.

THE ADULT ACQUIRED FLATFOOT AND SPRING LIGAMENT COMPLEX

4.
5.
6.
7.
8.
9.
10.
11.

135

Presented at Summer Meeting, American Orthopaedic Foot and Ankle Society, Puerto
Rico, 1999
Deland JT, Sung IH, Potter H: Posterior tibial tendon dysfunction: Which ligaments
are involved? Presented at Summer Meeting, American Orthopaedic Foot and Ankle
Society, Puerto Rico, 1999
Gazdag AE, Cracchiolo A: Rupture of the posterior tibial tendon: Evaluation of injury
of the spring ligament and clinical assessment of tendon transfer and ligament repair.
J Bone Joint Surg Am 79:675681, 1997
Mann RA, Thompson FM: Rupture of the posterior tibial tendon causing atfoot:
Surgical treatment. J Bone Joint Surg Am 67:556561, 1985
Myerson MS, Corrigan J, Thompson F, et al: Tendon transfer combined with calcaneal
osteotomy for posterior tibial tendon insufciency: A radiological investigation. Foot
Ankle Int 16:712718, 1995
Otis JC, Deland JT, Kenneally SM, et al: Medial arch strain following medial displacement calcaneal osteotomy: An in vitro study. Foot Ankle Int 20:222226, 1999
Pomeroy GC, Pike RH, Beals TC, et al: Current concepts review: Adult acquired
atfoot in adults due to dysfunction of the posterior tibial tendon. J Bone Joint Surg
Am 81:173182, 1999
Pomeroy GC, Manoli A II: A new operative approach for atfoot secondary to posterior
tibial tendon insufciency: A preliminary report. Foot Ankle Int 18:206212, 1997
Thorardson DB, Schmotzer H, Chon J: Reconstruction with tenodesis in an adult
atfoot model: A biomechanical evaluation of four methods. J Bone Joint Surg Am
77:15571564, 1995
Address reprint requests to
Jonathan T. Deland, MD
Hospital for Special Surgery
535 East 70th Street
New York, NY 10021

ADULT ACQUIRED FLATFOOT DEFORMITY

10837515/01 $15.00  .00

TREATMENT OF STAGE 3 ADULT


ACQUIRED FLATFOOT
Ian Peter Kelly, MCh, and Mark E. Easley, MD

In 1953, rupture of the posterior tibial tendon (PTT) was described


by Key34; in 1974, Goldner et al25 described the progressive talipes
equinovalgus deformity that is associated with trauma or degeneration
of the tendon and the medial plantar ligaments. Johnson and Strom32
recognized that PTT degeneration occurs as a continuum of disease and
introduced a 3-stage classication system in 1989. Stage 1 is represented
by tendinitis and early tendon degeneration. Stage 2 tendon insufciency
is associated with a passively correctable hindfoot valgus and forefoot
abduction deformity. Stage 3, which is the focus of this article, occurs
when chronic PTT insufciency results in xed hindfoot valgus and/or
xed forefoot abduction and supination.
PATHOPHYSIOLOGY
The PTT passes posterior to the ankle axis and medial to the subtalar
axis. Its powerful muscle has an excursion of only 2 cm; even a minor
increase in tendon length results in a major reduction in muscle power.60
The muscle functions to plantarex the ankle, invert the subtalar joint,
and dynamically maintain the medial longitudinal arch. Subtalar inversion locks the midfoot into a stable position for push-off. In the absence
of dynamic PTT support, subtalar inversion fails to occur, and the
transverse tarsal joint remains parallel or unlocked, rendering the forefoot
supple at push-off.19 During gait, gastrocnemius-soleus complex contrac-

From the Division of Orthopaedic Surgery, Department of Surgery, Duke University


Medical Center, Durham, North Carolina

FOOT AND ANKLE CLINICS


VOLUME 6 NUMBER 1 MARCH 2001

153

154

KELLY & EASLEY

tion results in excessive and abnormal stresses on the midtarsal and


subtalar joints, weakening the normal static ligamentous restraints of
the hindfoot. Over time, hindfoot valgus, forefoot abduction, and compensatory forefoot supination progressively worsen to create a xed
hindfoot deformity (that is not passively correctable).
CLINICAL EXAMINATION
Although stage 1 and 2 disease often present with medial ankle and
arch pain, the rigid stage 3 deformity presents differently. Symptoms
often migrate to the lateral side of the foot secondary to calcaneobular
impingement, lateral malleolar stress fracture, or intertarsal degenerative
arthrosis. Johnson and Strom32 described pain emanating from the sinus
tarsi, secondary to impingement of the anterior margin of the posterior
facet of the talus, as it projects inferiorly and impinges on the superior
aspect of the calcaneus.
A complete physical examination necessitates exposure of both
lower limbs and observation of the patient while walking and standing.
Clinical examination should include assessment of the knees because
symptomatic knee deformity merits correction before any hindfoot intervention. Normal pedal pulses and sensation should be conrmed. Hindfoot alignment is viewed best from behind the patient. In stage 3 disease,
the heel is in valgus, and more toes are visible on the lateral side of the
affected foot (too many toes sign).32 The medial longitudinal arch is attened, and the talar head often is prominent medially. In more advanced
deformity, there may be a callus on the medial arch, secondary to
abnormal loading on a plantar-exed and unsupported talar head. During double-limb heel rise, the normal heel moves into varus, while the
affected heel remains in valgus. The patient has great difculty performing a single-leg heel rise on the affected side. The patient should sit
with the legs hanging over the side of the bed. The clinician tests for an
inability to supinate the forefoot fully (supination lag) with attempted
active inversion in plantar exion.1 In stage 3 disease, motion is reduced
or absent in one or all of the hindfoot joints, and these joint are irreducible (the deformity is xed). It is possible to have a xed deformity of
one joint, while the other joints remain reducible. In this situation,
limited fusions, with preservation of the mobile joints, may be feasible.
A complete examination of adjacent joints is necessary to rule out other
sources contributing to the deformity, such as hypermobility of the
rst ray.
RADIOGRAPHIC EVALUATION
Although the diagnosis and staging of PTT insufciency is based
on clinical examination,50 radiographs are useful to quantify the degree
of deformity and to rule out associated pathology. The radiographic

TREATMENT OF STAGE 3 ADULT ACQUIRED FLATFOOT

155

evaluation should include bilateral weight-bearing anteroposterior and


lateral views of the foot and ankle. A variety of angular and linear
methods of measurement have been described.10, 50, 55 The anteroposterior
view of the ankle is obtained to rule out valgus tilt of the talus (stage
4), which may not be apparent clinically,15 and to delineate subbular
impingement. On the anteroposterior view of the foot, the talar coverage
angle (talar head coverage by the navicular bone), the talometatarsal
angle, and the talocalcaneal angle can be measured. The anteroposterior
talocalcaneal angle is difcult to determine because of overlapping bone
and reportedly is unreliable.54, 59 On the lateral weight-bearing view of
the foot, the lateral talocalcaneal pitch angle, lateral talometatarsal angle,
height of the medial cuneiform from the ground, and talonavicular sag
are noted.5 In stage 3 disease, hindfoot valgus and forefoot abduction
are seen, and secondary degenerative changes in the talar complex may
be noted. Gapping at the naviculocuneiform and tarsometatarsal joints
may be visualized on the plantar surface. There is no role for MR
imaging in the presence of xed deformity.13, 20, 44
Postoperative radiographs can be useful but may be difcult to
measure accurately and may not correlate with the clinical correction.8
Sangeorzan et al54 stressed the importance of objective radiographic
methods after arthrodesis and reported an improvement of the lateral
talocalcaneal angle of 11, the lateral talometatarsal angle of 17, and
the anteroposterior talometatarsal angle of 18 after triple arthrodeses.
Cracchiolo et al15 found the lateral talometatarsal angle was the most
effective measurement to assess clinical correction after triple arthrodesis. In a review of 132 triple arthrodeses performed for a variety of
pathologic conditions by Pell et al,49 the PTT group required the greatest
amount of radiographic correction.
NONOPERATIVE MANAGEMENT
A trial of anti-inammatory medications, casting, and orthoses can
be attempted for 3 to 6 months; however, the chance of success in
treating lateral foot pain secondary to xed hindfoot valgus nonoperatively is low. Any orthoses must be accommodative rather than corrective,64 and orthoses can be useful in patients who decline surgery or
have signicant medical comorbidities. The aim of an orthosis is to
accommodate the xed deformity, correct any correctable elements of
the deformity, and control pain. A block to further forefoot abduction
helps maintain the medial longitudinal arch and is tolerated better than
a medial arch support. Fixed deformity and bone prominences tend not
to tolerate orthoses well. Bracing options include a molded ankle foot
orthosis, an Arizona brace (Ernesto G. Castro Custom Footwear Inc,
Messa, AZ), and a Marzano brace (Yankee Bionics Inc, Akron, OH). Chao
et al11 showed that despite the progressive nature of the deformity, an
orthosis can arrest progression and provide symptomatic relief in sedentary elderly patients; however, it is likely that the degenerative changes
will continue because of the abnormal limb alignment. Steroid injections

156

KELLY & EASLEY

to the tendon should be avoided because the disease is degenerative


and not inammatory.43 In stage 3 disease, some authors suggest surgical
intervention as being the best form of conservative management.38
OPERATIVE TREATMENT
Fixed deformity usually is associated with symptomatic painful
arthrosis. The 2 treatment goals are rst to eliminate joint pain and
second to correct deformity in the hope of limiting progressive arthrosis
in contiguous joints. Corrective fusions are the treatment of choice in
stage 3 disease. Arthrodesis options are triple, double, or isolated fusions. A triple arthrodesis includes the subtalar, talonavicular, and calcaneocuboid joints. Double fusion involves the talonavicular and calcaneocuboid joints. Isolated subtalar and talonavicular fusions also have been
described. The fusion chosen must be able to fully correct all xed
deformities evident on clinical examination.7, 12
The concept of limited fusions, as opposed to the triple arthrodesis,
has evolved for 2 reasons. First, limited fusions can be sufcient to
correct the deformity and relieve pain, and second, they preserve motion
in the surrounding joints. Correction of the deformity ensures more
physiologic loading of the contiguous joints, whereas preservation of
motion limits the magnitude of the load on the joints through improved
shock absorption, decreasing the risk of progressive arthrosis. There are
some problems with the concept of limited fusions, however, and one
needs to be aware that the various limited fusions differ in their ability
to correct the atfoot deformity and at the same time preserve
motion.17, 47 First, fusing one component of the subtalar complex results
in a signicant stiffening of the other components.6 The actual motion
preserved for shock absorption may be little. Second, with the preserved
motion, the incidence of symptomatic nonunion is considerably higher.
Third, there have been no comparative studies proving a reduced rate
of symptomatic arthrosis with limited versus triple arthrodeses. Triple
arthrodesis is the gold standard in the treatment of stage 3 PTT dysfunction.
Triple Arthrodesis
A triple arthrodesis is the fusion of the subtalar, talonavicular, and
calcaneocuboid joints. The initial descriptions of this procedure through
medial and lateral incisions were by Davis16 and Ryerson53 in the pediatric population, and currently triple arthrodesis is the gold standard for
the treatment of stage 3 adult acquired atfoot. It is a salvage procedure
that is necessary only in the presence of painful, xed deformity and
arthrosis that is refractory to nonoperative management. The goals are
to reduce pain; correct the deformity; establish a plantigrade, stable,
weight-bearing foot; maintain the integrity of the adjacent joints, espe-

TREATMENT OF STAGE 3 ADULT ACQUIRED FLATFOOT

157

cially the ankle; and avoid incisional neuromas.45 The technical difculty
and potential complications of malunion, nonunion, accelerated arthrosis, and osteonecrosis of the talus have been emphasized.4, 26, 27, 33, 65
Likewise, residual discomfort and tarsometatarsal and ankle arthrosis
can be expected.63 With better understanding, minimal bone resection,
and use of rigid internal xation to achieve compression arthrodesis,
however, excellent overall results and high rates of patient satisfaction
are attainable.8, 23, 26, 54, 57
Operative Technique
The operation has been well described, and it is emphasized that a
meticulous operative technique is mandatory for success.3, 8, 23, 26, 38, 45, 54,
59, 63
Historically, correction of deformity was facilitated by resection of
large bone wedges. Currently the emphasis is on minimal bone resection,
preservation of subchondral bone, and deformity correction through
joint reduction.
The patient is positioned supine on the operating table, with a
bump under the ipsilateral buttock and thigh tourniquet applied to the
leg. To examine for an Achilles tendon contracture, the subtalar joint is
held in neutral with the knee exed to 30, and the ankle is passively
dorsiexed. If the foot remains in more than 5 of equinus, a percutaneous Achilles tendon lengthening should be considered.4, 49 One lateral
and 2 medial stab incisions, each involving 50% of the Achilles tendon,
are performed, and the tendon is lengthened in continuity. The triple
arthrodesis is performed through medial and lateral incisions, to improve visualization and to reduce the risk of neural injury, but either
incision can be performed rst. The lateral incision extends from the tip
of the lateral malleolus toward the base of the fourth metatarsal, parallel
with the peroneal tendons. This incision aims to be plantar to the
intermediate branch of the supercial peroneal nerve and dorsal to the
sural nerve; however, these nerves or communicating branches can be
encountered, especially more distally in the wound. The skin incision is
deepened through the inferior extensor retinaculum onto the extensor
digitorum brevis muscle, which is lifted subperiosteally from the calcaneus and retracted superiorly out of the wound; this may preserve the
muscles innervation from the deep peroneal nerve. A tagging suture is
useful to hold the muscle belly out of the operative eld. The contents
of the sinus tarsi are excised to expose the posterior facet of the subtalar
joint. The initial laminar spreader inverts the subtalar joint, which is
denuded of articular cartilage using sharp periosteal elevators and a
variety of laminar spreaders. The use of a curet as one progresses
medially is recommended to avoid injury to the exor hallucis longus
tendon. The subchondral bone of the posterior, middle, and anterior
facets can be perforated, but preservation of the anatomy assists the
reduction. The calcaneocuboid joint is exposed by capsular release and
adduction of the forefoot, and the joint is denuded of articular cartilage.
The technique of lateral column lengthening through the calcaneocuboid

158

KELLY & EASLEY

joint in conjunction with a triple arthrodesis has been reported for the
severe planovalgus deformity.30 The lateral aspect of the talonavicular
joint also is accessible through this incision. This point of 4-bone contact
(quadruple arthrodesis) is important because if osseous bridging can be
obtained at this point, the foot is stable, even if in the presence of
nonunion at one of the triple sites.31
The medial incision is between the anterior and posterior tibial
tendons, avoiding the saphenous nerve and vein, and extends from the
ankle joint to the naviculocuneiform joint. A curved instrument is helpful to open the talonavicular joint for exposure, and all cartilage is
denuded, while preserving the normal contour of the joint.
The deformity usually is correctable without bone wedges being
resected. The sequence in which the joints are reduced is not as important as the accuracy of the reduction. The talonavicular joint can be
reduced rst,49 but the authors opt rst to align the subtalar joint in 5
of valgus using a cannulated screw system. Position is conrmed clinically and in 2 planes using image intensication. The talonavicular joint
is then reduced, ensuring full coverage of the talar head and, most
importantly, correction of the forefoot supination deformity. Finally the
calcaneocuboid joint is stabilized by passing a screw from the base of
the anterior process of the calcaneus into the cuboid. Autograft or
allograft can be used to improve the likelihood of fusion. We pack any
gaps at the joint surfaces and the sinus tarsi with corticocancellous
allograft, but the need for graft is questionable.
Results
Long-term results of triple arthrodesis have been reported in children2, 4, 18, 46, 62, 65 and adults.8, 15, 22, 23, 26, 49, 54, 59 The studies in adults often
involved multiple disease processes, with rheumatoid arthritis and PTT
disease being the most common indications. Pell et al49 reported the
outcome after primary triple arthrodesis in 111 patients (132 feet), 63 of
whom had PTT dysfunction, at an average follow-up of 5.7 years.
The overall group reported high postoperative satisfaction rates and
improvements in the postoperative modied American Orthopaedic
Foot and Ankle Society score. There was no statistically signicant
difference between the PTT group and the other diagnostic groups
(posttraumatic, rheumatoid arthritis, neuromuscular, and miscellaneous)
with respect to the patients satisfaction or hindfoot scores. Fortin and
Walling23 reported results of triple arthrodesis in PTT dysfunction in 25
adult patients with 32 fused hindfeet, all with stage 3 or 4 disease. The
average follow-up in this retrospective review was 4.3 years postoperatively. Iliac crest bone graft was used in 25 feet and allograft in 7 feet.
The American Orthopaedic Foot and Ankle Society scores improved on
average 36 points, and 24 of 25 patients were satised with the surgery.
There were 1 nonunion and 2 varus malunions needing revision surgery.
Two patients had stage 4 disease (and were offered a pantalar arthrodesis), and one patient had progression of ankle symptoms, necessitating

TREATMENT OF STAGE 3 ADULT ACQUIRED FLATFOOT

159

a subsequent ankle fusion. One patient had progression of naviculocuneiform sag that was asymptomatic, and 2 patients needed removal of
the calcaneal screw. Twenty-two patients rated their surgery as a success,
2 patients rated their surgery as a success with reservations, and 1
patient was dissatised.
Graves et al26 reported 17 patients after triple arthrodesis, 10 of
whom had PTT dysfunction. In the PTT group, 9 patients had reduced
pain, and 8 patients were satised. Dissatisfaction with the procedure
was because of one painful nonunion of the talonavicular joint and
progression of ankle arthrosis in a second case. The average time to
pain relief was 10 months. These investigators found no loss of ankle
dorsiexion and little loss of plantar exion postoperatively. Sangeorzan
et al54 reviewed 40 adult patients with 44 fused feet, on average 4.9 years
after triple arthrodesis. Nine feet were fused for PTT rupture with
hindfoot arthritis, but they were not reported as a separate group. All
operations were performed using rigid internal xation with screws. Of
44 feet, 34 had a good result, 6 had a fair result, and 4 had failures. The
failures were 2 nonunions (talonavicular and calcaneocuboid), which
united with further surgery, and 2 malunions, one needing a calcaneal
osteotomy. Most patients could function painlessly on at surfaces and
had only occasional mild pain with more vigorous activities. Graves et
al26 concluded that when carefully performed, triple arthrodesis can
restore an adult patient with a disabling foot disorder to normal function
and minimal pain. Bennett et al8 retrospectively reviewed 22 adult patients who underwent a triple arthrodesis because of hindfoot pain or
deformity (or both), with a minimum 3-year follow-up. The 8 patients
with PTT dysfunction were not reported separately. Patient satisfaction
with the procedure was high, and 21 of 22 (95%) patients believed
that they were improved overall and would have the surgery again if
indicated.
Discussion
Triple arthrodesis is a technically demanding procedure that can
improve pain and function and produce a stable, plantigrade foot. With
improved operative techniques and instrumentation and better awareness of the complications, good results can be achieved. Patients must
be warned of the prolonged recovery period, however, and that a triple
arthrodesis does not cure their problem but is a good salvage operation
for end-stage disease. The main complications of the triple arthrodesis
are accelerated arthrosis in surrounding mobile joints, nonunion, malunion, and osteonecrosis of the talus.
Accelerated arthrosis of the ankle and midfoot joints usually is
asymptomatic but can leave residual postoperative pain. Patients must
be forewarned that pain may not be resolved completely.4, 26, 63 Radiographic evidence of accelerated ankle and midfoot arthrosis has been
reported in greater than 50% of patients.4, 8, 15, 26, 28, 49, 65 The reasons are
thought to be increased motion demands, increased load demands, loss

160

KELLY & EASLEY

of shock absorption, and abnormal joint alignment.4, 8, 26, 49, 54 Lengthening


of the Achilles tendon to enable more ankle dorsiexion and consequently reducing the load on the midfoot may be important in limiting
the progression of arthrosis.23 If symptomatic, bracing occasionally is
successful; however, further fusion may be necessary.
Nonunion has been reported in more than 20% of cases in which
inconsistent means of internal xation were employed.4 Wilson et al66
investigated the reasons for nonunion and concluded that absence of
bone contact in one or more joints was predictive of failure. Currently,
with compression arthrodesis and the use of the 2 incisions for better
visualization as well as improved instrumentation and image intensication, high fusion rates can be expected.9, 26, 38, 54, 56, 58 Currently, nonunion
rates of 4.5% have been reported,54 and the need for bone graft is
debatable.15, 59 The talonavicular joint is the most common site for nonunion, but this often can remain asymptomatic.8 If necessary, further
fusion with compression and bone grafting generally is successful.
Although the risk of nonunion can be reduced with internal xation,
close attention to detail intraoperatively is the primary cure for malunion. Complete correction of all deformities is difcult and one of the
main reasons why this procedure is technically demanding. Incomplete
correction or in situ fusion causes abnormal joint loading, leading to a
more rapid progression of arthrosis in the adjacent joints.23 Residual heel
valgus results in valgus tilting of the talus in the ankle mortise, with
accelerated lateral ankle joint wear.26, 51, 62 Likewise, overcorrection with
varus malalignment of the heel is tolerated poorly. Malalignment can
result despite attention to detail. The deformity most often forgotten is
forefoot supination, with consequent weight bearing on the lateral column. This deformity can be secondary to hypermobility of the rst ray
or incomplete derotation of the forefoot through the talonavicular joint.
This deformity is difcult to quantify preoperatively and is seen most
easily after correction of forefoot abduction and hindfoot valgus intraoperatively. Excessive forefoot varus causes lateral overload with pain
under the fth metatarsal and cuboid, lateral ankle instability, and ankle
arthrosis. The results of revision surgery for correction of malunion after
a triple arthrodesis have been reported.27 This complex revision surgery
is best avoided by close attention to detail at the time of the index surgery.
Complete talar dome osteonecrosis after triple arthrodesis has been
described with rates as high as 24%.4, 18, 36, 42 Previous techniques with
aggressive talar neck dissection and resection lead to a considerable
incidence of talar osteonecrosis. With present-day techniques, triple arthrodesis does not necessitate aggressive talar neck dissection, and the
incidence of complete talar dome osteonecrosis is less.54 Segmental talar
dome osteonecrosis can exist and appear similar to progressive ankle
arthrosis. The true incidence of segmental osteonecrosis may be underestimated. Jones and Nunley33 reported 3 cases of segmental osteonecrosis
of the lateral aspect of the talar dome after triple arthrodesis. The lateral
talar dome is supplied by the artery of the sinus tarsi from the peroneal

TREATMENT OF STAGE 3 ADULT ACQUIRED FLATFOOT

161

artery.24 This artery is violated during the lateral approach to the posterior facet. In the Jones and Nunley study,33 the arthrodeses united within
3 to 4 months, but the patients presented with ankle pain and bular
stress fractures 5 to 12 months postoperatively. There was radiologic
evidence of lateral dome osteonecrosis and valgus tilting of the talus at
that time. One patient needed a tibiotalar fusion, whereas the other 2
were managed with orthoses. The authors hypothesized that the talarto-calcaneal direction of the compression screw may have contributed to
the development of the osteonecrosis. Concomitant risk factors in all 3
patients in that study included rheumatoid arthritis and oral steroids.
The method of screw placement for subtalar fusion has been described.41
The screw may be directed from the talar neck into the calcaneus or
from the tuberosity of the calcaneus into the talus. The posterior-toanterior screw seems preferable in terms of avoiding potential osteonecrosis.33, 54
Double Arthrodesis
Double arthrodesis is the simultaneous fusion of the talonavicular
and calcaneocuboid joints. The indications are xed forefoot abduction
and xed forefoot varus greater than 10 to 1539 in patients with a
passively correctable and painless subtalar joint. In older patients with
the same clinical ndings, one might consider an isolated talonavicular
fusion. The addition of the calcaneocuboid fusion imparts further stability to the talonavicular arthrodesis and is more appropriate in younger,
active individuals. It also eases exposure of the lateral aspect of the
talonavicular joint and enables symmetric shortening of both columns
of the foot for the compression arthrodesis. The calcaneocuboid fusion
does not reduce hindfoot motion, which had been eliminated already by
the talonavicular fusion. Lateral midfoot pain is a common complaint
after isolated talonavicular fusion but was encountered less frequently
after the addition of a calcaneocuboid fusion.39
Mann and Beaman39 compared the results of double fusion in PTT
disease (n  16) with double fusion in patients with other pathology
(n  8). There were considerable improvements in pain and function in
both groups. Complications were more frequent in the PTT group and
included 4 of 16 patients with a talonavicular nonunion, 3 of which
needed revision surgery (1 repeat talonavicular fusion and 2 triple arthrodeses). The 25% talonavicular nonunion rate was high and seen in
patients with a greater degree of deformity and when staples were used
to stabilize the talonavicular joint. The prolonged recovery rate was the
biggest disadvantage from the patient viewpoint. Clain and Baxter12
reported good results of the double fusion in 16 feet, 7 of which had
PTT insufciency. These investigators encountered one talonavicular
nonunion, progressive arthrosis of the ankle in 6 of 16 feet, and progressive arthrosis of the naviculocuneiform joints in 7 of 16 feet. They
stressed the importance of lengthening the Achilles tendon if the ankle

162

KELLY & EASLEY

does not dorsiex, to reduce the stress imparted on the surrounding


joints. They did not advocate double fusion in symptomatic exible
atfeet, suggesting that their patient group had stage 3 disease. There
are few long-term reports of double fusion for PTT insufciency in the
literature, and patient selection is paramount.
Isolated Subtalar Fusion
Isolated subtalar fusion is indicated only when the subtalar joint
has a xed deformity while the transverse tarsal joint remains correctable
and pain-free. On clinical examination, forefoot abduction and varus are
fully reducible, and the transverse tarsal joint is neither hypermobile nor
painful secondary to arthrosis. Isolated subtalar arthrodesis is indicated
in a select group of patients that are not commonly seen.
The operative technique has been well described, and one can
expect a high fusion rate with a low complication rate.21, 35, 40, 57 One must
aim for correction of the hindfoot deformity rather than perform a fusion
in situ.21 The precise position of fusion is essential. Subtalar varus is
tolerated poorly because it produces a rigid foot with excessive stress
on the lateral border and the lateral ankle ligaments.19 Excessive valgus
can lead to painful calcaneobular impingement, unlocking of the midfoot with consequent collapse, and a poor cosmetic appearance of the
foot. The recommended position for fusion is 5 of valgus. The important
distinction between in situ fusion versus a corrective or repositional arthrodesis was highlighted by Stephens et al.61 These authors discussed the
relative lengthening of the medial column of the foot, which they propose is owing to anterior subluxation of the talus on the calcaneus. They
stress the need to reduce the subtalar joint anatomically in all planes to
correct the pes planus deformity.
The concept of the limited fusion is to enable correction of the
deformity, with maximal preservation of motion in the surrounding
joints. How much motion is preserved after an isolated subtalar fusion?
Mann et al40 fused in situ and found that patients retained 60% of
transverse tarsal motion, 70% of ankle dorsiexion, and 90% of ankle
plantar exion. After a simulated subtalar fusion in a cadaver model,
Astion et al6 showed only a 26% retention of motion at the talonavicular
joint and a 56% retention of motion at the calcaneocuboid joint. In
contrast, Fellmann and Zollinger21 reported the same or an improved
range of transverse tarsal motion after subtalar fusion in 69% of their
patients. Different measurement techniques were used in each report.
OMalley et al47 showed that because the subtalar fusion allows transverse tarsal motion, it was unable to maintain correction of the atfoot
deformity consistently. The key factor that determines residual motion
at the transverse tarsal joint is the position of the fused subtalar joint. A
valgus fusion maintains a supple, mobile transverse tarsal joint, whereas
a varus fusion locks the midfoot.
Mann et al40 reviewed 48 cases of isolated subtalar fusion for a

TREATMENT OF STAGE 3 ADULT ACQUIRED FLATFOOT

163

variety of indications, including 11 cases with PTT dysfunction. Patients


were limited to those with hindfoot valgus without major forefoot deformity. The number of patients with xed valgus (stage 3 disease) was not
stated. One of the 11 patients had a poor result owing to hindfoot valgus
of 12 and progressive talar tilt at the ankle, necessitating an ankle-foot
orthosis. All fusions were performed in situ. There was a 36% and 41%
incidence of mild radiographic progression of arthrosis in the ankle and
transverse tarsal joint. Mann et al40 concluded that isolated subtalar
arthrodesis provided a highly successful result in carefully selected
patients with PTT insufciency.
Kitaoka and Patzer35 reviewed 21 patients after corrective subtalar
arthrodesis for stage 2 PTT dysfunction. The importance of reevaluation
intraoperatively was stressed because the distinction between late stage
2 disease and early stage 3 disease can be difcult. After subtalar
correction, forefoot position must be assessed carefully because there is
always the possibility of proceeding to a triple arthrodesis if xed
forefoot deformity exists.14, 35 Of the 21 patients, 11 continued to have
some pain secondary to preexisting arthrosis of adjacent joints. Patient
selection and intraoperative reevaluation of the forefoot were emphasized. Mangone et al37 reviewed 34 patients, 24 of whom were evaluated
fully. Eleven of the patients had PTT insufciency, and these patients
had excellent subjective and objective results. The stage of the disease
was not described. Fellmann and Zollinger21 reported excellent results
in 36 patients, of whom 12 had PTT insufciency. Neither the Johnson
staging nor the group-specic results were described. Russotti et al52
reported 45 cases of isolated subtalar fusion; 15 were performed for
PTT dysfunction. All patients were satised with the result and united
radiologically, but the stage of the disease was not described. These
investigators reported no radiographic evidence of progressive transverse tarsal arthrosis with 57 months of follow-up.
In PTT insufciency, isolated subtalar fusion can give high satisfaction and high fusion rates, with a low complication rate. The crucial
decision relates to patient selection and staging of the disease because
this is a rare population of patients. The position of the forefoot always
should be reevaluated after reduction of the hindfoot, with the potential
to proceed to a triple arthrodesis. For more meaningful results, future
investigations need to describe the PTT group with greater accuracy,
preferably using the Johnson staging system.
Isolated Talonavicular Fusion
Isolated talonavicular fusion is indicated only when the forefoot
deformity is xed or the talonavicular joint is unstable, in conjunction
with a mobile, painless subtalar joint. These patients are encountered
rarely. Although deformity correction is possible through the talonavicular joint,47 there is little preservation of motion in the surrounding joints
for shock absorption,6 negating one of the advantages of the limited

164

KELLY & EASLEY

fusion. The advantages of an isolated talonavicular fusion are ease


of the operation and its ability to achieve correction of the foot. The
disadvantages are the elimination of hindfoot motion and the high
potential for nonunion. Because of the high nonunion rate, autogenous
bone graft often is harvested, increasing the complexity of the operation.29 This fusion is more appropriate in older patients with a lower
functional demand, whereas the double fusion (talonavicular and calcaneocuboid) should be considered in the younger, more active patient
with the same clinical picture.39, 48
Harper29 reported the results of isolated talonavicular arthrodesis
for acquired atfoot deformity. The number of patients with xed as
opposed to exible transverse tarsal joints was not identied. Of 29
patients, 25 were satised with no or minor reservations and achieved
good or excellent results. These were typically nonathletic, older patients, who were not limited by the loss of motion. Otherwise, there are
few reports in the literature of isolated talonavicular fusion in PTT
insufciency.
SUMMARY
Stage 3 adult acquired atfoot occurs when chronic posterior tibial
tendon insufciency results in xed hindfoot valgus or xed forefoot
abduction and supination. Nonoperative management results in limited
success. Corrective fusion is the treatment of choice. Although a variety
of arthrodeses have been employed, triple arthrodesis remains the gold
standard.
References
1. Abboudi J, Kupcha P: Supination lag as an indication of posterior tibial tendon
dysfunction. Foot Ankle Int 19:570, 1998
2. Adelaar RS, Dannelly EA, Meunier PA, et al: A long term study of triple arthrodesis
in children. Orthop Clin North Am 7:895, 1976
3. Amis J: The Foot And Ankle. Philadelphia, Lippincott-Raven, 1997
4. Angus PD, Cowell HR: Triple arthrodesis: A critical long-term review. J Bone Joint
Surg Br 68:260, 1986
5. Aronson J, Nunley J, Frankovitch K: Lateral talocalcaneal angle in assessment of
subtalar valgus: Follow-up of seventy Grice-Green arthrodeses. Foot Ankle Int 4:56,
1983
6. Astion DJ, Deland JT, Otis JC, et al: Motion of the hindfoot after simulated arthrodesis.
J Bone Joint Surg Am 79:241, 1997
7. Beals TC, Pomeroy GC, Manoli A, 2nd: Posterior tendon insufciency: Diagnosis and
treatment. J Am Acad Orthop Surg 7:112, 1999
8. Bennett GL, Graham CE, Mauldin DM: Triple arthrodesis in adults. Foot Ankle Int
12:138, 1991
9. Carr J, Hansen ST, Benirschke S: Subtalar distraction bone block union for late complications of calcaneus fractures. Foot Ankle Int 9:81, 1988
10. Chadha H, Pomeroy G, Manoli A, 2nd: Radiologic signs of unilateral pes planus. Foot
Ankle Int 18:603, 1997

TREATMENT OF STAGE 3 ADULT ACQUIRED FLATFOOT

165

11. Chao W, Wapner KL, Lee TH, et al: Nonoperative management of posterior tibial
tendon dysfunction. Foot Ankle Int 17:736, 1996
12. Clain MR, Baxter DE: Simultaneous calcaneocuboid and talonavicular fusion: Longterm follow-up study. J Bone Joint Surg Br 76:133, 1994
13. Conti S, Michelson J, Jahss M: Clinical signicance of magnetic resonance imaging in
preoperative planning for reconstruction of posterior tibial tendon ruptures. Foot
Ankle Int 13:208, 1992
14. Conti SF: Posterior tibial tendon problems in athletes. Orthop Clin North Am 25:109,
1994
15. Cracchiolo AD, Pearson S, Kitaoka H, et al: Hindfoot arthrodesis in adults utilizing a
dowel graft technique. Clin Orthop 257:193203, 1990
16. Davis G: Treatment of hollow foot (pes cavus). Am J Orthop Surg 11:231, 1913
17. Deland J, Otis JC, Lee KT, et al: Lateral column lengthening with calcaneocuboid
fusion: Range of motion in the triple joint complex. Foot Ankle Int 16:729, 1995
18. Duncan JW, Lovell WW: Hoke triple arthrodesis. J Bone Joint Surg Am 60:795, 1978
19. Elftman H: Transverse tarsal joint and its control. Clin Orthop 16:41, 1960
20. Feighan J, Towers J, Conti S: The use of magnetic resonance imaging in posterior tibial
tendon dysfunction. Clin Orthop 365:2338, 1999
21. Fellmann J, Zollinger H: Isolated talocalcaneal interposition fusion: A prospective
follow-up study. Foot Ankle Int 18:616, 1997
22. Figgie MP, OMalley MJ, Ranawat C, et al: Triple arthrodesis in rheumatoid arthritis.
Clin Orthop 292:250254, 1993
23. Fortin PT, Walling AK: Triple arthrodesis. Clin Orthop 365:9199, 1999
24. Gelberman RH, Mortensen WW: The arterial anatomy of the talus. Foot Ankle Int
4:64, 1983
25. Goldner JL, Keats PK, Bassett FHD, et al: Progressive talipes equinovalgus due to
trauma or degeneration of the posterior tibial tendon and medial plantar ligaments.
Orthop Clin North Am 5:39, 1974
26. Graves SC, Mann RA, Graves KO: Triple arthrodesis in older adults: Results after
long-term follow-up. J Bone Joint Surg Am 75:355, 1993
27. Haddad SL, Myerson MS, Pell RFT, et al: Clinical and radiographic outcome of revision
surgery for failed triple arthrodesis. Foot Ankle Int 18:489, 1997
28. Haritidis JH, Kirkos JM, Provellegios SM, et al: Long-term results of triple arthrodesis:
42 cases followed for 25 years. Foot Ankle Int 15:548, 1994
29. Harper MC: Talonavicular arthrodesis for the acquired atfoot in the adult. Clin
Orthop 365:6568, 1999
30. Horton GA, Olney BW: Triple arthrodesis with lateral column lengthening for treatment of severe planovalgus deformity. Foot Ankle Int 16:395, 1995
31. Jahss MH: Disorders of the Foot and Ankle: Medical and Surgical Management.
Philadelphia, WB Saunders, 1991
32. Johnson KA, Strom DE: Tibialis posterior tendon dysfunction. Clin Orthop 239:196
206, 1989
33. Jones CK, Nunley JA: Osteonecrosis of the lateral aspect of the talar dome after triple
arthrodesis: A report of three cases. J Bone Joint Surg Am 81:1165, 1999
34. Key J: Partial rupture of the tendon of the posterior tibial muscle. J Bone Joint Surg
Am 35:1006, 1953
35. Kitaoka HB, Patzer GL: Subtalar arthrodesis for posterior tibial tendon dysfunction
and pes planus. Clin Orthop 345:187194, 1997
36. MacKenzie I: Lambrinudis arthrodesis. J Bone Joint Surg Br 41:738, 1959
37. Mangone PG, Fleming LL, Fleming SS, et al: Treatment of acquired adult planovalgus
deformities with subtalar fusion. Clin Orthop 341:106112, 1997
38. Mann R: Surgery of the Foot and Ankle, ed 7. St Louis, Mosby, 1999
39. Mann RA, Beaman DN: Double arthrodesis in the adult. Clin Orthop 365:7480, 1999
40. Mann RA, Beaman DN, Horton GA: Isolated subtalar arthrodesis. Foot Ankle Int
19:511, 1998
41. Manoli A, Beals TC, Hansen ST: Technical factors in hindfoot arthrodesis. Instr Course
Lect 46:347356, 1997
42. Marek FASA: Aseptic necrosis of the astragalus following arthrodesing procedures of
the tarsus. J Bone Joint Surg Am 27:587, 1945

166

KELLY & EASLEY

43. Mosier SM, Lucas DR, Pomeroy G, et al: Pathology of the posterior tibial tendon in
posterior tibial tendon insufciency. Foot Ankle Int 19:520, 1998
44. Myerson M: Current therapy in foot and ankle surgery. St Louis, Mosby-Year Book,
1993
45. Myerson MS: Adult acquired atfoot deformity: Treatment of dysfunction of the
posterior tibial tendon. Instr Course Lect 46:393, 1997
46. Olney BW, Menelaus MB: Triple arthrodesis of the foot in spina bida patients. J Bone
Joint Surg Br 70:234, 1988
47. OMalley MJ, Deland JT, Lee KT: Selective hindfoot arthrodesis for the treatment of
adult acquired atfoot deformity: An in vitro study. Foot Ankle Int 16:411, 1995
48. Pedowitz WJ, Kovatis P: Flatfoot in the adult. J Am Acad Orthop Surg 3(5):293302,
1995
49. Pell RF, Myerson MS, Schon LC: Clinical outcome after primary triple arthrodesis. J
Bone Joint Surg Am 82:47, 2000
50. Pomeroy GC, Pike RH, Beals TC, et al: Acquired atfoot in adults due to dysfunction
of the posterior tibial tendon. J Bone Joint Surg Am 81:1173, 1999
51. Resnick RB, Jahss MH, Choueka J, et al: Deltoid ligament forces after tibialis posterior
tendon rupture: Effects of triple arthrodesis and calcaneal displacement osteotomies
[published erratum appears in Foot Ankle Int 1995; 16(5):314]. Foot Ankle Int 16:14,
1995
52. Russotti GM, Cass JR, Johnson KA: Isolated talocalcaneal arthrodesis: A technique
using moldable bone graft. J Bone Joint Surg Am 70:1472, 1988
53. Ryerson E: Arthrodesing operations on the feet. J Bone Joint Surg 5:453, 1923
54. Sangeorzan B, Smith D, Veith R, et al: Triple arthrodesis using internal xation in
treatment of adult foot disorders. Clin Orthop 294:299, 1993
55. Sangeorzan BJ, Mosca V, Hansen ST Jr: Effect of calcaneal lengthening on relationships
among the hindfoot, midfoot, and forefoot. Foot Ankle 14:136, 1993
56. Scranton P, Fu F, Brown T: Ankle arthrodesis: A comparative clinical and biomechanical
evaluation. Clin Orthop 151:234, 1980
57. Scranton PE Jr: Comparison of open isolated subtalar arthrodesis with autogenous
bone graft versus outpatient arthroscopic subtalar arthrodesis using injectable bone
morphogenic protein-enhanced graft. Foot Ankle Int 20:162, 1999
58. Scranton PE Jr: An overview of ankle arthrodesis. Clin Orthop 268:96101, 1991
59. Scranton PE Jr: Results of arthrodesis of the tarsus: Talocalcaneal, midtarsal, and
subtalar joints. Foot Ankle Int 12:156, 1991
60. Silver RL, de la Garza J, Rang M: The myth of muscle balance: A study of relative
strengths and excursions of normal muscles about the foot and ankle. J Bone Joint
Surg Br 67:432, 1985
61. Stephens HM, Walling AK, Solmen JD, et al: Subtalar repositional arthrodesis for adult
acquired atfoot. Clin Orthop 365:6973, 1999
62. Tenuta J, Shelton YA, Miller F: Long-term follow-up of triple arthrodesis in patients
with cerebral palsy. J Pediatr Orthop 13:713, 1993
63. Wapner KL: Triple arthrodesis in adults. J Am Acad Orthop Surg 6:188, 1998
64. Wapner KL, Chao W: Nonoperative treatment of posterior tibial tendon dysfunction.
Clin Orthop 365:3945, 1999
65. Wetmore RS, Drennan JC: Long-term results of triple arthrodesis in Charcot-MarieTooth disease. J Bone Joint Surg Am 71:417, 1989
66. Wilson F, Fay GF, Lamotte P, et al: Triple arthrodesis: A study of the factors affecting
fusion after three hundred and one procedures. J Bone Joint Surg Am 47:340, 1965
Address reprint requests to
Mark E. Easley, MD
Division of Orthopaedic Surgery
Department of Surgery, Box 2950
Duke University Medical Center
5330 Hospital South
Durham, NC 27710
e-mail: easle004@mc.duke.edu

ADULT ACQUIRED FLATFOOT DEFORMITY

10837515/01 $15.00  .00

TREATMENT OF STAGE 4 ADULT


ACQUIRED FLATFOOT
Ian Peter Kelly, MCh, and James A. Nunley, MD

Although Goldner et al16 often are credited with identifying degeneration of the posterior tibial tendon (PTT) and the medial plantar
ligaments as the cause of acquired adult atfoot in their classic article in
1974, understanding of PTT insufciency was enhanced greatly by Johnson and Strom,24 who in 1989 described 3 clinical stages of this disease.
In stage 1, there is degeneration without elongation of the tendon and no
clinical deformity; in stage 2, there is tendon elongation and correctable
deformity; and in stage 3, the deformity is no longer correctable. Although Johnson and Strom24 did allude to probable stage 4 disease,
which they described as a hindfoot that is xed in eversion, producing
valgus tilting of the talus within the ankle mortise and lateral tibiotalar
degeneration, Myerson39 is credited with the rst description of stage 4
PTT degeneration. Stage 4 PTT degeneration is not encountered commonly, and there is little published literature related to the available
treatment options and long-term outcome.

DEFINITION
For the purpose of this article, stage 4 PTT dysfunction is dened as
an anatomic condition in which there is xed hindfoot eversion associated with valgus tilting of the talus within the ankle mortise (Fig. 1).24

From the Division of Orthopaedic Surgery, Department of Surgery, Duke University


Medical Center, Durham, North Carolina

FOOT AND ANKLE CLINICS


VOLUME 6 NUMBER 1 MARCH 2001

167

168

KELLY & NUNLEY

Figure 1. Standing anteroposterior (AP) view of the ankle demonstrates valgus tilt of the
talus within the ankle mortise.

ANATOMY AND PATHOGENESIS


The deltoid ligament is the medial collateral ligament of the ankle
(Fig. 2).8, 55 The anterior border of this ligament lies under the tibialis
anterior tendon and is in continuity with the thin anterior capsule of the
ankle. The posterior border is in continuity with the posterior ankle
capsule and the posterior tibiobular ligament. The ligamentous complex is invested by the deep crural fascia in continuity with the deep
exor retinaculum. A large segment of the deltoid ligament is covered
by the posterior tibial and exor digitorum longus tendons and the
brosynovial oor of these tendon sheaths, which blend with the ligament. The deltoid ligament is divided anatomically into supercial and
deep portions; the supercial portion is broad and triangular and has 5
fascicles based on their insertions. The most anterior is the anterior
supercial tibiotalar fascicle, which originates from the anterior border of
the anterior colliculus and inserts onto the dorsum of the talar neck, just
proximal to the talonavicular capsule. The tibionavicular fascicle has the
same point of origin from the anterior colliculus and inserts onto the
dorsomedial aspect of the navicular bone. The tibioligamentous fascicle
originates from the anterior colliculus to insert onto the superior border
of the superomedial calcaneonavicular ligament.8 The spring ligament
complex is described as the combination of the tibioligamentous fascicle
of the supercial deltoid and the superomedial and inferior calcaneonav-

TREATMENT OF STAGE 4 ADULT ACQUIRED FLATFOOT

169

Figure 2. Anatomy of the supercial deltoid and the superomedial calcaneonavicular


ligaments (A) demonstrates advancement of the attenuated deltoid ligament (arrows) proximally and posteriorly on the medial malleolus (B).

icular (spring) ligaments.8 The 3 anterior fascicles of the supercial


ligament represent the broader but weaker components of the supercial
deltoid ligament. The tibiocalcaneal fascicle is the strongest component
and the only component to cross the ankle and subtalar joints.29 Injury
to this fascicle has more inuence on tibiotalar contact characteristics
than injury to any of the other fascicles.10 The tibiocalcaneal fascicle
originates from the medial aspect of the anterior colliculus and descends
vertically onto the sustentaculum tali, after blending with bers of the
superomedial calcaneonavicular ligament. The fth fascicle is the posterior supercial tibiotalar fascicle, which originates from the posterior aspect

170

KELLY & NUNLEY

of the medial surface of the anterior colliculus and inserts on the posteromedial tibiotalar tubercle, reaching as far posteriorly as the bro-osseous
tunnel of the exor hallucis longus tendon.
The deep portion of the deltoid ligament is the tibiotalar ligament.
This is a short, strong ligament with 2 components. The anterior tibiotalar
ligament originates from the anterior colliculus and runs to the medial
talar surface. The posterior tibiotalar ligament, the strongest component of
the whole deltoid ligament, originates from the posterior aspect of the
anterior colliculus, the intercollicular fossa, and the anterior aspect of
the posterior colliculus. This ligament runs inferiorly, posteriorly, and
laterally onto a medial elevation of the medial talar surface. It is intraarticular and extrasynovial, and it functions as a medial and posterior
stabilizer of the ankle.
The static medial ligaments and the dynamic peroneal and posterior
tibial tendons play a role in the pathogenesis of the atfoot deformity.16
Insufciency of the dynamic musculature alone does not produce the
adult acquired atfoot deformity,36 and the additional attrition of the
medial ligaments is necessary for the typical atfoot posture.9 When the
PTT is not functional and the peroneal muscles are functional, there is
increased stress on the medial ligaments, suggesting that attrition over
time may be part of the presumed pathogenesis; however, this remains
to be proved.15, 26, 34, 38, 51, 52 When both PTT and the peroneals are not
functional, the atfoot deformity usually is not seen.36
In stage 4 disease, xed hindfoot eversion is seen in conjunction
with valgus tilt of the talus within the ankle mortise. Because the deltoid
ligament is the primary restraint to valgus talar tilt,10, 21, 29, 50, 52, 58 this
deformity is thought to develop because of deltoid ligament attrition,
aggravated by the abnormal mechanical axis of the prolonged pes planus
deformity.11 Because the subtalar joint has been in xed valgus for a
prolonged time, the tibiocalcaneal fascicle of the supercial deltoid must
have lengthened by attrition. The only remaining stabilizer against valgus tilt of the talus is the deep deltoid ligament. Cadaver studies have
showed that the supercial and deep components of the deltoid ligament
are of equal importance in the prevention of talar tilt.10, 21, 48, 50 When one
component of the ligament was released, no tilt was shown; however,
when the supercial and deep components of the deltoid were released,
signicant talar tilt was seen in the cadaver model. Stage 4 PTT insufciency implies the additional attrition and elongation of the deep
deltoid ligament, enabling valgus tilting of the talus within the ankle
mortise.9, 40, 52
DIAGNOSIS
Clinical examination differentiates stages 1, 2, and 3 PTT degeneration in terms of the presence and exibility of the atfoot deformity.
Because the clinical features of stage 3 and stage 4 disease are identical,
stage 4 disease is a radiographic diagnosis seen on the standing anteroposterior ankle view as valgus tilting of the talus within the ankle

TREATMENT OF STAGE 4 ADULT ACQUIRED FLATFOOT

171

mortise (See Fig. 1). The normal ankle should show minimal talar tilt on
valgus stress radiographs.31 Talar tilt of greater than 2 on stress lms
can be considered highly probable of signicant deltoid ligament injury.
Obvious tilt is pathologic. A rare nding is an ossicle on the deltoid
ligament, which may be associated with failure of the deltoid.4
NONOPERATIVE MANAGEMENT OF
STAGE 4 DISEASE
Most patients with symptomatic PTT insufciency should have a
trial of nonoperative management, including the use of orthoses. This
approach may be the only option in patients with signicant medical
comorbidities prohibiting surgery, whereas other patients may refuse
surgical intervention. The goals of bracing are to relieve pain, arrest
progression, and accommodate xed deformity.47 In stage 4 disease, the
hindfoot is xed in eversion, and any attempt to correct xed deformity
with an orthotic device causes pain and skin breakdown.34 The orthosis
must be accommodative rather than corrective.61 Although this device
may provide symptomatic relief and arrest progression, the joints still
are being abnormally loaded, and progressive arthrosis is almost inevitable. A trial of 3 to 6 months is recommended, and if successful, a brace
may be used indenitely. If unsuccessful or if the patient cannot tolerate
a long-term orthosis, operative management may be indicated. Orthotic
management is contraindicated in patients with balance or ambulation
difculty, such as patients with Parkinsons disease or signicant arthritis in other lower limb joints.7 Success with nonoperative management
tends to be the exception rather than the rule.60
Orthotic options include a molded nonarticulating ankle-foot orthosis (AFO), an Arizona brace (Ernesto G. Castro Custom Footwear Inc,
Messa, AZ), a Marzano brace (Yankee Bionics Inc, Arkon, OH), or a
patellar tendonbearing brace.61 The Arizona brace is a lace-up molded
leather and polypropylene AFO, whereas the Marzano brace combines
the University of California Biomechanics Laboratory brace with a custom-molded total contact anterior shell on the distal tibial crest, secured
with a posterior strap. The heel portion is molded in situ for xed
deformities. The Marzano brace is shorter than an AFO, but because it
is articulated at the ankle, it may not be appropriate in stage 4 PTT
dysfunction. Because of the talar tilt in stage 4 PTT degeneration, many
patients already may have ankle pain secondary to arthritis. By combining an AFO to accommodate the hindfoot position with an extension to
the knee, a patellar tendonbearing brace may unload the ankle and
provide more relief than an AFO alone.
There are few articles in the literature reporting the results of nonoperative management for all stages of PTT insufciency, and most of the
articles are from the same unit.7, 32, 56, 61 Chao et al7 reported on 53 feet
treated with orthoses over a 4-year period. Forty feet were treated with
a molded AFO and 13 with a University of California Biomechanics

172

KELLY & NUNLEY

Laboratory brace insert with medial posting. The nonobese patients with
exible deformities were treated with the latter, whereas the more severe
deformities were managed with the molded AFO. Of the 40 feet managed with the AFO, 26 were rated as excellent to good, 10 were rated as
fair to poor, and 4 were considered failures. Most of the satised patients
were older with low functional demands.
SURGICAL MANAGEMENT
The surgical management of stage 4 PTT degeneration is divided
into patients with deltoid ligament insufciency and no symptomatic
ankle arthritis and patients with deltoid ligament insufciency and
symptomatic ankle arthritis. In the rst group of patients, a triple arthrodesis is proposed to correct the hindfoot malalignment and deltoid
ligament advancement and plication to stabilize the ankle valgus. If the
ankle valgus had not been recognized until after the triple arthrodesis,
one must perform at a later date a medial calcaneal sliding osteotomy
and deltoid ligament reconstruction or revise the triple arthrodesis and
repair the deltoid ligament to correct malalignment.
When patients with stage 4 PTT degeneration present with symptomatic ankle arthritis, the 2 main surgical options are a tibiotalocalcaneal fusion or a pantalar fusion. These technically demanding operations
can produce satisfactory results and provide an alternative to amputation.35, 43 These operations remain salvage procedures in patients with
complex deformity and signicant pain because of a high complication
rate, stiffening of the hindfoot, a protracted recovery period, and no
arthroplasty option. Preoperatively, it is important to determine the
source of the deformity and the source of the pain.47 Selective local
anesthetic blocks can localize the pain source and estimate the relief
possible through fusion of the particular joint. If the pain is arising from
all 4 joints (ankle, subtalar, talonavicular, and calcaneocuboid joints), a
pantalar fusion is indicated. If the pain emanates from isolated regions,
however, limited arthrodeses may sufce. On occasion, supplementary
management with a brace may prevent further deformity, such as the
use of a patellar tendonbearing brace after a triple arthrodesis. The
arthrodesis must be corrective rather than in situ, to reduce the abnormal
loading of the surrounding joints.
Because these arthrodeses are not commonly performed, there is
continued debate regarding 1-stage versus 2-stage procedures, the surgical approach, and the instrumentation employed. In 1992, Papa and
Myerson43 reported on 21 patients after extended hindfoot arthrodeses
(13 tibiotalocalcaneal and 8 pantalar fusions). None of the procedures
were performed for stage 4 PTT dysfunction, highlighting the rarity of
this problem. The tibiotalocalcaneal group was more mobile and functioned at a higher level than the pantalar group.43 Any extended hindfoot
fusion would have signicant potential for complications, including
osteonecrosis of the talus, wound breakdown, deep infection, nonunion,

TREATMENT OF STAGE 4 ADULT ACQUIRED FLATFOOT

173

and malunion.3, 35, 44, 62 Because ankle and hindfoot motion is eliminated,
there are no compensatory mechanisms to accommodate inaccurate positioning of the foot, meaning malunion is tolerated poorly.
Tibiotalocalcaneal Arthrodesis
Tibiotalocalcaneal arthrodesis is a fusion of the ankle and subtalar
joints. The ankle is maintained at neutral, the subtalar joint is
arthrodesed in 5 of valgus, and the degree of external rotation should
be equivalent to the contralateral normal foot. This fusion is indicated
only if the talonavicular and calcaneocuboid joints are painless. This
clinical picture rarely is seen in PTT insufciency, and most articles on
this fusion relate to trauma or other hindfoot pathology.33 A tibiotalocalcaneal fusion can be achieved with external xation,53 with internal
xation,1, 18, 33, 35, 43, 54 or by a closed retrograde intramedullary nailing
technique.27, 49 The operative techniques are described fully in the literature.23, 33, 35, 53 Russotti et al53 reported satisfactory results in 75% of 21
patients after tibiotalocalcaneal fusion, using open bone grafting and
external xation. Mann and Chou33 reported on tibiocalcaneal fusion in
9 patients, using an open technique and lag screw xation. Union was
achieved in all patients at an average of 5 months. No patient in either
study had stage 4 PTT dysfunction. The use of retrograde intramedullary
nailing to achieve tibiotalocalcaneal fusion was popularized by Kile et
al in 1994.27 Again, accurate joint preparation and alignment are vital.
Advantages of this instrumentation include rotational control and its
load-sharing properties. The anatomy pertinent to reaming through the
heel has highlighted the proximity of the lateral plantar artery, lateral
plantar nerve, and nerve to abductor digiti quinti13, 46; these must be
protected to avoid postoperative neurogenic heel pain. The recommended incision is transverse and slightly posterior to the midtibial line
with the dissection remaining parallel to the nerve to abductor digiti
quinti.13 The mechanical bending and torsional properties of intramedullary nail xation and lag screw xation for tibiotalocalcaneal arthrodesis
have been compared, and nail xation was found to be signicantly
stiffer in all planes of bending and torsion.6 The authors postulated that
the more rigid construct may help maintain alignment and encourage a
higher rate of union.
Pantalar Arthrodesis
A pantalar arthrodesis is a fusion of the ankle, subtalar, talonavicular, and calcaneocuboid joints. In the past, this fusion was indicated
primarily for severe foot deformity in children. Barrett et al3 reported on
83 pantalar arthrodeses in a pediatric population with a follow-up of 1
to 33 years. The main indications were poliomyelitis and myelodysplasia, and these investigators concluded that a single-stage procedure

174

KELLY & NUNLEY

achieved as good a fusion rate as a 2-stage procedure. Waugh et al62


reported on 97 patients after 116 pantalar arthrodeses, with a minimum
follow-up of 2 years. Poliomyelitis was the most common indication,
and these authors concluded that a 1-stage procedure was effective.
Miehlke et al35 reported good results on 54 hindfoot fusions, 15 of which
were pantalar arthrodeses in patients with rheumatoid arthritis. The
authors have not found any reports in the literature relating the results
of pantalar arthrodesis specically for stage 4 PTT insufciency, highlighting the rarity of this advanced stage of disease.
PostTriple Arthrodesis Progression to
Stage 4 Disease
Residual hindfoot valgus can lead to valgus tilting of the talus
within the ankle mortise, a change in ankle joint contact characteristics,10,
11, 57
and a poor result after a triple arthrodesis.12 The residual valgus may
represent a malunion or progression of the deformity despite surgical
correction. This situation emphasizes the need to examine and repair as
necessary the deltoid ligament at the time of a triple arthrodesis. Articles
detailing the long-term results of triple arthrodeses report a worrying
rate of radiographic ankle arthrosis and stress the importance of accurate
alignment to reduce the likelihood of progressive arthrosis.2, 5, 14, 17, 20, 45, 63
Pell et al45 reported the results of 111 patients (132 feet) after a triple
arthrodesis at an average follow-up of 5.7 years. These investigators
noted a high prevalence of ankle arthritis clinically and radiographically;
however, they could detect no association between patient satisfaction
and the presence of ankle arthritis. Bennett et al5 detailed the symptoms,
clinical examination, and radiographic features of 22 patients at a minimum of 3 years after triple arthrodesis. Eleven patients (50%) had mild
ankle tenderness, and 11 had progression of ankle arthrosis within the
short follow-up period of this study.
The management options in this difcult situation include accommodative bracing, revision of the triple arthrodesis, medial displacement
calcaneal osteotomy, deltoid ligament reconstruction, ankle fusion, and
total ankle replacement. Anteroposterior standing views of both ankles
rule out lateral talar dome segmental osteonecrosis or a bular stress
fracture,25 either of which could mimic stage 4 valgus talar tilting. Stress
radiographs attempting to reduce the ankle joint should be obtained.31
If the joint is irreducible, the options are bracing, ankle fusion, or a total
ankle arthroplasty (Fig. 3). If the ankle joint is reducible and the hindfoot
has been fused in excessive valgus, the fusion could be revised.19 Otherwise, a medial displacement calcaneal osteotomy,39, 45, 52 deltoid ligament
reconstruction,28 or these procedures combined can be considered; however, there are no reports of the long-term results available. The medial
displacement calcaneal osteotomy realigns the valgus heel under the
mechanical axis of the tibia,30 acts as a gastrocnemius-soleus tendon
transfer to a more biomechanically stable position for the ankle joint,41

TREATMENT OF STAGE 4 ADULT ACQUIRED FLATFOOT

175

Figure 3. Management of post triple arthrodesis valgus talar tilt. MDCO  medial displacement calcaneal osteotomy; OA  osteoarthritis.

reduces the forefoot abducting force of peroneus longus,59 alters the


tibiotalar joint contact characteristics,11, 57 and reduces the stress on the
deltoid52 and calcaneonavicular42 ligaments. It does not act through the
plantar fascia,22 which previously was thought to be its means of medial
arch restoration through the windlass mechanism.37 Using cadaver specimens, a deltoid ligament reconstruction restored hindfoot alignment and
potentially avoided the need for an ankle fusion.28 Goldner et al16 were
the rst to detail the advancement and plication of the superomedial
calcaneonavicular ligament, with the tibiotalar and tibiocalcaneal portions of the supercial deltoid ligament, in the treatment of early atfoot
deformity. No clinical results of this procedure have been reported to
the authors knowledge in the management of stage 4 PTT disease,
however. Likewise, total ankle arthroplasty results in this patient population have not been reported, but with newer prostheses this may be a
preferable option to the extended arthrodeses. There also may be a role
for medial displacement calcaneal osteotomy in addition to a triple
arthrodesis if the hindfoot cannot be positioned properly.45, 52
SUMMARY
Stage 4 PTT dysfunction is a rare anatomic condition in which xed
hindfoot valgus is associated with valgus tilting of the talus within the

176

KELLY & NUNLEY

ankle mortise. Success with nonoperative management is the exception


rather than the rule. The surgical options are a tibiotalocalcaneal fusion
or a pantalar fusion; however, there are few results reported in the adult
acquired atfoot population. Valgus talar tilting after triple arthrodesis
may be the challenge of the future.

References
1. Alvarez RG, Barbour TM, Perkins TD: Tibiocalcaneal arthrodesis for nonbraceable
neuropathic ankle deformity. Foot Ankle Int 15:354, 1994
2. Angus PD, Cowell HR: Triple arthrodesis: A critical long-term review. J Bone Joint
Surg Br 68:260, 1986
3. Barrett G, Meyer LC, Bray EW, et al: Pantalar arthrodesis: A long-term follow-up. Foot
Ankle Int 1:279, 1981
4. Beals TC, Pomeroy GC, Manoli A II: Posterior tendon insufciency: Diagnosis and
treatment. J Am Acad Orthop Surg 7:112, 1999
5. Bennett GL, Graham CE, Mauldin DM: Triple arthrodesis in adults. Foot Ankle Int
12:138, 1991
6. Berend ME, Glisson RR, Nunley JA: A biomechanical comparison of intramedullary
nail and crossed lag screw xation for tibiotalocalcaneal arthrodesis. Foot Ankle Int
18:639, 1997
7. Chao W, Wapner KL, Lee TH, et al: Nonoperative management of posterior tibial
tendon dysfunction. Foot Ankle Int 17:736, 1996
8. Davis WH, Sobel M, DiCarlo EF, et al: Gross, histological, and microvascular anatomy
and biomechanical testing of the spring ligament complex. Foot Ankle Int 17:95, 1996
9. Deland JT, Arnoczky SP, Thompson FM: Adult acquired atfoot deformity at the
talonavicular joint: Reconstruction of the spring ligament in an in vitro model. Foot
Ankle Int 13:327, 1992
10. Earll M, Wayne J, Brodrick C, et al: Contribution of the deltoid ligament to ankle joint
contact characteristics: A cadaver study. Foot Ankle Int 17:317, 1996
11. Fairbank A, Myerson MS, Fortin P, et al: The effect of calcaneal osteotomy on contact
characteristics of the tibiotalar joint. Foot 5:137, 1995
12. Figgie MP, OMalley MJ, Ranawat C, et al: Triple arthrodesis in rheumatoid arthritis.
Clin Orthop 292:250254, 1993
13. Flock TJ, Ishikawa S, Hecht PJ, et al: Heel anatomy for retrograde tibiotalocalcaneal
roddings: A roentgenographic and anatomic analysis. Foot Ankle Int 18:233, 1997
14. Fortin PT, Walling AK: Triple arthrodesis. Clin Orthop 365:9199, 1999
15. Funk DA, Cass JR, Johnson KA: Acquired adult at foot secondary to posterior tibialtendon pathology. J Bone Joint Surg Am 68:95, 1986
16. Goldner JL, Keats PK, Bassett FHD, et al: Progressive talipes equinovalgus due to
trauma or degeneration of the posterior tibial tendon and medial plantar ligaments.
Orthop Clin North Am 5:39, 1974
17. Graves SC, Mann RA, Graves KO: Triple arthrodesis in older adults: Results after
long-term follow-up. J Bone Joint Surg Am 75:355, 1993
18. Gruen GS, Mears DC: Arthrodesis of the ankle and subtalar joints. Clin Orthop
268:1520, 1991
19. Haddad SL, Myerson MS, Pell RFT, et al: Clinical and radiographic outcome of revision
surgery for failed triple arthrodesis. Foot Ankle Int 18:489, 1997
20. Haritidis JH, Kirkos JM, Provellegios SM, et al: Long-term results of triple arthrodesis:
42 cases followed for 25 years. Foot Ankle Int 15:548, 1994
21. Harper MC: Deltoid ligament: An anatomical evaluation of function. Foot Ankle Int
8:19, 1987
22. Horton GA, Myerson MS, Parks BG, et al: Effect of calcaneal osteotomy and lateral
column lengthening on the plantar fascia: A biomechanical investigation. Foot Ankle
Int 19:370, 1998

TREATMENT OF STAGE 4 ADULT ACQUIRED FLATFOOT

177

23. Johnson K: Master Techniques in Orthopaedic Surgery. New York, Raven Press, 1994
24. Johnson KA, Strom DE: Tibialis posterior tendon dysfunction. Clin Orthop 239:196
206, 1989
25. Jones CK, Nunley JA: Osteonecrosis of the lateral aspect of the talar dome after triple
arthrodesis: A report of three cases. J Bone Joint Surg Am 81:1165, 1999
26. Kaye RA, Jahss MH: Tibialis posterior: A review of anatomy and biomechanics in
relation to support of the medial longitudinal arch. Foot Ankle Int 11:244, 1991
27. Kile TA, Donnelly RE, Gehrke JC, et al: Tibiotalocalcaneal arthrodesis with an intramedullary device. Foot Ankle Int 15:669, 1994
28. Kitaoka HB, Luo ZP, An KN: Reconstruction operations for acquired atfoot: Biomechanical evaluation. Foot Ankle Int 19:203, 1998
29. Kjaersgaard-Andersen P, Wethelund JO, Helmig P, et al: Stabilizing effect of the
tibiocalcaneal fascicle of the deltoid ligament on hindfoot joint movements: An experimental study. Foot Ankle Int 10:30, 1989
30. Koutsogiannis E: Treatment of mobile at foot by displacement osteotomy of the
calcaneus. J Bone Joint Surg Br 53:96, 1971
31. Leith JM, McConkey JP, Li D, et al: Valgus stress radiography in normal ankles. Foot
Ankle Int 18:654, 1997
32. Lin S, Lee TH, Chao W, et al: Nonoperative treatment of patients with posterior tibial
tendonitis. Foot Ankle Clin 1:261, 1996
33. Mann R, Chou LB: Tibiocalcaneal arthrodesis. Foot Ankle Int 16:401, 1995
34. Mann RA, Thompson FM: Rupture of the posterior tibial tendon causing at foot:
Surgical treatment. J Bone Joint Surg Am 67:556, 1985
35. Miehlke W, Gschwend N, Rippstein P, et al: Compression arthrodesis of the rheumatoid
ankle and hindfoot. Clin Orthop 340:7586, 1997
36. Mizel MS, Temple HT, Scranton PE Jr, et al: Role of the peroneal tendons in the
production of the deformed foot with posterior tibial tendon deciency. Foot Ankle
Int 20:285, 1999
37. Mosca VS: Calcaneal lengthening for valgus deformity of the hindfoot: Results in
children who had severe, symptomatic atfoot and skewfoot. J Bone Joint Surg Am
77:500, 1995
38. Mosier SM, Pomeroy G, Manoli A II: Pathoanatomy and etiology of posterior tibial
tendon dysfunction. Clin Orthop 365:1222, 1999
39. Myerson M: Adult acquired atfoot deformity: Treatment of dysfunction of the posterior tibial tendon. J Bone Joint Surg Am 78:780, 1996
40. Myerson MS: Adult acquired atfoot deformity: Treatment of dysfunction of the
posterior tibial tendon. Instr Course Lect 46:393, 1997
41. Myerson MS, Corrigan J, Thompson F, et al: Tendon transfer combined with calcaneal
osteotomy for treatment of posterior tibial tendon insufciency: A radiological investigation. Foot Ankle Int 16:712, 1995
42. Otis JC, Deland JT, Kenneally S, et al: Medial arch strain after medial displacement
calcaneal osteotomy: An in vitro study. Foot Ankle Int 20:222, 1999
43. Papa JA, Myerson MS: Pantalar and tibiotalocalcaneal arthrodesis for post-traumatic
osteoarthrosis of the ankle and hindfoot. J Bone Joint Surg Am 74:1042, 1992
44. Patterson R, Parrish FF, Hathaway EN: Stabilizing operations of the foot: A study of
the indications, techniques used, and the end results. J Bone Joint Surg Am 32:1, 1950
45. Pell RFT, Myerson MS, Schon LC: Clinical outcome after primary triple arthrodesis. J
Bone Joint Surg Am 82:47, 2000
46. Pochatko DJ, Smith JW, Phillips RA, et al: Anatomic structures at risk: Combined
subtalar and ankle arthrodesis with a retrograde intramedullary rod. Foot Ankle Int
16:542, 1995
47. Pomeroy GC, Pike RH, Beals TC, et al: Acquired atfoot in adults due to dysfunction
of the posterior tibial tendon. J Bone Joint Surg Am 81:1173, 1999
48. Quiles M, Requena F, Gmez L, et al: Functional anatomy of the medial collateral
ligament of the ankle joint. Foot Ankle Int 4:73, 1983
49. Quill G: Tibiotalocalcaneal and pantalar arthrodesis. Foot Ankle Clin 1:199, 1996
50. Rasmussen O, Kromann-Andersen C, Boe S: Deltoid ligament: Functional analysis of

178

51.
52.

53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.

KELLY & NUNLEY

the medial collateral ligamentous apparatus of the ankle joint. Acta Orthop Scand
54:36, 1983
Reeck J, Felten N, McCormack AP, et al: Support of the talus: A biomechanical
investigation of the contributions of the talonavicular and talocalcaneal joints, and the
superomedial calcaneonavicular ligament. Foot Ankle Int 19:674, 1998
Resnick RB, Jahss MH, Choueka J, et al: Deltoid ligament forces after tibialis posterior
tendon rupture: Effects of triple arthrodesis and calcaneal displacement osteotomies
[published erratum appears in Foot Ankle Int 1995; 16(5):314]. Foot Ankle Int 16:14,
1995
Russotti GM, Johnson KA, Cass JR: Tibiotalocalcaneal arthrodesis for arthritis and
deformity of the hind part of the foot. J Bone Joint Surg Am 70:1304, 1988
Sanders R, Pappas J, Mast J, et al: The salvage of open grade IIIB ankle and talus
fractures. J Orthop Trauma 6:201, 1992
Sarraan S: Anatomy of the Foot and Ankle, ed 2. Philadelphia, JB Lippincott, 1993
Sferra J, Rosenberg GA: Nonoperative treatment of posterior tibial tendon pathology.
Foot Ankle Clin 2:261, 1997
Steffensmeier SJ, Saltzman CL, Berbaum KS, et al: Effects of medial and lateral displacement calcaneal osteotomies on tibiotalar joint contact stresses. J Orthop Res 14:980, 1996
Stormont DM, Morrey BF, An KN, et al: Stability of the loaded ankle: Relation between
articular restraint and primary and secondary static restraints. Am J Sports Med
13:295, 1985
Thordarson DB, Schmotzer H, Chon J, et al: Dynamic support of the human longitudinal arch: A biomechanical evaluation. Clin Orthop 316:165172, 1995
Walling A: Editorial comment. Clin Orthop 365:24, 1999
Wapner KL, Chao W: Nonoperative treatment of posterior tibial tendon dysfunction.
Clin Orthop 365:3945, 1999
Waugh TR, Wagner J, Stincheld FE: An evaluation of pantalar arthrodesis: A followup study of one hundred and sixteen operations. J Bone Joint Surg Am 47:1315, 1965
Wetmore RS, Drennan JC: Long-term results of triple arthrodesis in Charcot-MarieTooth disease. J Bone Joint Surg Am 71:417, 1989
Address reprint requests to
James A. Nunley, MD
Division of Orthopaedic Surgery
Department of Surgery, Box 2923
Duke University Medical Center
5330 Hospital South
Durham, NC 27710
e-mail: nunle001@me.duke.edu

Update on Stage IV Acquired Adult


Flatfoot Disorder
When the Deltoid Ligament Becomes
Dysfunctional
Jeremy T. Smith, MD, Eric M. Bluman, MD, PhD*

KEYWORDS
Posterior tibial tendon dysfunction Acquired adult flatfoot deformity Stage IV
Deltoid ligament Ankle joint
KEY POINTS
Stage IV acquired adult flatfoot deformity (AAFD) is present when tibiotalar valgus occurs
in association with AAFD.
Stage IV AAFD can be divided into stage IV-A (flexible deformity without advanced
tibiotalar arthritis) and stage IV-B (rigid deformity or flexible deformity with advanced
tibiotalar arthritis).
Stage IV-A can be treated with joint-sparing procedures that include reconstruction of the
deltoid ligament.
Stage IV-B should be treated with a joint-sacrificing procedure.
Surgical procedures that address stage IV AAFD should be accompanied by procedures to
achieve a plantigrade foot.

The posterior tibial tendon plays a central role in maintaining proper foot alignment.
Dysfunction of the posterior tibial tendon leads to acquired adult flatfoot deformity
(AAFD), a condition that can be both painful and disabling. There are four stages of
AAFD, which correlate to increasing deformity. Johnson and Strom1 described the
first three stages in 1989. Stage I AAFD is paratenonitis of the posterior tibial tendon
without deformity. The central components of stage II disease are flexible flatfoot
deformity with hindfoot valgus, forefoot abduction, and forefoot varus. Stage III is
characterized by a fixed hindfoot valgus and often fixed forefoot varus deformity.
Myerson2,3 described a fourth stage, which occurs when the talus tilts into valgus
within the ankle mortise secondary to deltoid ligament insufficiency (Fig. 1). Stage IV
Department of Orthopaedic Surgery, Brigham Foot and Ankle Center at the Faulkner, Brigham
and Womens Hospital, 1153 Centre Street, Suite 56, Boston, MA 02130, USA
* Corresponding author.
E-mail address: ebluman@partners.org
Foot Ankle Clin N Am 17 (2012) 351360
doi:10.1016/j.fcl.2012.03.011
1083-7515/12/$ see front matter 2012 Elsevier Inc. All rights reserved.

foot.theclinics.com

352

Smith & Bluman

Fig. 1. Stage IV acquired adult flatfoot deformity due to posterior tibial tendon dysfunction.
(A) Lateral view of the foot. (B) Anteroposterior view of the ankle with tibiotalar joint valgus
angulation.

ankle malalignment is typically accompanied by the deformities present in stages II


and III. Stage IV AAFD is subclassified into stage IV-Aflexible ankle valgus without
substantial tibiotalar arthritis, and stage IV-Brigid ankle valgus or flexible ankle
valgus with significant tibiotalar arthritis.4
ANATOMY AND PATHOGENESIS

The posterior tibialis muscle originates on the posterior tibia, fibula, and interosseous
membrane. The posterior tibial tendon (PTT) passes behind the medial malleolus and
has a broad insertion at the medial midfoot, a large part of which is on the navicular
tuberosity. Additional points of insertion include the sustentaculum tali, cuboid,
cuneiforms, and the metatarsal bases of the second, third, and fourth rays. The PTT
functions as a powerful inverter of the foot5 and helps to maintain the foots
longitudinal arch.6 During normal gait, the PTT pulls the hindfoot into varus, which
facilitates locking of the transverse tarsal joint, providing a rigid foot for push-off.
Dysfunction of the posterior tibial tendon has been associated with progressive
AAFD.7 This deformity is thought to relate to the degree and duration of tendon
dysfunction. Contributing dynamic deforming factors include the unopposed action of
the peroneus brevis (the principal antagonist of the posterior tibialis) and contracture
of the triceps surae. The latter exacerbates hindfoot valgus by acting as a hindfoot
everter. As hindfoot valgus progresses, the mechanical axis of the leg assumes a
medial position relative to the foot with resultant increased tension on soft tissues of
the medial ankle. This process leads to progressive failure of the static stabilizers of
the medial hindfoot and midfoot, including the spring ligament, plantar fascia, and
deltoid ligament complex.8 12 In cases of advanced AAFD, the spring ligament
complex often attenuates or ruptures. Failure of this ligament complex allows the talar
head to shift medial and plantar relative to the navicular. As the deltoid ligament
complex fails, the talus may assume a valgus position within the ankle mortise.
The deltoid ligament complex consists of superficial and deep components. It
serves a critical role in supporting the spring ligament, tibiotalar joint, subtalar joint,
and talonavicular joint. The superficial deltoid ligament has been shown to be the
primary restraint to tibiotalar valgus angulation.1315 Descriptions of the course of the

Stage IV Acquired Adult Flatfoot Update

superficial deltoid vary, which may be because of its wide fan-shaped insertion from
the navicular to the posterior tibiotalar capsule.1519 The tibiocalcaneal ligament is the
strongest component of the superficial deltoid.18 The deep deltoid ligament, originating from the intercollicular groove and posterior colliculus and inserting on the talus,
has been shown to principally prevent axial rotation of the talus within the ankle
mortise.1315
The deltoid ligament has an important role in preventing valgus deformity of the
ankle.
Sequential sectioning of the deltoid ligament complex has demonstrated that both
the superficial and deep components prevent tibiotalar tilt. When both components
are sectioned, significant valgus talar tilt may occur17,20,21 and decrease ankle joint
contact area by up to 43%.20
Chronic deltoid ligament insufficiency can be caused by AAFD,2,3 trauma, sportsrelated injuries, triple arthrodesis with valgus malunion,22 and a poor result after ankle
arthroplasty.23,24 Some data suggest that the deltoid ligament is stressed even after
a properly positioned triple arthrodesis.25,26 Most patients with stage IV AAFD have
progressed through stage III disease, although a subset of patients develop valgus
talar tilt without a rigid flatfoot deformity, suggesting a possible progression directly
from stage II to stage IV AAFD. As described by Bluman and colleagues4 in 2007,
stage IV AAFD may present with or without lateral ankle instability, tibiotalar arthritis,
and/or a rigid valgus ankle deformity. Treatment approaches for stage IV AAFD vary
depending on the flexibility of the deformity and the degree of ankle arthritis. Stage
IV-A is characterized by flexible ankle valgus without significant tibiotalar arthritis.
Stage IV-B can be seen with either rigid ankle valgus or flexible ankle valgus with
significant tibiotalar arthritis. Patients diagnosed with early stage IV disease may be
more likely to have a flexible deformity than those who present later in their disease
course.27
CLINICAL PRESENTATION AND EVALUATION

Patients with deltoid deficiency and AAFD typically present similarly to patients with
stage III AAFD, with the addition of occasional medial-sided ankle pain. Medial-sided
pain may have been present earlier in the disease course from PTT paratenonitis.
Often this pain will abate after rupture of the PTT and then recur in late-stage disease
as the medial ligaments are stressed. Tarsal tunnel neuropathy may also occur with
severe valgus deformity.28,29
Upon examination, the most likely findings are a pronounced valgus deformity of
the hindfoot, an antalgic gait with shortened stride length, and an inability to heel rise.
Evaluation should include inspection for abrasions from shoe wear or orthoses. Pain
or skin changes beneath the talar head may relate to dorsolateral peritalar subluxation
causing prominence at the medial plantar midfoot. Lateral pain may be due to sinus
tarsi or subfibular impingement and/or lateral ankle joint arthritis. Other common
findings include triceps surae contracture and forefoot varus. Ankle joint range of
motion may be severely limited, and care should be taken to accurately measure true
ankle dorsiflexion because the fixed hindfoot deformity may falsely elevate perceived
ankle dorsiflexion. The ipsilateral knee should also be evaluated because genu
valgum can contribute to the foot and ankle deformity.30
Proper radiographic evaluation includes imaging both the foot and ankle. Foot
radiographs are used to assess for loss of talar head coverage by the navicular
(anteroposterior view), dorsolateral peritalar subluxation (anteroposterior and lateral
view), increased talofirst metatarsal angle (anteroposterior and lateral views), and
decreased calcaneal pitch (lateral view). Weightbearing ankle radiographs will reveal

353

354

Smith & Bluman

increased talar tilt (anteroposterior and mortise view), the upper limit of which has
been reported to be 2.3134 Ankle radiographs are also reviewed to assess tibiotalar
arthritis. If talar valgus tilt is present, the flexibility of this deformity must be assessed.
Fluoroscopic manipulation is used to evaluate both the flexibility of a valgus deformity
and to assess for varus instability, which may be present with concomitant lateral
ligament rupture. Comparison contralateral extremity radiographs may be of use.
MANAGEMENT

Some investigators have reported good outcomes with nonsurgical treatment of


early-stage AAFD.3537 For stage I and II disease, a structured nonoperative protocol
with orthoses and physical therapy was shown to have successful subjective and
functional outcomes in 83% of patients.35 For stage III and IV disease, custom
molded orthoses such as an Arizona brace may slow the progression of valgus ankle
angulation.36 Yet whereas bracing may control talar tilt, its success in minimizing the
development of tibiotalar arthritis is in question. The altered ankle joint biomechanics
that accompany deltoid incompetence decreased joint contact area and increased
joint contact forces20may not be adequately corrected with orthoses.
Because of concerns that orthoses do not provide adequate biomechanical
correction, some investigators espouse operative treatment of stage IV AAFD in all
patients medically fit for surgery.27,32,38 Severe valgus ankle deformity has historically
been treated with tibiotalar fusion. The problem with this approach is that tibiotalar
fusion is often accompanied by procedures to correct associated foot deformities
such as fusions of the subtalar, talonavicular, calcaneocuboid, and/or midfoot joints.
The result of these combined procedures may be a tibiotalocalcaneal or pantalar
fusion. Although tibiotalocalcaneal or pantalar fusions are successful in reestablishing
a plantigrade foot, many patients report residual pain.39 41 Objectively, there is
increased ambulatory energy expenditure and decreased functionality.41
The subclassification of stage IV AAFD into stage IV-A (flexible ankle valgus without
significant tibiotalar arthritis) and stage IV-B (rigid ankle valgus or flexible ankle valgus
with significant tibiotalar arthritis) has helped to clarify the surgical indications for
stage IV AAFD.4 It is paramount to determine which of these two subclassifications a
patient falls into, because stage IV-A is generally treated with ankle jointsparing
procedures and stage IV-B with joint-sacrificing procedures.
Joint-Sparing Operative Management

Joint-sparing operations preserve ankle range of motion, which would otherwise be


lost with tibiotalocalcaneal or pantalar fusion. Prior to performing a joint-sparing
procedure, the surgeon must understand the deformity. A joint-sparing operation for
stage IV AAFD is appropriate for stage IV-A disease. Preoperative or intraoperative
fluoroscopic assessment can help to establish the flexibility of the tibiotalar deformity
and also identify associated varus instability, if present. If there is associated lateral
ligamentous incompetence, an additional lateral ligament reconstruction should be
performed. This reconstruction can be accomplished with a variety of techniques,
including calcaneofibular ligament imbrication or free gracilis grafting.42
Joint-sparing operations rely on reconstruction of the failed deltoid ligament
complex. Flatfoot realignment procedures must be also be performed,43,44 typically
prior to medial ligamentous reconstruction in order to facilitate proper tensioning and
fixation. The importance of these realignment procedures cannot be overstated,
because medial ligament reconstructive attempts will fail if residual coronal plane
deformity persists.

Stage IV Acquired Adult Flatfoot Update

Multiple investigators have described techniques for deltoid ligament reconstruction,4555 although to date little data exist regarding the outcomes of these various
procedures. Ligament reconstruction procedures can be divided into two categories:
host-tissue repair operations and graft substitution operations. Although several
host-tissue repair procedures have been found to have good short-term results,50,53
there is concern that chronically injured deltoid ligament tissue may not allow for
durable repair over a longer time period.56 In the absence of long-term data
demonstrating the effectiveness of host-tissue repair, several investigators have
proposed autograft or allograft tendon reconstructive procedures as an alternative.27,45,47,54,55 This review focuses on two techniques that have demonstrated
reasonable clinical results.47,48,55
In 2004, Delands47 group published a description of deltoid ligament reconstruction using peroneus longus tendon autograft in patients with stage IV flatfoot.47 In this
procedure, the peroneus longus tendon is transected proximally and then passed
through a tunnel in the talus from lateral to medial. The tendon is then passed from
medial to lateral through a tunnel in the medial malleolus to replicate the fibers of the
deep deltoid. Fixation is achieved either by suturing to a screw at the fibula or by
stapling at the lateral tibia. In a study of 5 patients with a minimum 2-year follow-up,
average correction of valgus talar tilt was from 10 preoperative valgus to 4
postoperative valgus.47 A follow-up publication examining the same five patients at a
mean 8.9 years postoperatively reported a mean valgus talar tilt of 2, mean Foot and
Ankle Orthopaedic Surgery score of 68 (maximum, 100), mean SF-36v2 score of 76
(maximum,100), mean visual analogue scale of 4 (maximum, 10), and mean ankle
range of motion of 47.48 A criticism of this procedure is that it fails to reconstruct the
superficial component of the deltoid ligament complex, which has been shown to be
the primary restraint to valgus ankle angulation.1315 Furthermore, there is concern
about donor-site morbidity from the use of autograft tendon, and the senior author of
this technique now reportedly uses Achilles tendon allograft instead of peroneal
tendon autograft.48
In 2011, Jeng and colleagues55 reported the outcomes of a minimally invasive
deltoid ligament reconstruction for stage IV-A flatfoot deformity (Fig. 2). This technique, described initially in 2007,27 uses hamstring allograft tendon to reconstruct
both the deep and superficial components of the deltoid ligament complex. A 20-cm
graft is split longitudinally to create two limbs, leaving 6 cm intact as a single limb. The
single limb is passed through a tibial tunnel parallel to the ankle joint surface at the
distal tibial physeal scar and fixed using a bioabsorbable PLLA interference screw.
The graft is subcutaneously passed distally and one limb is secured by interference
screw through a talar tunnel to recreate the deep deltoid fibers. The remaining limb is
then passed through a calcaneal tunnel and fixed with an interference screw to
recreate the superficial deltoid (Fig. 3).
Candidates for this minimally invasive deltoid ligament reconstruction had (1) ankle
pain refractory to nonsurgical care, (2) rigid hindfoot valgus and forefoot abduction
deformity consistent with stage IV AAFD, (3) valgus talar tilt of at least 3 that was (4)
passively correctable tibiotalar valgus documented fluoroscopically and had (5) at
least 2 mm of lateral tibiotalar cartilage remaining.55 At an average of 36 months
postoperatively, 5 of 8 patients were deemed to have had a successful outcome with
a correction to less than 3 of valgus talar tilt, preserved lateral tibiotalar joint space,
SF-12 functional scores equivalent to age-matched normative scores, and mean
ankle range of motion of 42. Patients with a preoperative valgus talar tilt of greater
than 10 had inferior outcomes. The investigators concluded that this procedure is
appropriate for a carefully selected population with stage IV-A AAFD.

355

356

Smith & Bluman

Fig. 2. Stage IV-A acquired adult flatfoot deformity reconstruction using a minimally invasive
deltoid ligament reconstruction. (A) Preoperative anteroposterior view of the ankle showing
a valgus deformity of the tibiotalar joint. (B) Postoperative anteroposterior view showing
correction of the deformity.

Another technique that shows early promise in biomechanical studies was described by Haddad and colleagues54 in 2010. This reconstruction attempts to
anatomically recreate both limbs of the deltoid ligament complex using a looped
semitendinosis allograft passed through a drill hole at the intercollicular groove of the
medial malleolus to the anterior distal tibia.54 One end of the tendon is then secured
to the talus through a tunnel to recreate the deep tibiotalar limb whereas the other is
secured to the calcaneus to recreate the superficial tibiocalcaneal limb. A cadaveric
biomechanical analysis showed no statistical difference in angular displacement
under load between reconstructed ankles and cadaveric ankles with an intact native
deltoid ligament. The conclusions from this cadaveric study were that this technique
restores eversion and external rotation stability to the talus under low torque. The
investigators themselves mention one of the biggest concerns with this technique,
which is that to date there are no published clinical outcomes. Although the
biomechanical data are promising, in vivo healing and durability are not yet known.
Joint-Sacrificing Operative Management

Patients with rigid valgus talar tilt or severe tibiotalar arthritis are not appropriate for
an ankle jointsparing procedure, but rather should be considered candidates for
either a tibiotalocalcaneal or pantalar arthrodesis.4,27 These procedures are reliable
but are not without potential problems because many patients have residual pain,
nonunion, increased ambulatory energy expenditure, and decreased functionality.39 41 An alternative approach to stage IV-B AAFD is to realign the weightbearing
axis with a hindfoot fusion and then perform a total ankle arthroplasty.38,57,58
Although this approach is an attractive option, persistent valgus malalignment has
been reported in 50% of patients after this combination of procedures.58 Another
option for a stage IV-B AAFD is a combined total ankle arthroplasty and deltoid
ligament reconstruction. Although little has been published on this topic, the

Stage IV Acquired Adult Flatfoot Update

Fig. 3. Minimally invasive deltoid ligament reconstruction. Disarticulated superior view of


the talus (A) and calcaneus (B) demonstrating the trajectory of graft limbs with arrows
marking the entry site of the graft in the bones. (C) Coronal sectional view showing how graft
limbs are anchored. (D) Medial view of the ankle showing tibial (black), talar (light gray), and
calcaneal (dark gray) limbs of graft.

minimally invasive ligament reconstruction technique described by Jeng and colleagues55 would be a good choice because of the small incision and avoidance of
large skin bridges. Proper implant selection is critical, because a large tibial keel will
interfere with proper graft tunnel placement.

357

358

Smith & Bluman

SUMMARY

Deltoid ligament complex insufficiency is a fundamental pathologic component of


stage IV AAFD. Failure of the deltoid ligament allows the talus to tilt into valgus within
the ankle mortise. If left untreated, ankle joint biomechanics are altered and may lead
to debilitating tibiotalar arthritis. All surgical treatments that address the valgus talar
tilt seen with stage IV AAFD require accompanying procedures to properly realign the
hindfoot. Stage IV AAFD can be subdivided into two groups. Patients with a flexible
ankle deformity without advanced tibiotalar arthritis (stage IV-A) can be considered for
a joint-sparing procedure. A variety of procedures have been described, but longterm follow-up studies have yet to determine which of these techniques is optimal.
Patients with a rigid valgus ankle deformity or a flexible deformity accompanied by
advanced tibiotalar arthritis (stage IV-B) should be considered for a joint-sacrificing
procedure. To date, the most reliable results for stage IV-B AAFD have been reported
with either tibiotalocalcaneal or pan-talar arthrodesis.
REFERENCES

1. Johnson KA, Strom DE. Tibialis posterior tendon dysfunction. Clin Orthop 1989;(239):
196 206.
2. Myerson MS. Adult acquired flatfoot deformity; treatment of dysfunction of the
posterior tibial tendon. Inst Course Lect 1997;46:393 405.
3. Myerson MS. Adult acquired flatfoot deformity. Treatment of dysfunction of the
posterior tibial tendon. J Bone Joint Surg Am 1996;78:780 92.
4. Bluman EM, Title CI, Myerson MS. Posterior tibial tendon rupture: a refined classification system. Foot Ankle Clin 2007;12:233 49.
5. Mann R, Inman VT. Phasic activity of intrinsic muscles of the foot. J Bone Joint Surg
Am 1964;46:469 81.
6. Kitoaka HB, Luo ZP, An KN. Effect of the posterior tibial tendon on the arch of the foot
during simulated weightbearing: biomechanical analysis. Foot Ankle Int 1997;18(1):
43 6.
7. Mann RA, Thompson FM. Rupture of the posterior tibial tendon causing flat foot:
surgical treatment. J Bone Joint Surg Am 1985;67:556 61.
8. Deland JT. The acquired flatfoot and spring ligament complex. Foot Ankle Clin
2001;6:129 35.
9. Deland JT, de Asla RJ, Sung IH, et al. Posterior tibial tendon insufficiency: which
ligaments are involved? Foot Ankle Int 2005;26(6):42735.
10. Balen PF, Helms CA. Association of posterior tibial tendon injury with spring ligament
injury, sinus tarsi abnormality, and plantar fasciitis on MR imaging. AJR Am J
Roentgenol 2001;176(5):1137 43.
11. Yao L, Gentili A, Cracchiolo A. MR imaging findings in spring ligament insufficiency.
Skeletal Radiol 1999;28(5):24550.
12. Gazdag AR, Cracchiolo A 3rd. Rupture of the posterior tibial tendon. Evaluation of
injury of the spring ligament and clinical assessment of tendon transfer and ligament
repair. J Bone Joint Surg Am 1997;79(5):675 81.
13. Rasmussen O. Stability of the ankle joint. Analysis of the function and traumatology of
the ankle ligaments. Acta Orthop Scand Suppl 1985;211:175.
14. Rasmussen O, Kromann-Andersen C. Experimental ankle injuries. Analysis of the
traumatology of the ankle ligaments. Acta Orthop Scand 1983;54:356 62.
15. Rasmussen O, Kromann-Andersen C, Boe S. Deltoid ligament. Functional analysis of
the medial collateral ligamentous apparatus of the ankle joint. Acta Orthop Scand
1983;54:36 44.

Stage IV Acquired Adult Flatfoot Update

16. Boss AP, Hintermann B. Anatomical study of the medial ankle ligament complex. Foot
Ankle Int. 2002;23:54753.
17. Harper MC. Deltoid ligament: an anatomical evaluation of function. Foot Ankle
1987;8:19 22.
18. Pankovich AM, Shivaram MS. Anatomical basis of variability in injuries of the medial
malleolus and the deltoid ligament. I. Anatomical studies. Acta Orthop Scand 1979;
50:21723.
19. Siegler S, Block J, Schneck CD. The mechanical characteristics of the collateral
ligaments of the human ankle joint. Foot Ankle 1988;8:234 42.
20. Earll M, Wayne J, Brodrick C, et al. Contribution of the deltoid ligament to ankle joint
contact characteristics: a cadaver study. Foot Ankle Int 1996;17(6):31724.
21. Close JR. Some applications of the functional anatomy of the ankle joint. J Bone Joint
Surg Am 1956;38:761 81.
22. Haddad SL, Myerson MS, Pell RF, et al. Clinical and radiographic outcome of revision
surgery for failed triple arthrodesis. Foot Ankle Int 1997;18(8):489 99.
23. Haskell A, Mann RA. Ankle arthroplasty with preoperative coronal plane deformity:
short-term results. Clin Orthop 2004;424:98 103.
24. Stamatis ED, Myerson MS. How to avoid specific complications of total ankle
replacement. Foot Ankle Clin 2002;7(4):178 84.
25. Song S, Lee S, OMalley M, et al. Deltoid ligament strain after correction of acquired
flatfoot deformity by triple arthrodesis. Foot Ankle Int 2000;21(7):5737.
26. Resnick RB, Jahss MH, Choueka J, et al. Deltoid ligament forces after tibialis posterior
tendon rupture: effects of triple arthrodesis and calcaneal displacement osteotomies.
Foot Ankle Int 1995;16(1):14 20.
27. Bluman EM, Myerson M. Stage IV posterior tibial tendon rupture. Foot Ankle Clin
2007;12(2):341 62.
28. Francis H, March L, Terenty T, et al. Benign joint hypermobility with neuropathy:
documentation and mechanism of tarsal tunnel syndrome. J Rheumatol 1987;14(3):
577 81.
29. Daniels TR, Lau JT, Hearn TC. The effects of foot position and load on tibial nerve
tension. Foot Ankle Int 1988;19(2):73 8.
30. Sobel M, Stern S, Manoli A. The association of posterior tibialis tendon insufficiency
with valgus osteoarthritis of the knee. Am J Knee Surg 1992;5:59.
31. Michelson JD, Helgemo SL Jr. Kinematics of the axially loaded ankle. Foot Ankle Int
1995;16(9):577 82.
32. Kelly IP, Nunley JA. Treatment of stage 4 adult acquired flatfoot. Foot Ankle Clin
2001;6(1):16778.
33. Karlsson J, Bergsten T, Lansinger O, et al. Surgical treatment of chronic lateral
instability of the ankle. A new procedure. Am J Sports Med 1989;17(2):268 73.
34. Karlsson J, Bergsten T, Lansinger O, et al. Reconstruction of the lateral ligaments of
the ankle for chronic lateral instability. J Bone Joint Surg Am 1988;70(4):581 8.
35. Alvarez RG, Marini A, Schmitt C, et al. Stage I and II posterior tibial tendon dysfunction
treated by a structured nonoperative management protocol: an orthosis and exercise
program. Foot Ankle Int 2006;27(1):2 8.
36. Wapner KL, Chao W. Nonoperative treatment of posterior tibial tendon dysfunction.
Clin Orthop 1999;365:39 45.
37. Chao W, Wapner KL, Lee TH, et al. Non-operative treatment of posterior tibial tendon
dysfunction. Foot Ankle Int 1996;17:736 41.
38. Bohay DR, Anderson JG. Stage IV posterior tibial tendon insufficiency: the tilted ankle.
Foot Ankle Clin 2003;8(3):619 36.

359

360

Smith & Bluman

39. Barrett GR, Meyer LC, Bray EW 3rd, et al. Pantalar arthrodesis: a long-term follow-up.
Foot Ankle 1981;1(5):279 83.
40. Chou LB, Mann RA, Yaszay B, et al. Tibiotalocalcaneal arthrodesis. Foot Ankle Int
2000;21(10):804 8.
41. Papa JA, Myerson MS. Pantalar and tibiotalocalcaneal arthrodesis for post-traumatic
osteoarthrosis of the ankle and hindfoot. J Bone Joint Surg Am 1992;74(7):10429.
42. Coughlin MJ, Schenck RC Jr, Grebing BR, et al. Comprehensive reconstruction of the
lateral ankle for chronic instability using a free gracilis graft. Foot Ankle Int 2004;25(4):
231 41.
43. Haddad SL, Myerson MS, Younger A, et al. Adult acquired flatfoot deformity. Foot
Ankle Int 2011;32(1):95111.
44. Anderson R, Davis W. Management of the adult flatfoot deformity. In: Myerson MS,
editor. Foot and ankle disorders. 1st edition. Philadelphia (PA): WB Saunders and Co;
2000. p. 101739.
45. Kitoaka HB, Luo ZP, An KN. Reconstruction operations for acquired flatfoot: biomechanical evaluation. Foot Ankle Int 1998;19(4):2037.
46. Goldner JL, Keats PK, Bassett FH 3rd, et al. Progressive talipes equinovalgus due to
trauma or degeneration of the posterior tibial tendon and medial plantar ligaments.
Orthop Clin North Am 1974;5(1):39 51.
47. Deland JT, de Asla RJ, Segal A. Reconstruction of the chronically failed deltoid
ligament: a new technique. Foot Ankle Int 2004;25(11):7959.
48. Ellis SJ, Williams BR, Wagshul AD, et al. Deltoid ligament reconstruction with peroneus longus autograft in flatfoot deformity. Foot Ankle Int 2010;31(9):7819.
49. Boyer MI, Bowen V, Weiler P. Reconstruction of a severe grinding injury to the medial
malleolus and the deltoid ligament of the ankle using a free plantaris tendon graft and
vascularized gracilis free muscle transfer: a case report. J Trauma 1994;36(3):454 7.
50. Raikin SM, Myerson MS. Surgical repair of ankle injuries to the deltoid ligament. Foot
Ankle Clin 1999;4(4):74553.
51. Kelikian K, Kelikian A. Disruptions of the deltoid ligament. In: Disorders of the ankle.
1st edition. Philadelphia: WB Saunders; 1985. p. 339 70.
52. McCormack AP, Ching RP, Sangeorzan BJ. Biomechanics of procedures used in
adult flatfoot deformity. Foot Ankle Clin 2001;6(1):15.23, v.
53. Hintermann B, Valderrabano V, Boss A, et al. Medial ankle instability: an exploratory,
prospective study of fifty-two cases. Am J Sports Med 2004;32(1):18390.
54. Haddad SL, Dedhia S, Ren Y, et al. Deltoid ligament reconstruction: a novel technique
with biomechanical analysis. Foot Ankle Int 2010;31(7):639 51.
55. Jeng CL, Bluman EM, Myerson MS. Minimally invasive deltoid ligament reconstruction
for stage IV flatfoot deformity. Foot Ankle Int 2011;32(1):2130.
56. Hurschler C, Provenzano PP, Vanderby R Jr. Scanning electron microscopic characterization of healing and normal rat ligament microstructure under slack and loaded
conditions. Connect Tissue Res 2003;44(2):59 68.
57. Walling AK. Chapter 19: stage IV flatfoot: alternatives to tibiocalcaneal or pantalar
procedures. In: Nunley JA, Pfeffer GB, Sanders R, et al, editors. Advanced reconstruction of the foot and ankle. Rosemont (IL): American Academy of Orthopaedic
Surgeons; 2004. p. 123 6.
58. Hall C, Agel J, Hansen S, et al. Complications in the treatment of stage IV adultacquired flatfoot. Presented at the American Orthopaedic Foot and Ankle 22nd
Summer Meeting. La Jolla (CA), July 14-16, 2006.

Youngs Procedure for the


Treatment of Valgus Flatfoot
Deformity Caused by a Posterior
Tibial Tendon Dysfunction, Stage II
Nuri Schinca, MDa,*, Alicia Lasalle, MDb, Josefina Alvarez, MDa

KEYWORDS
Flatfoot Supple flatfoot Tendon translocation
Tibialis posterior tendon insufficiency
KEY POINTS
Youngs procedure achieves a good muscle balance.
The tibial anterior tendon is not detached therefore, it works actively.
The technique improves the medial arch providing it with active components.
It causes an insufficiency of the ATT, which results in a predominance of the
peroneus longus.
It reinforces soft parts of the plantar and medial sector.
It does not sacrifice the joint mobility.

Flatfoot in adults caused by a dysfunction of the posterior tibial tendon is a frequent


pathology for which several therapies have been proposed, for instance tendon
transfers and bone procedures.
In Uruguay, in 1974, Dr Selva Ruiz and Professor Dr Liber Mauro1 began to use the
procedure described by Young2 to treat supple flatfoot in children and adolescents,
because they understood it contained an action mechanism that worked better than
other techniques on the pathophysiology of the deformity. This technique consisted
of the translocation of the tendon from the tibial anterior muscle, changing its route by
means of a tunnel carved in the navicular.
Even when the use of flexors is advised all over the world for the treatment of adult
flatfoot3 caused by a posterior tibial tendon (PTT) dysfunction, in view of the good
a
Foot and Ankle Surgery Service, British Hospital, Italia Avenue 2420, CP 11600, Montevideo,
Uruguay; b Foot and Ankle Surgery Service, Police Hospital, Jos Batlle y Ordoez 3574, CP
11700, Montevideo, Uruguay
* Corresponding author. San Salvador 2274/301. CP 11200, Montevideo, Uruguay.
E-mail address: florin@chasque.net

Foot Ankle Clin N Am 17 (2012) 227245


doi:10.1016/j.fcl.2012.02.002
1083-7515/12/$ see front matter 2012 Elsevier Inc. All rights reserved.

foot.theclinics.com

228

Schinca et al

results obtained with Youngs procedure in supple flatfoot deformity (FFD) of children and
adolescents, and considering that it was a technique we were familiarized with, we
decided to use it also for the treatment of FFD caused by PTT dysfunction in adults.
PATHOPHYSIOLOGY

Navarro4 claimed that the study of the anatomy alone is not enough to understand the
modifications that characterize the lesions or illnesses, as the pathology does not
only alter the form or structure, but it also affects the function. Understanding the
anatomy of the foot allows us to comprehend how the foot adapted itself to the
changes necessary for humans to stand and walk.
Among those changes, the medial arch was shaped, which is characteristic of
human beings and represents a static platform in the foot support and a propulsion
lever in its dynamic function. Mauro5 states that the dome is structured by arches.
These are true springs constituted by juxtaposed bone pieces, supported by
capsuleligament systems. The main support areas are eccentric: The calcaneus on
the back, and metatarsal heads and toes on the front. This submits the columns to
tensile and compressive forces. The compressive forces follow the trabecular bone
systems, the tensile forces are resisted by the plantar capsuleligament systems,
which constitute powerful tensors.
From a pathophysiologic point of view, we consider the foot as constituted of 4
links, 3 peripheral and 1 central (the talus). The peripheral links are active, that is to
say, they are subject to the influence of forces capable of unbalancing the foot. These
forces are represented by gravity and/or muscle driving forces.
1. The upper link is constituted by the tibial plafond and the ankle mortise, and it is
through.
2. This link that the pressure forces arrive. The mortise is not rigid, owing to the inferior
tibiofibular joint. The lateral malleolus is an external stabilizer of the hind foot, it is
stronger and more posterior, and it descends further than the medial malleolus.
3. The single posterior link is represented by the calcaneus, preceded in its dynamics
by a sole motor: The powerful triceps surae.
4. The anterior link, with a more complex structure, multisegmental, is constituted by
the pre talus foot.
5. The central link is represented by the talus. This bone has key characteristics:
a. It constitutes the central organ of the hind foot.
b. It is intra-articular, surrounded by wide cartilaginous surfaces.
c. It resembles a patella bone: embedded in the malleolar pincer, it receives and
distributes the forces transmitted by the tibia towards the plantar architecture.
d. It is a bone relais of the 2 ligament chains that originate in the malleoli6;
therefore, its stability is ensured by the fact that it is embedded and the richness
of the ligament insertions.
e. It is a passive link in the kinetic chain of the foot, because it does not offer
insertion for any muscle.
f. It is not the leader of any movement. Connected to the 3 peripheral links, it is
subjected to their influences.
THE JOINT COMPLEX OF THE HINDFOOT

The center of the hindfoot architecture is the talus. Around it, a ball joint is constituted,
the powerful peritalar joint. This joint comprises 3 other articulations, anatomically
separated but closely linked from the functional point of view:

Youngs Procedure for Valgus Flatfoot Deformity

1. The tibiofibular talar joint.


2. The posterior subtalar joint.
3. The anterior subtalar and the talonavicular joint, both form a complex structure.
The head of talus is embedded in a cotyloid fossa built of bone structures: Glenoid
cavity of the navicular and sustentacular facet, all of them structure the coxa pedis.
4. A ligament coat constituted by the spring ligament, covered by cartilage and
reinforced by the tibialis posterior, and the deltoid ligament or tibionavicular gleno
sustentacular ligament. The cavity is completed by the dorsal talonavicular
ligament and the Y-ligament. The navicular bone has a glenoid reception cavity,
wider than the corresponding articular surface of the talus, thus facilitating a broad
rotation movement.4
At the same time, the talus head is parabolic. This suggests a blocking mechanism
with the navicular, capable of reaching a higher position than the calcaneocuboid
joint, resulting in the development of the arch. Here is where the PTT takes part,
blocking the subtalar joint before the foot is lifted, so that the triceps elevates the heel
from the ground. The calcaneus plays an anatomic role on the joint complex of the
hindfoot, in its lateral sector, by means of the posterior subtalar joint. However, at the
same time, it is closely linked, from the functional point of view, with the anterior
sector of the coxa pedis, because its vast tuberosity articulates with the cuboid bone.
In this way, even when it belongs to the outer edge of the foot, it constitutes the
rotation sector of the hindfoot, together with the joints referred to as peritalar joint.
The muscle systems that control foot mobility are very complex:
1. All the muscles are polyarticular.
2. All of them are inserted distally through the front of the midtarsal joint, with the
exception of the triceps surae.
3. They all act on the heterokinetic cardan joint of the hindfoot. Their actions take
place in accordance with the flexion-extension and the inversion-eversion
movements.
4. According to Kapandjis scheme,6 the muscle dynamic in relation to the cardan
axis allows for the following muscle categorization:
a. Anterior-external: Dorsiflexors evertors: Extensor digitorum longus and peroneus brevis;
b. Posterior-external: Plantar flexors evertors: Peroneus brevis and longus, which
control the foot curvature;
c. Posterior-internal: Plantar flexorsinvertors: Triceps surae, tibialis posterior and
flexor digitorum; and
d. Anterior-internal: Dorsiflexorssupinators: Extensors hallux and tibial anterior.
As we see, there are synergyantagonism actions. The muscle system of the first
metatarsal is constituted by 2 antagonist muscles: The tibialis anterior tendon (ATT)
that lifts it and gets it into a horizontal position, determining the supination of the entire
forefoot,5 and the peroneus longus, which is a depressor of the first metatarsal, and
abductor and pronator of the forefoot. It depresses the forefoot and moves it outward
joining the internal and external metatarsals, superimposing them, facilitating and
optimizing the plantar flexion function of the triceps surae.
The unbalance of this couple is among the pathogenic factors of the flatfoot. The
main cause of adult flatfoot is the loss of the muscular function. Several pathogenic
mechanisms combine causing isolated or combined effects.7,8 The flatfoot is not only
a depression or reduction in height of the anteroposterior dome in standing position,
but is a clinical radiologic complex that adds to this remarkable sign, the valgus

229

230

Schinca et al

position of the heel, the emergence of the head of the talus in the inner edge, the
retraction of the Achilles tendon, and the abduction and supination of the forefoot.
The injured PTT loses its supportive and cavus action, producing a valgus hindfoot
and a flattening of the foot arch. The PTT is no longer able to block the hindfoot in
varus to allow the Achilles tendon and the hindfoot to lift the heel, thus becoming
unstable. Instead of being transmitted to the anterior and posterior pillars, the forces
arriving to the foot are now affecting the conflict and stress area of the inner edge.
This results in a progressive elongation and/or rupture of the medial and plantar
capsular ligamentous complex, becoming an aggressor to the static of the foot.9
The valgus displacement of the calcaneus restrains the main support of the talus,
and allows for its inwardforward translocation in plantar flexion, misdirecting the
entire peritalar joint: The tibiotalar, subtalar, and midtarsal relation. With the heel bone
in pronation, the Achilles tendon will not produce a supinator effect over the hindfoot
(as it occurs with a normal foot); instead, because it is inserted in a valgus heel bone,
it becomes a pronator.
The triceps has the property of reinforcing the pronation or supination movement,
responding to the behavior of the varus or valgus calcaneus. What is more, with time,
it retracts, resulting in a hindfoot equinus. Keeping the heel in valgus and the ankle in
plantar flexion shortens the tendon and, with time, diminishes the push off or lifting
force of the triceps. Delayed heel-off is characteristic of valgus flatfoot stage II.10 15
The forefoot is displaced in abduction, with the navicular and cuboid located outside
the posterior tarsus. A subluxation of the talonavicular joint takes place. Subsequently, the metatarsal shafts may try to compensate by assuming an abduction and
supination position, and this breaks the talusnavicular cuneiform metatarsal axis.
The valgus hindfoot secondarily determines an increase of the pressure under the
head of the first metatarsal, which becomes horizontal. The retraction of the tibialis
anterior aggravates the situation. With the hindfoot in valgus deviation, the supination
is caused by accommodative changes that allow the medial and lateral columns of the
forefoot to remain in contact with the ground.16
There are authors, like Hintermann and colleagues,17 who state that the supination
attitude of the forefoot appears earlier than the insufficiency of the tibialis posterior,
so that in the semiological analysis, this sign appears before the tiptoe sign, which
tells of the importance of searching for the first ray elevation sign.
We believe that the first ray elevation sign is secondary to the appearance of the
flatfoot deformity. What aspects should be considered by a technique that pretends
to correct the valgus flatfoot deformity caused by a PTT dysfunction?
1.
2.
3.
4.
5.

Insufficiency of the PTT.


Valgus hind foot.
Elongation or rupture of the internal capsuleligament structures.
Forefoot supination.
Forefoot abduction.

How is it possible to correct this muscle imbalance, caused by the posterior tibialis
insufficiency and how is Youngs procedure able to achieve it?
TECHNIQUE

We perform a medial approach from the medial maleollus up to the ATT distal
insertion. The PTT is explored. If there is substantial damage, it is resected; otherwise,
its borders are fixed to other flexors, remembering that they are synergic. The ATT is
then freed beginning in its distal insertion until the ankle, moving it without detaching
it (Figs. 13).

Youngs Procedure for Valgus Flatfoot Deformity

Fig. 1. The ATT.

A horizontal cut is performed over the inner side of the navicular, at a central point
between the dorsal and the sole of the foot, following its horizontal axis. Two
osteoperiosteal flaps are carved upper (Fig. 4) and lower (Fig. 5) using a chisel
and trying to carve them deep, so that the navicular remains with the entire cancellous

Fig. 2. Proximal release of the ATT up to the ankle.

231

232

Schinca et al

Fig. 3. Distal release of the ATT, without detaching it.

bone exposed. A slot is sculpted along the navicular, as plantar as possible (Fig. 6).
The proximal side of the navicular slot is made round to avoid a possible lesion
caused by the sharp edge to the ATT while being translocated to its new position.
The ATT is lowered and located in the slot sculpted in the navicular in the inner
edge of the foot (Fig. 7) and both osteoperiosteal flaps are closed, tensing them and
suturing them over the tendon (Fig. 8). The ATT is pulled underneath the navicular,
because it is a tremendous support for the medial arch of the foot.
Depending on the surgeons preference, it is possible to add in the depth of the
canal an anchor, with 1 or 2 metal spears, if there are doubts that the osteoperiosteal
flap of fragile structure may not be able to keep the tendon in its new position. It is
worth remembering that the bone structure of the adult patients is not the same as
that of younger patients, for whom the procedure was originally described.

Fig. 4. Carving or creation of the great upper and lower flaps.

Youngs Procedure for Valgus Flatfoot Deformity

Fig. 5. The lower flap leans the entire medial capsularligamentous apparatus.

The patient is kept in postoperative care immobilized with an open cast. After 15
days, the splint and the skin stitches are removed, and the patient is immobilized
again, this time with a closed cast with the ankle in neutral position. One month after
the operation, the patient is allowed to bear weight completely. Two months after
surgery, the cast is removed and an Arizona splint is applied for another month, and
the patient begins hydrotherapy to start mobilizing the foot.
It was necessary to modify Youngs original procedure. Youngs procedure2 sculpts a
vertical slot in the navicular, so that the ATT has a horizontal trajectory under the navicular

Fig. 6. Creation of the slot in the navicular.

233

234

Schinca et al

Fig. 7. The ATT is lowered, changing its route.

and vertical in its proximal sector. However, we found it difficult to lower the ATT given
that, even when this foot is mobile, it does not present the flexibility of a child. Therefore,
it was necessary to make a horizontal trajectory, keeping it as plantar as possible so that
it supports the internal and plantar capsular and ligamentous apparatus.
The second reason why the trajectory was altered was that the majority of these
patients having a very osteoporotic navicular, which may break while trying to sculpt
the vertical slot. Is this isolated procedure enough to correct all the components of the

Fig. 8. Overlapping of slots, tensing the medial capsular ligamentous apparatus.

Youngs Procedure for Valgus Flatfoot Deformity

pathology? The purpose of this tendinous translocation is to find a substitute for the
insufficient PTT and achieve an adequate muscle balance.
1. We must take into account that the ATT is not detached, so its function mechanism
is still active. Only its distal 10 or 15 cm where moved to redirect it, so that the
proximal area has a more vertical trajectory in the inner edge of the foot. Because
of this trajectory change, its lifting action increases the upward pull of the inner
arch, reinforcing the action that the tibialis posterior had in its sling function at the
level of the navicular.18
2. The horizontal trajectory of the ATT is performed with an almost plantar route, and
it is covered by osteoperiosteal flaps that are tensed on top of it, constituting a
powerful inner capsulartendinousligamentous support.
3. The ATT translocation makes this tendon unable to lift the first metatarsal, causing
it to pull the navicular. Therefore, an insufficiency of the ATT is created, which results
in a predominance of the peroneus lateral longus that descends and prones the
forefoot causing an active increase in the height of the arch in the inner side of the foot.
What are the advantages over other procedures? First, The muscle used is not
detached; therefore, it works actively. Second, it achieves a good muscle balance,
which results in an improvement of the medial arch providing it with active
components, reinforcing the action of the PTT and granting an insufficiency of the
ATT, which results in a predominance of the peroneus lateral longus. Third, it
reinforces soft parts of the plantar and medial sector. Fourth, it does not sacrifice
the joint mobility. Finally, it is an efficient procedure that acts anatomically and
functionally.
COMPARISON WITH OTHER PROCEDURES
Flexor Digitorum Longus

Although the flexor digitorum longus (FDL) is a plantar flexor and supinator,
moving it to the navicular results in a reinforcement of the insufficient PTT, but
diminishes the flexorplantar action, which is very important.19 23 It is worth
remembering that the FDL is one third of the size of the PTT and capable of
applying one third of the force of the PTT; therefore the question posed is this: Is
this tendon enough to transpose or perform a tenodesis?
Comparative Force and Tendon Excursion

Silver and associates24 affirm the values in Table 1. The ATT provides for 80% of the
dorsiflexion force of the ankle. Hintermann and co-workers25 conducted an in vitro

Table 1
Comparative force and tendon excursion
Force

Excursion

Tibialis anterior tendon

5.6

2.9

Posterior tibial tendon

6.4

1.6

Flexor hallucis longus

3.6

1.7

Flexor digitorum longus

1.8

1.2

Peroneus longus

5.5

1.6

Peroneus brevis

2.6

1.4

Achilles

49.1

4.0

235

236

Schinca et al

study of foot movement and tendon excursion. They speculated that rotation is
provided by a low insertion and high excursion, whereas the stabilizing function is
more effective with several insertions acting to move bones together. The stabilizing
function was attributed mainly to those muscles with numerous insertions and low
maximum tendon excursion (eg, the PTT). The major contribution of the PTT to foot
stability results from the locking-in mechanism of the midfoot.
The main function of the flexor hallucis longus and FDL may be attributed to toe
flexion extension. They additionally exert an inverting effect on the rearfoot. They do
not have such influence on midfoot motion; their stabilizing function over of the foot
eversioninversion axis may be substantially less than the one of the tibialis posterior.
In relation to the tibialis anterior inversion eversion axes, it is not as high as the one
the tibialis posterior can produce.
The ATT has an invertor moment about 0.6 when the referential is tibialis posterior
(as the FLD). Its moment arm is twice as long with regards to the flexion extension
axis, making it a strong extensor. In an attempt to increase the leverage as an invertor,
ATT transfer may, therefore, not be an adequate method for the treatment of FFD.
Such surgery may result additionally in the loss of the extensor function.
Our hypothesis is that with the translocation, the force of the ATT will be canalized
as an invertor and the loss of its extensor force will help to correct the forefoot
supination. Cobb26 described a method of reconstruction in Johnson and Strom type
II PTT27 dysfunction, performing a transfer of a split of the ATT.28,29 In 1990, Helal30
described satisfactory results in 5 of 8 patients with use of the split ATT procedure in
patients with a stage 2 PTT.
The goal of the Cobb procedure26 is to reconstruct a functional ATT associated
with a torn PTT. In the Cobb method, half of the ATT is left attached at its insertion and
it is passed throughout a hole drilled in the cuneiform to the PTT compartment,
performing an anastomosis in the proximal end of the PTT. The advantage of the
Cobb procedure over the FDL is that the loss of a plantar flexor is avoided, what at
the same time, generates a partial insufficiency of the ATT when dividing it in half.
Later, this produces the fall of the first metatarsal and the pronation of the forefoot.
Mann and Thompson22 affirm that performing the slot in the cuneiform instead of
the navicular increases the lever arm of the transferred tendon and prevents any
collapse in the navicular cuneiform, which is sometimes associated with the collapse
of the talonavicular joint.
Baravarian and co-workers31 said that the Cobb technique is not ideal for all
patients because if the muscletendinous unit of the tibialis posterior is left nonfunctional or severely scarred owing to prolonged disuse, the anastomosis will not have a
proper functioning because it depends on the condition of the tibial posterior muscle
(PTT) for its action. Valdebarrano and associates32 carried out a study to determine
the recovery potential of the PTT after tendon rupture. They state that the recovery
potential of the PTT seemed to be significant even after delayed repair.
Knupp and Hinterman33 carried out a prospective study with the Cobb procedure
and included 22 consecutive patients with supple flatfoot deformity caused by PTT
dysfunction. Average follow-up was 24 months. The overall clinical results obtained
were excellent (41%), good (54%), fair (5%), and poor (none). They stated that the
advantage is that the Cobb procedure decreases the tension of the ATT.
This dynamic correction may allow the patient to adapt the forefoot to the ground
as required. The Cobb procedure may, to a certain extent, correct the functional
amputation of the first ray that occurs with the forefoot supination deformity.33 The
resulting improved and preserved weight distribution in the forefoot is thought to give
the patient better control and stability in stance of gait.

Youngs Procedure for Valgus Flatfoot Deformity

Fig. 9. Preoperative left FFD. Too many toes sign.

BONE PROCEDURES THAT MAY BE ASSOCIATED WITH YOUNGS TECHNIQUE


Calcaneal Osteotomy

Osteotomies provide the foot with a more stable posture. It allows the transferred or
translocated tendon to function more effectively.34,35 The biomechanical improvement of

Fig. 10. At 7 years follow up, the valgus hindfoot correction persists and the too many toes
sign disappears.

237

238

Schinca et al

a medial translocation of the calcaneus is believed to be a result of the restoration of the


gastrosoleus medial axis with the ankle joint.36 As a result of the change of the subtalar
vector, the hindfoot forces change diminishing the stress in the internal repair, thereby
protecting the tendinous transference.
This osteotomy is extra-articular, and for that reason it does not impact or
compromise any joint, avoiding rigidity. As we said, a fully movable foot is fundamental for Youngs procedure to be effective. Moreover, the internal displacement of
the calcaneus bone reduces stress over the deltoid ligament. For all these reasons,
we prefer this technique over other procedures.
Lengthening of the External Column

This technique is used to correct forefoot abduction, produced in FFD, determining a


shortening of the external column.37 40 The technique is carried out by lengthening
the mentioned sector by distraction, placing a graft in the calcaneal osteotomy, or
elongation at the level of the calcaneal cuboid joint and placing a graft in it. Evans,
modified41 by Mosier,42 states that the anterior calcaneal osteotomy permits the
internal rotation by means of the heel, through heel bone adduction with simultaneous
supination and plantar flexion. He also documented that it corrects the radiologic
values of FDD, even the talonavicular coverage; however, the mechanism is unknown.
What are the drawbacks of this technique? It is a purely morphologic correction;
shortening of the external column is caused by a functional lengthening of the medial

Fig. 11. At 7 years follow-up. Good morphology and no signs of forefoot abduction.

Youngs Procedure for Valgus Flatfoot Deformity

column produced by the insufficiency of the PTT and the medial capsule ligament
elements. Our hypothesis is that if we are able to control the medial functional
insufficiency we will achieve the coverage of the head of the talus with the navicular,
not anatomically but functionally.
Following authors such as Coetzee and Castro,36 we agree that this technique
causes improper rotation of the forefoot in varus and supination, and probably
worsens the supination already present in the forefoot, which is what Youngs
procedure tries to correct. Lengthening the external column increases pressure over
the joints, which may cause a rigidity not adequate when combining this technique
with Youngs, given that Youngs procedure requires a mobile foot to obtain
successful results. Mosier and colleagues42 state that 65% of patients present with
calcaneocuboid arthritis, loss of correction over time, and loss of mobility in the rest
of the joints of the hind foot owing to such arthritis.
The placement of the graft in the external sector,36 near the posterior subtalar joint,
produces an impact over this joint, generating sinus tarsi pain. Nonunion is still the
most common complication. It occurs in 10% to 20% of cases and usually requires
a revision procedure.36
Cuneiform Osteotomy With Bone Graft

Supination of the forefoot in this pathology is a complication that has not been
considered by many authors. However, many others propose acting over the first
cuneiform to correct it, by means of a cuneiform osteotomy with bone graft. This
technique43 performs an osteotomy in the center of the first cuneiform, placing a
triangular graft in the dorsal base, trying to provide flexion to the unstable medial

Fig. 12. Absence of tip toe sign.

239

240

Schinca et al

Fig. 13. With the foot in varus, it is possible to observe the new direction of the ATT.

column of the foot and correct supination. However, we insist that it is purely an
anatomic correction, because it does not add motors to contribute with plantar
flexion, as Youngs procedure does, generating an insufficiency of the anterior tibialis
tendon.
Talonavicular Arthrodesis

Youngs procedure is contraindicated in a foot with some sort of rigidity.44

Fig. 14. Forefoot pronation.

Youngs Procedure for Valgus Flatfoot Deformity

Arthroerisis

The biomechanics of the implant function have not been fully elucidated. They present
the problem that they may determine, eventually, a decrease in the subtalar
mobility.45,46 Schon47 says that the arthroerisis seems beneficial with a low-risk
profile. The high implant removal rate is a concern, but implant removal brings about
a resolution of the pain without loss of correction.
In the future, better studies will help to define the outcomes, and determine whether
it does not diminish the subtalar mobility or causes its arthritis.
MATERIAL AND METHODS

We retrospectively studies 35 patients with supple flatfoot deformity caused by PTT


dysfunction treated with Youngs procedure.2 The operations were carried out
between 1994 and 2010 with a minimum follow-up of 1 year, a maximum follow-up of
17 years, and median of 6.5 years.
There were 2 groups with similar characteristics in relation to gender, age, weight;
and previous treatment (Youngs procedure, 14 patients [42.4%] among others).
Youngs procedure and medial displacement calcaneal osteotomy 21 patients
(57.6%). The American Orthopaedic Foot and Ankle Society score system48 was used
in the analysis of the results. This score increased from a preoperative value of
50.77% to 94.14% points at the latest follow-up. The postoperative score for Youngs
technique was 92.76 and Youngs technique and calcaneal osteotomy was 95 points
(Figs. 9 16).
The overall clinical results were graded as excellent (American Orthopaedic
Foot and Ankle Society score, 100%) in 11 (31.43%). In this group, all patients
were treated with Youngs technique and calcaneal osteotomy. Results were good
(score range, 90 99) in 17 (48.57%), fair (score range, 80 89) in 6 (17.14%), and

Fig. 15. Uncovered talus. Forefoot abduction.

241

242

Schinca et al

Fig. 16. Covered talus, 7 years postoperative.

poor (score, 80) in 1 (2.86%). The poor outcome was the only patient with PTT
rupture. Clinical ATT dysfunction occurred in the early postoperative period. The
patients had decreased tibial anterior tendon power, but after 6 months none had
clinical ATT dysfunction.
No patient had first ray metatarsalgia caused by an ATT dysfunction at latest
follow-up. Forefoot postoperative supination occurred in 6 patients (17.1%). In these
cases, Youngs technique was performed without medial displacement calcaneal
osteotomy. Although some patients had persistent forefoot supination and hindfoot
valgus, 97% were satisfied with the results. They manifested they could feel the
ground and, after surgery, recover foot control.
Of our cases, 85% had no pain and only 15% occasional and moderate pain. All
patients were able to perform daily living activities without the use of canes and
only 12% had some limitations at recreational activities. Postoperatively, 33
patients (94.2%) were able to wear shoes without insoles; 6 (17.1%) preferred to
have insoles.
Comparing both groups, Youngs procedure is useful to reconstruct a supple FFD,
but additional procedures, such as medial displacement calcaneal osteotomy, should
be considered to correct the entire deformity. Of our patients, 97% did not experience
a progression of the flatfoot deformity at follow-up. This procedure improves
functionality, gait capability, and subjective stability, which may be explained by an
improved control of proprioception. In all, 97.1% of patients are satisfied with their
results. We believe that Youngs technique respects the anatomy and biomechanics
of the foot to reach the necessary muscular balance.

Youngs Procedure for Valgus Flatfoot Deformity

SUMMARY

Youngs procedure contains an action mechanism that works better than other
techniques on the pathophysiology of FFD. It respects the anatomy and biomechanics of the foot to reach the necessary muscular balance. The benefits of this technique
include that the ATT is not detached, so its function mechanism is still active; the new
trajectory of the ATT provides a powerful sling function at the level of the navicular;
and the horizontal trajectory of the ATT and the osteoperiosteal flaps constitute a
powerful inner capsulartendinousligamentous support. What is more, an insufficiency of the ATT is created, which results in a predominance of the peroneus lateral
longus, that descends and prones the forefoot. Additional procedures, such as medial
displacement calcaneal osteotomy, should be considered to correct the entire
deformity. The combination of these techniques do not sacrifice the joint mobility.
ACKNOWLEDGMENTS

The authors thank Mara Noel Maquieira for translation of the manuscript.
REFERENCES

1. Ruiz S, Mauro L. Tratamiento quirrgico del pie plano. Revista de Ciruga del Uruguay
1974;43:5435.
2. Young CS. Operative treatment of pes planus. Surg Gynecol Obstet 1939;68:1099
101.
3. Van Boerum D, Sangeorzan B. Biomechanics and pathophysiology of flatfoot. Foot
Ankle Clin 2003;8:419 30.
4. Navarro A. Nuevos conceptos sobre la fisiologia del pie. Deducciones patolgicas
Anales del Instituto de Clnica quirrgica y ciruga experimental. Tomo 5. Montevideo:
Rosgal; 1943. p. 9 81.
5. Mauro L. Biomecnica del pie plano esttico en nios y adolescentes. In X Congreso
Latinoamericano de Ortopedia y Traumatologa. Ro de Janeiro; 1977.
6. Kapandji JA. Miembro inferior. In: Cuadernos de fisiologa articular, vol. 2. Barcelona:
Toray-Masson; 1982. p. 136 219.
7. Llorens Isidro A, Nuez Samper M, Llanos Alcazar LF, editors. Filogenia del pie. In:
Biomecnica Medicina y Ciruga del pie. Barcelona: Masson; 1997. p. 8 13.
8. Ness ME, Long J, Marks R, et al. Foot and ankle kinematics in subjects with PTT
dysfunction. Gait Posture 2008;27:3319.
9. Cracchiolo A. Evaluation of spring ligament pathology in patients with posterior tibial
tendon rupture, tendon transfer and ligament repair. Foot Ankle Clin 1997;2:297309.
10. Giza E, Cush G, Schon L. The flexible flat foot in the adult. Foot Ankle Clin 2007;12:
25172.
11. Huang CK, Kitaoka HB, An KN. Biomechanics evaluation of longitudinal arch stability.
Foot Ankle Int1993;14:3537.
12. Kitaoka HB, Ahn TK, Luo ZP, et al. Stability of the arch of the foot. Foot Ankle Int;
1997;18:644 8.
13. Otis JC. Clinical and applied biomechanics. In: Myerson M, editor. Foot and ankle
disorders. Philadelphia: WB Saunders Company; 2000. p. 18194.
14. DiGiovanni C, Langer P. The role of the isolated gastrocnemius and combined Achilles
contractures in the flat foot. Foot Ankle Clin 2007;12:36379.
15. Neptune RR, Kantz SA, Zajac FE. Contributions of the individual ankle plantar flexors
to support forward progression and swing initiation during walking. J Biomech
2001;34:138798.

243

244

Schinca et al

16. Bluman EM, Title CI, Myerson MS. Posterior tibial tendon rupture. A refined classification system. Foot Ankle Clin 2007;12:233 49.
17. Hintermann B, Gachter A. The first metatarsal rise sign. A simple sensitive sign of tibial
posterior tendon dysfunction. Foot Ankle Int 1996;17:236 41.
18. El-Tayeby HM. The severe flexible flatfoot: a combined reconstructive procedure with
rerouting of the tibialis anterior tendon. J Foot Ankle Surg 1999;38:419.
19. DiPaola M, Raikin S. Tendon transfers and realignment osteotomies for treatment of
Stage II posterior tibial tendon dysfunction. Foot Ankle Clin 2007;12:273 82.
20. Guyton G, Jeng C, Krieger L, et al. Flexor digitorum longus transfer and medial
displacement calcaneal osteotomy for posterior tibial tendon dysfunction. A middleterm clinical follow up. Foot Ankle Int 2001;22:627 63.
21. Thompson MR. Rupture of the posterior tibial tendon causing flat foot. Surgical
treatment. J Bone Joint Surg Am 1985;67:556 61.
22. Myerson MS, Badekas A, Schon LC. Treatment of stage II posterior tibial tendon
deficiency with flexor digitorum longus transfer and calcaneal osteotomy. Foot Ankle
Int 2004;25:44550.
23. Parsons S, Naim S, Richards P, et al. Correction and prevention of deformity in type
II tibialis posterior dysfunction. Clin Orthop Relat Res 2010;468:102532.
24. Silver RL, de la Garza J, Rang M. The myth of muscle balance. A study of relative
strengths and excursions of normal muscles about the foot and ankle. J Bone Joint Br
1985;67:4327.
25. Hintermann B, Nigg BM, Sommer C. Foot movement and tendon excursion: an in
vitro study. Foot Ankle Int 1994;15:386 95.
26. Cobb N. Tibialis posterior disorders. In: Helal B, Rowley DI, Cracchiolo A III, et al,
editors. Surgery of disorders of the foot and ankle. London (UK): Martin Dunley; 1996.
p. 291301.
27. Johnson KA, Strom DE. Tibialis posterior tendon dysfunction. Clin Orthop Rel Res
1989;239:196 206.
28. Basmajian JV, Stecko G. The role of muscles in arch support of the foot J Bone Joint
Surg 1963;45:1184 90.
29. Biga N, Moulier D, Mabit C. Pied plat valgus statique. EMQ. Paris: Elsevier; 1999.
30. Helal B. Cobb repair for tibialis posterior tendon rupture. J Foot Surg 1990;29:349 58.
31. Baravarian B, Zgonis T, Lowery C. Use of the Cobb procedure in the treatment of
posterior tibial tendon dysfunction. Clin Podiatr Med Surg 2002:19:371 89.
32. Valdebarrano V, Hintermann B, Wischer T, et al. Recovery of the posterior tibial muscle
after late reconstruction following tendon rupture. Foot Ankle Int 2004;25:8595.
33. Knupp M, Hintermann B. The Cobb procedure for treatment of acquired flatfoot
deformity associated with stage II insufficiency of the posterior tibial tendon. Foot
Ankle Int 2007;28:416 21.
34. Hiller L, Pinney SJ. Surgical treatment of acquired flatfoot deformity: what is the state
of practice among academic foot and ankle surgeons in 2002? Foot Ankle Int
2003;24:7015.
35. Vora AM, Tien TR, Parks BG, et al. Correction of moderate and severe acquired
flexible flatfoot with medializing calcaneal osteotomy and flexor digitorum longus
transfer. J Bone Joint Surg Am 2006;88:1726 34.
36. Coetzee C, Castro M. The indications and biomechanical rationale for various hindfoot
procedures in the treatment of posterior tibialis tendon dysfunction. Foot Ankle Clin
2003;8:4539.
37. Anderson RB, Davis WH. Calcaneocuboid distraction arthrodesis for the treatment of
the adult acquired flatfoot. The modified Evans procedure. Foot Ankle Clin 1996;1:
279 94.

Youngs Procedure for Valgus Flatfoot Deformity

38. Krans A, Louwerens JW, Anderson P. Adult acquired flexible flatfoot, treated by
calcaneo-cuboid distraction arthrodesis, posterior tibial tendon augmentation, and
percutaneous Achilles tendon lengthening: a prospective outcome study of 20
patients. Acta Orthop 2006;77:156 63.
39. Sands A, Tansey JP. Lateral column lengthening. Foot Ankle Clin 2007;12:30130.
40. Sangeorzan B, Mosca V, Hansen ST. Effect of calcaneal lengthening on relationships
among the hindfoot, midfoot, and forefoot. Foot Ankle Int 1993;14:136 41.
41. Evans D. Calcaneus valgus deformity. J Bone Joint Surg Br 1975;57:270 8.
42. Mosier SM, Pomeroy G, Manoli A. Pathoanatomy and etiology of posterior tibial
tendon dysfunction. Clin Orthop Rel Res 1999;365:1222.
43. Tankson CJ. The Cotton osteotomy: indications and techniques lengthening. Foot
Ankle Clin 2007;12:309 16.
44. Cohen B, Ogden F. Medial column procedures in the acquired flatfoot deformity. Foot
Ankle Clin 2007;12:287301.
45. Giannini S, Ceccarelli F, Benedetti M. Surgical treatment of flexible flatfoot in children:
a four years follow up study. J Bone Joint Surg Am 2001;83(Suppl 2):739.
46. Viladot A. Surgical treatment of the child flatfoot. Clin Orthop1992;283:34.
47. Schon L. Subtalar arthroerisis: a new exploration of an old concept. Foot Ankle Clin
2007;12:329 39.
48. Kitaoka HB, Alexander IJ, Adalaar RS, et al. Clinical rating systems for the anklehindfoot, midfoot, hallux and lesser toes. Foot Ankle Int 1994;15:349 53.

245