Anda di halaman 1dari 8


discussions, stats, and author profiles for this publication at:

Treatment of Lisfranc Fracture-Dislocations with

Primary Partial Arthrodesis

Article in Foot & Ankle International January 2012

DOI: 10.3113/FAI.2012.0050 Source: PubMed


15 237

5 authors, including:

Keith R Reinhardt Luke S Oh

Northwell Health Harvard Medical School


All content following this page was uploaded by Luke S Oh on 28 January 2015.

The user has requested enhancement of the downloaded file.

Copyright 2012 by the American Orthopaedic Foot & Ankle Society
DOI: 10.3113/FAI.2012.0050

Treatment of Lisfranc Fracture-Dislocations with Primary Partial Arthrodesis

Keith R. Reinhardt, MD; Luke S. Oh, MD; Patrick Schottel, MD; Matthew M. Roberts, MD; David Levine, MD
New York, NY

ABSTRACT Level of Evidence: III, Retrospective Comparative Study

Background: The optimal method of treatment for Lisfranc Key Words: Tarsometatarsal; Lisfranc; Primary Partial Arthro-
desis; Fracture-Dislocation; Midfoot
fracture-dislocations remains controversial, and the role of
primary partial arthrodesis for combined osseous-ligamentous
Lisfranc injuries is unclear. This study reviewed the outcomes INTRODUCTION
of Lisfranc injuries treated by primary partial arthrodesis.
Methods: Patients who underwent primary partial arthrodesis Lisfranc (tarsometatarsal) fracture-dislocations are un-
for a primarily ligamentous or combined osseous and ligamen-
common injuries, but have classically been a difficult
tous Lisfranc fracture-dislocation were reviewed retrospectively
problem for the orthopaedic surgeon. Because of the
and assessed at followup according to radiographic, clinical and
high potential for chronic disability, prompt diagnosis and
standardized patient-based outcomes. Twenty-five patients (12
ligamentous, 13 combined), median age of 46 (range, 20 to 73)
precise anatomic reduction are essential.2,4,6,9,12,15,23 The
years, were followed for an average of 42 (range, 24 to 96) various treatments for this injury include closed reduc-
months. Results: The average American Orthopedic Foot and tion and immobilization, closed reduction with percutaneous
Ankle Society (AOFAS) score was 81 points (scale 0 to 100), with pinning,10,21,27 open reduction and internal fixation (ORIF)
patients in general losing points for mild pain, limitations of either trans-articular2,3,15,18 or extra-articular,1 flexible liga-
recreational activities, and fashionable footwear requirements. mentous fixation,5 primary partial (first three tarsometatarsal
There was no statistical difference between ligamentous and [TMT] joints) arthrodesis,11,16,18 and primary complete (all
combined injuries with regard to the physical or mental compo- five TMT joints) arthrodesis.9,12
nent scores on the SF-36. At latest followup, patients reported an Most authors have traditionally considered open reduction
average return to 85% of their preinjury activity level (range, and internal fixation to be the treatment of choice for Lisfranc
50% to 100%). Twenty-one patients (84%) expressed satis- fracture-dislocations.3,6,8,15 22,27 More recently, however,
faction with their outcome and at latest followup, the mean some authors have advocated primary partial arthrodesis as
visual analog pain scale (VAS) score was 1.8 out of 10 (range, an alternative to ORIF.11,16 Previously, partial or complete
0 to 8). Three patients showed radiographic signs of post- arthrodesis was not advocated as a primary treatment for
traumatic arthritis of adjacent joints. Conclusion: Treatment of Lisfranc fracture-dislocations, and most considered them to
both primarily ligamentous and combined osseous and ligamen-
be salvage operations,14,17,19,24,25 However, trends toward
tous lisfranc injuries with primary partial arthrodesis produced
worse results using ORIF in subsets of patients with primarily
good clinical and patient-based outcomes.
ligamentous Lisfranc injuries led to the proposal of partial
No benefits in any form have been received or will be received from a commercial arthrodesis as a primary treatment in these patients,15 which
party related directly or indirectly to the subject of this article. has been supported by recent Level I evidence.11,16
Corresponding Author:
The applicability of primary partial arthrodesis for com-
Keith R. Reinhardt, MD bined osseous-ligamentous injuries in addition to the pri-
Hospital for Special Surgery marily ligamentous Lisfranc injuries remains unclear. The
Orthopaedic Surgery
purpose of this study was to evaluate and compare the clinical
535 East 70th Street
8th Floor and radiographic outcomes of patients with primarily liga-
New York, NY 10021 mentous and combined osseous-ligamentous Lisfranc injuries
E-Mail: after operative treatment with primary partial arthrodesis.
For information on pricings and availability of reprints, email Secondary aims of the study were to determine if outcomes
or call 410-494-4994, x232. following primary partial arthrodesis were affected by the

Foot & Ankle International/Vol. 33, No. 1/January 2012 PRIMARY ARTHRODESIS OF LISFRANC INJURIES 51

subtype of Lisfranc injury sustained, and to identify poten- injury included twisting a plantarflexed forefoot (n = 13),
tially predictive variables of outcomes following partial motor vehicle accidents (n = 5), falls (n = 4), and crush
arthrodesis regardless of injury subtype. injury (n = 3). Five patients had involvement of all five
tarsometatarsal joints, four had involvement of four, twelve
had involvement of three, and four had involvement of
two. In the combined osseous-ligamentous group, 10 patients
(77%) had metatarsal fractures, eight patients (62%) had
Operative records of two foot and ankle fellowship-trained associated cuneiform fractures, and two patients (15%)
orthopaedic surgeons (M.M.R. and D.S.L.) were reviewed to had associated fractures of the cuboid. In the primarily
identify patients who underwent primary partial arthrodesis ligamentous group, two patients (17%) had an associated
for a Lisfranc fracture-dislocation between 1998 and 2006. fleck sign. The demographic and clinical data according to
During this time period, primary partial arthrodesis was injury pattern subtype is further elucidated in Table 1.
performed on all patients presenting to our institution with There were 12 patients (four males, eight females) in the
an operative Lisfranc injury. Patients were included in the primarily ligamentous injury group with a median age of 51.5
study if they were skeletally mature at the time of injury, (range, 20 to 71) years. In this group, one patient was on a
had undergone primary arthrodesis of the first, second, bisphosphonate for osteoporosis, one patient was a smoker,
and/or third tarsometatarsal joints within 6 weeks of the and one was on chronic steroids. In the combined osseous-
time of injury, and had a minimum followup of 2 years. ligamentous injury group there were 13 patients (four males,
Exclusion criteria were prior ipsilateral foot or ankle surgery nine females) with a median age of 46 (range, 20 to 73) years.
or infection, concurrent ipsilateral lower extremity fractures, In this group, one patient was a smoker, one was on chronic
chronic injury of greater than 6 weeks duration, and any steroids, and two other patients were both smokers and on
previous attempt at operative management of the same injury. chronic steroids. The groups were similar with regard to
All participating patients signed informed written consent the demographic variables of age, gender, comorbidities, and
and the study protocol was approved by the Institutional extent of injury as indicated by the number of tarsometatarsal
Review Board. joints involved (p > 0.05).
Thirty-four patients met our initial inclusion criteria. Of Primary partial arthrodesis consisted of the first TMT joint
those, 25 patients with 25 Lisfranc fracture-dislocations alone in three patients, the first and second TMT joints in two
consented to participate in the study and returned for patients, and the first three TMT joints in 20 patients. The
followup at a mean of 42 (range, 24 to 96) months. The medial intercuneiform joint was also fused in six patients.
study population consisted of 8 males and 17 females, with a Fixation of the lateral TMT joints with temporary Kirschner
median age at the time of injury of 46 (range, 20 to 73) years. wires was required for the fourth TMT joint alone in two
Patients were separated into two groups based on their pattern patients, and for both the fourth and fifth TMT joints in three
of injury. Ligamentous injuries with associated fractures patients. Proximal tibia or calcaneal bone graft was used in
of the metatarsal bases, cuneiforms, or the cuboid were all patients.
considered combined osseous-ligamentous injuries, whereas
an isolated avulsion of the Lisfranc ligament from the Postoperative management
base of the second metatarsal (fleck sign) was included The postoperative protocol consisted of routine followup
with purely ligamentous injuries in a primarily ligamentous visits at 2 weeks, 6 weeks, 3 months, 6 months, 1 year, and
injury subgroup. Overall, there were no open injuries, none yearly thereafter. Immediately postoperatively the patients
of the injuries were work-related, and the mechanisms of were placed in a posterior splint for 10 days, at which

Table 1: Patient Demographic and Clinical Data by Tarsometatarsal Injury Subtype

Primarily Osseous &
Ligamentous Ligamentous p Value
Age: median(range) 51.5 (20 71) 46.0 (20 73) 0.91332
Female: n (%) 8 (66.7) 9 (69.2) 1.00003
TMT joints injured: median (range) 3 (2 5) 3 (2 5) 1.00002
Patients with comorbidities: n (%)1 3 (25) 4 (31) 0.77693

1 , Comorbidities considered relevant to arthrodesis included smoking, chronic steroid use, and the use of

bisphosphonates for osteoporosis; 2 , p value for Wilcoxon ranksum test; 3 , p value for Fishers Exact Test.

Copyright 2012 by the American Orthopaedic Foot & Ankle Society

52 REINHARDT ET AL. Foot & Ankle International/Vol. 33, No. 1/January 2012

point the splint and sutures were removed, and the patient sports or activities via a clinical questionnaire. Specifically,
was converted to a controlled ankle motion (CAM) boot. they were asked to estimate their current level of activity as
Patients were kept nonweightbearing for the first 6 weeks, a percentage of their activity level before the injury, and to
but were encouraged to work on active ankle range of describe any activities that they were prevented from doing
motion. At 6 weeks, patients began a progressive weightbear- because of the current function of their foot.
ing and physical therapy protocol, with full weight-
bearing achieved by 8 weeks postoperatively. Initial Statistical analysis
weightbearing was performed in the CAM boot, which was Descriptive statistics were calculated for variables of
weaned based on residual pain. Six weeks of gait, balance, interest. Continuous variables were represented by means and
and proprioceptive training was then followed by a strength- standard deviations or medians and ranges, while frequencies
ening protocol. Standard AP, lateral, and oblique radio- and percentages were reported for categorical and binary
graphs were obtained at each visit beginning with the 6-week outcome variables. The Wilcoxon ranksum test was used to
followup. If the fourth and/or fifth tarsometatarsal joints had analyze continuous variables and Fishers Exact Test was
been fixed with Kirschner wires, the Kirschner wires were used to evaluate binary and categorical variables. Univariate
removed in the office at the 6-week visit prior to initiating analysis of variance (ANOVA) was performed to assess the
weightbearing. association of variables of interest to the primary outcome
of AOFAS score. A p value less than 0.05 was considered
Followup to be statistically significant. Post hoc power analysis, using
Patients were contacted via mail and telephone to partici- alpha at 0.05 and beta at 0.8, for a 20% difference in the
pate in the study and requested to return for clinical followup AOFAS score (mean, 81 + 17) revealed 17 patients would
as part of the study. During the followup visit all patients be needed per group. Analyses were conducted using SAS
underwent a physical exam, had radiographs, and answered software version 9.1 (SAS Institute, Cary, NC).
a series of questionnaires related to their pain and function.
The physical exam consisted of an assessment of the appear- RESULTS
ance and alignment of the foot, an analysis of their gait, and
range of motion of adjacent joints. Standing AP, lateral, and Overall 84% of patients (21 out of 25) were either
oblique radiographs were taken of both feet in each patient somewhat satisfied or very satisfied with their outcome
for comparison to the contralateral side. Intraoperative, early following primary partial arthrodesis (Figure 1). Only one
followup (6 weeks and 3 months), and final visit radiographs patient across both groups was very dissatisfied. Within the
were assessed for alignment of the tarsometatarsal joints, primarily ligamentous group seven patients were very satis-
fracture union, status of the fusion, integrity of retained hard- fied, two were somewhat satisfied, and three were some-
ware, and adjacent joint arthrosis. Two independent observers what dissatisfied. In the combined injury group, eleven
used previously described radiographic parameters to assess patients were very satisfied, one was somewhat satis-
tarsometatarsal joint alignment.26 The quality of the reduc- fied, and one was very dissatisfied. Patient satisfaction
tion was judged by the medial column line on the AP radio- did not differ between groups (p = 0.1820). The average
graph, which is a line tangential to the medial aspect of VAS pain score overall at latest followup was 1.8 points out
the navicular and the medial cuneiform.7 In a normal foot of 10 (range, 0 to 8 points). The average pain scores did
this line should intersect the base of the first metatarsal. The not differ between groups, with an average primarily liga-
reduction was considered anatomic if this criterion was met, mentous group pain score of 1.5 points out of 10 (range,
nearly anatomic if it was within 2 mm of the base of the first
metatarsal, and non-anatomic if it was more than 2 mm. 12
At latest followup, functional outcomes were assessed
using the American Orthopaedic Foot and Ankle Society
(AOFAS) score13 for the midfoot, which is based on a 8
# of Patients

0 to 100 point scale, with a score of 100 indicating the

most optimal function. This is a widely-used, although not
validated, functional outcome score. Patients also completed 4
the SF-36 form, which is a short questionnaire with 36
items used to measure physical and mental health on a scale
from 0 to 100, with higher scores indicating better perceived 0
functioning.28 Patients rated their pain levels on a visual Very dissastisfied Somewhat
Somewhat Satisfied Very Satisfied

analog pain scale (VAS) from 0 (no pain) to 10 (worst pain). Patient Satisfaction Level
Finally, patients were asked about their satisfaction with their Primarily ligamentous Combined osseous + ligamentous

outcome (very dissatisfied, somewhat dissatisfied, somewhat

satisfied, or very satisfied) and return to previous levels of Fig. 1: Patient satisfaction level according to injury subtype.

Copyright 2012 by the American Orthopaedic Foot & Ankle Society

Foot & Ankle International/Vol. 33, No. 1/January 2012 PRIMARY ARTHRODESIS OF LISFRANC INJURIES 53

0 to 4 points) and an average osseous + ligamentous not report being limited from doing any activities at latest
group pain score of 2.1 points out of 10 (range, 0 to 8 followup.
points) (p = 0.7835). In the primarily ligamentous group, The association of a number of variables and the
six patients were taking over-the-counter pain medication final AOFAS outcome score was analyzed to identify
intermittently at last followup and six patients were not variables that could potentially predict outcome. When all
taking any pain medication. In the combined group, six 25 patients were considered together, patient age, number
patients were taking over-the-counter pain medication inter- of tarsometatarsal joints injured, length of followup, and the
mittently, one patient was taking prescriptive pain medica- presence of a comorbidity were not found to be predictive
tion, and another patient was being managed by a chronic of the AOFAS score (p > 0.05 for all variables). According
pain service. to the radiographic criteria, the quality of the reduction of
At the time of final followup, the overall average AOFAS the tarsometatarsal joints was anatomic in 12 patients, nearly
midfoot score was 81 (range, 25 to 100) points. Within the anatomic in ten, and nonanatomic in three. The quality of
primarily ligamentous group the mean AOFAS score was tarsometatarsal reduction was also not associated with the
83.3 (range, 49 to 100) points, and in the combined injury final AOFAS score (p = 0.76).
group it averaged 78.5 (range, 25 to 100) points (p > 0.05). Of the 25 patients treated with primary partial arthrodesis,
Patients lost points most frequently for responses of mild six (24%) underwent additional surgical procedures. Four
pain, limitation of recreational activities, and inability to patients underwent removal of hardware. Reasons for
maintain fashionable footwear. On the SF-36 there was no removal of hardware included patient preference (n = 2),
difference between groups on either the physical or mental a broken screw (n = 1), and painful prominent hardware
component summary scores (Table 2). (n = 1). Two patients underwent revision arthrodesis for
Patients reporting returning to an average of 85% (range, tarsometatarsal nonunion. One was an avid surfer who devel-
50% to 100%) of their pre-injury activity level at latest oped a nonunion of the first and third TMT joints (Figure 2,
followup. By group, primarily ligamentous patients returned A and B). His AOFAS score improved from 41 to 82
to 86.9% (range, 50% to 100%) and combined injury patients following revision arthrodesis and bone grafting with prox-
returned to 83% (range, 50% to 100%) of their pre-injury imal tibia autograft, and he returned to surfing at latest
activity levels (p = 0.6582) (Table 3). Three patients in followup (Figure 2C). The other occurred in a smoker with
the ligamentous group reported being unable to perform rheumatoid arthritis on chronic prednisone, who had a persis-
high-level activities because of the function of their foot, tent nonunion of the second TMT joint despite attempted
which included skiing, soccer, and modern dance. Limita- revision arthrodesis with proximal tibia autograft. At latest
tion of high-level activities was reported by four patients in followup she returned to only 50% of her pre-injury activity
the combined injury group and consisted of skiing, hiking, level, was undergoing pain management, and had a final
running, and dance. The remaining 18 patients (72%) did AOFAS score of 25. In total, there were four nonunions,

Table 2: Physical and Mental Component Summary Scores by

Tarsometatarsal Injury Subtype

Primarily Osseous +
Ligamentous Ligamentous p Value
Physical Component 48.1 9.5 51.4 11.9 0.41
Score (mean SD)
Mental Component 54.9 8.0 53.6 9.3 0.59
Score (mean SD)

Table 3: Percentage of Pre-Injury Activity Level at Latest Followup as Estimated by


Primarily Osseous +
Overall Ligamentous Ligamentous p Value
Average activity level 85% 86.9% 83.0% 0.6582

Copyright 2012 by the American Orthopaedic Foot & Ankle Society

54 REINHARDT ET AL. Foot & Ankle International/Vol. 33, No. 1/January 2012


Fig. 2: A, Weightbearing AP, lateral, and oblique radiographs demonstrating a first and third tarsometatarsal joint nonunion (arrows) in a 38-year-old avid
surfer 15 months following primary partial arthrodesis. B, Sagittal cut CT scan of this patient confirming the persistence of the first TMT joint space. C,
Weightbearing AP, lateral, and oblique radiographs of the same patient with evidence of solid fusion of the first three TMT joints 9 months following revision
arthrodesis with proximal tibia autograft (of note, the patient achieved TMT fusion by 3 months following revision surgery).

three of which were either smokers, on chronic steroids, or rather than the fixation technique was responsible for the
both. Patients who developed a nonunion were more likely to outcome. While post-traumatic arthritis of the TMT joints
have a pre-existing comorbidity (smoking, steroid use) than following primary arthrodesis is not an issue, there remains
those who did not (p = 0.05). Of the two nonunions that uncertainty regarding the incidence of adjacent joint arthritis.
did not undergo revision surgery, one developed a fibrous In our study, three patients (12%) across both groups at
union that became asymptomatic with a final AOFAS score an average 42 months showed radiographic evidence of
of 90, and the other deferred surgery because of mild symp- adjacent joint post-traumatic arthritis using primary partial
toms and returned to 85% of his pre-injury activity level. arthrodesis for both combined and primarily ligamentous
Post-traumatic arthrosis of adjacent joints was seen radio- injury subtypes. Further, the incidence of adjacent joint
graphically in three patients (12%). arthritis did not differ based on injury pattern, whereby two
of those patients were in the combined osseous-ligamentous
DISCUSSION group (15.4%) and one was in the primarily ligamentous
group (8.3%) (p > 0.05).
Lisfranc fracture-dislocations continue to be a difficult With a lack of consensus in the literature regarding optimal
injury to treat by the orthopaedic surgeon with poten- treatment methods for Lisfranc injuries, recent literature has
tial for significant morbidity for the patient. Tradition- emphasized the need to compare ORIF to primary partial
ally, these patients were treated with ORIF, while partial arthrodesis.11,16 Ly and Coetzee prospectively randomized
arthrodesis was reserved as a salvage procedure for post- patients to either primary partial arthrodesis or ORIF for
traumatic arthrosis. However, the literature over time has the treatment of primarily ligamentous Lisfranc injuries,
revealed that not all Lisfranc injuries are created equal. excluding patients with combined osseous-ligamentous inju-
Outcomes of particular interest for this injury addressed ries. In their study, the average AOFAS outcome score
in the literature have included incidence of post-traumatic was significantly better in the 21 patients who underwent
arthritis, functional outcome scores, return to activity, pain primary partial arthrodesis (86.9) than in the 20 patients who
scores, patient satisfaction, and the need for further surgical underwent ORIF (57.1) (p < 0.0001).16 We found similar
procedures. results in our patients following primary partial arthrodesis
In 2000, Kuo et al. reported on the outcomes of 48 patients with an average overall AOFAS score of 81 points. In our
following ORIF for either primarily ligamentous or combined study, patients with combined osseous-ligamentous injuries
osseous-ligamentous Lisfranc injuries. At a mean followup scored similarly on the AOFAS to those with primarily
of 52 months they found evidence of post-traumatic arthritis ligamentous injuries (78.5 versus 83.3, respectively).
in only 18% percent of the combined osseous-ligamentous Ly and Coetzee also found that patients in the primary
patients compared to 40% in the patients with a primarily arthrodesis group estimated their activity level to be a higher
ligamentous injury.15 This led the authors to suggest primary percentage of their pre-injury activity level than those in the
arthrodesis as an alternative primary treatment for primarily ORIF group [92% versus 65%, respectively (p < 0.005)].16
ligamentous injuries, theorizing that the injury pattern itself Again, our results are reflective of their primary arthrodesis
Copyright 2012 by the American Orthopaedic Foot & Ankle Society
Foot & Ankle International/Vol. 33, No. 1/January 2012 PRIMARY ARTHRODESIS OF LISFRANC INJURIES 55

group in that our patients overall returned to 85% of the return to athletic performance. Finally, although a minimum
pre-injury activity level. Furthermore, the type of Lisfranc clinical followup of 2 years was used, the mean followup of
injury pattern in our study did not impact functional recovery, 42 months can be argued to be too soon to properly evaluate
with combined injury patients returning to 83% and primarily for post-traumatic arthritis of adjacent joints.
ligamentous patients returning to 86.9% of their pre-injury
activity levels (p = 0.6582). CONCLUSION
Regarding pain scores, Ly and Coetzees patients rated
their pain on the VAS, reporting less pain in the primary Primary partial arthrodesis for both combined osseous-
arthrodesis group compared to ORIF at latest followup ligamentous and primarily ligamentous Lisfranc fracture-
[1.2 versus 4.1, respectively (p = 0.0002)].16 Similarly, the dislocations allowed patients in general to return to pre-injury
overall average VAS pain score following primary partial levels of activity, and was associated with low pain scores,
arthrodesis in our study was 1.8 out of 10, and we found high patient satisfaction rates, and low re-operation rates.
no differences in pain scores between primarily ligamentous
and combined injury groups. ACKNOWLEDGMENT
Following Ly and Coetzee, Henning et al. recently
published a randomized trial comparing primary arthrodesis The authors wish to thank Huong T. Do for assistance
to ORIF, which included patients with either combined in designing and performing the statistical analyses for this
osseous-ligamentous or primarily ligamentous Lisfranc inju- study.
ries. At a mean followup of 53 months, telephone surveys
revealed patient satisfaction rates of 90% in the ORIF group REFERENCES
and 92% in the primary partial arthrodesis group.11 Our
finding of an overall patient satisfaction rate of 84% is similar 1. Alberta, FG; Aronow, MS; Barrero, M; et al.: Ligamentous
following primary partial arthrodesis. Henning et al. also Lisfranc joint injuries: a biomechanical comparison of dorsal plate and
transarticular screw fixation. Foot Ankle Int. 26:462 473, 2005.
reported on additional surgical procedures. In their study 2. Arntz, CT; Veith, RG; Hansen, ST, Jr: Fractures and fracture-
79% of the ORIF patients and 17% of the primary partial dislocations of the tarsometatarsal joint. J. Bone Joint Surg. Am.
arthrodesis patients underwent hardware removal.11 This was 70:173 181, 1988.
a significant difference, but routine hardware removal had 3. Arntz, CT; Hansen, ST, Jr: Dislocations and fracture dislocations of
been built into their ORIF protocol. The incidence of further the tarsometatarsal joints. Orthop. Clin. North Am. 18:105 114, 1987.
4. Blair, WF: Irreducible tarsometatarsal fracture-dislocation. J. Trauma
surgical procedures seen in our study (24%) is similar, and 21:988 990, 1981.
this is further supported by that shown in Ly and Coetzees 00016
study whereby 80% of the ORIF and 19% of the primary 5. Brin, YS; Nyska, M; Kish, B: Lisfranc injury repair with the TightRope
partial arthrodesis patients underwent additional surgical device: a short-term case series. Foot Ankle Int. 31:624 627, 2010.
6. Buzzard, BM; Briggs, PJ: Surgical management of acute
Our study has several limitations. Although patients tarsometatarsal fracture dislocation in the adult. Clin. Orthop. Relat. Res.
returned for clinical followup, identification of patients for (353):125 133, 1998.
inclusion in the study was done retrospectively and is there- 00014
fore subject to selection bias. Of note is the fact that 7. Coss, HS; Manos, RE; Buoncristiani, A; Mills, WJ: Abduction stress
and AP weightbearing radiography of purely ligamentous injury in the
during the time period of the study all patients presenting
tarsometatarsal joint. Foot Ankle Int. 19:537 541, 1998.
to the senior authors with Lisfranc injuries were treated with 8. Faciszewski, T; Burks, RT; Manaster, BJ: Subtle injuries of the
primary partial arthrodesis regardless of injury severity or Lisfranc joint. J. Bone Joint Surg. Am. 72:1519 1522, 1990.
pattern. The sample size of the study is relatively small, 9. Hardcastle, PH; Reschauer, R; Kutscha-Lissberg, E; Schoffmann,
and while this is somewhat dictated by the incidence of this W: Injuries to the tarsometatarsal joint. Incidence, classification and
treatment. J. Bone Joint Surg. Br. 64:349 356, 1982.
injury in the general population, it nonetheless introduces the 10. Hardcastle, PH; Reschauer, R; Kutscha-Lissberg, E; Schoffmann,
potential for type II errors (not finding a significant differ- W: Injuries to the tarsometatarsal joint. Incidence, classification and
ence when one truly exists). Post hoc power analysis showed treatment. J. Bone Joint Surg. Br. 64:349 356, 1982.
that 17 patients per group would be needed to detect a clin- 11. Henning, JA; Jones, CB; Sietsema, DL; Bohay, DR; Anderson, JG:
ically relevant difference in our primary outcome, AOFAS Open reduction internal fixation versus primary arthrodesis for lisfranc
injuries: a prospective randomized study. Foot Ankle Int. 30:913 922,
score. We did not find an association between the quality of 2009.
the reduction and outcomes, yet authors before us have.15 12. Hunt, SA; Ropiak, C; Tejwani, NC: Lisfranc joint injuries: diagnosis
Furthermore, although radiographic followup assessments and treatment. Am J. Orthop. 35:376 385, 2006.
were performed by independent examiners, clinical assess- 13. Kitaoka, HB; Alexander, IJ; Adelaar, RS; et al.: Clinical rating
ments were done by the operating surgeon, which allowed systems for the ankle-hindfoot, midfoot, hallux, and lesser toes. Foot
Ankle Int. 15:349 353, 1994.
for potential bias. In addition, this study did not assess 14. Komenda, GA; Myerson, MS; Biddinger, KR: Results of arthrodesis
the athletic or specific sports participation of the partic- of the tarsometatarsal joints after traumatic injury. J. Bone Joint Surg.
ipants, making it difficult to draw conclusions regarding Am. 78:1665 1676, 1996.

Copyright 2012 by the American Orthopaedic Foot & Ankle Society

56 REINHARDT ET AL. Foot & Ankle International/Vol. 33, No. 1/January 2012

15. Kuo, RS; Tejwani, NC; Digiovanni, CW; et al.: Outcome after open 21. Myerson, MS; Fisher, RT; Burgess, AR; Kenzora, JE: Fracture
reduction and internal fixation of Lisfranc joint injuries. J. Bone Joint dislocations of the tarsometatarsal joints: end results correlated with
Surg. Am. 82-A:1609 1618, 2000. pathology and treatment. Foot Ankle 6:225 242, 1986.
16. Ly, TV; Coetzee, JC: Treatment of primarily ligamentous Lisfranc 22. Resch, S; Stenstrom, A: The treatment of tarsometatarsal injuries. Foot
joint injuries: primary arthrodesis compared with open reduction and Ankle 11:117 123, 1990.
internal fixation. A prospective, randomized study. J. Bone Joint Surg. 23. Rosenberg, GA; Patterson, BM: Tarsometatarsal (Lisfrancs) fracture-
Am. 88:514 520, 2006. dislocation. Am J. Orthop. Suppl:7 16, 1995.
17. Mann, RA; Prieskorn, D; Sobel, M: Mid-tarsal and tarsometatarsal 24. Sangeorzan, BJ; Veith, RG; Hansen, ST, Jr: Salvage of Lisfrancs
arthrodesis for primary degenerative osteoarthrosis or osteoarthrosis tarsometatarsal joint by arthrodesis. Foot Ankle 10:193 200, 1990.
after trauma. J. Bone Joint Surg. Am. 78:1376 1385, 1996. 25. Sangeorzan, BJ; Hansen, ST, Jr: Early and late posttraumatic foot
18. Mulier, T; Reynders, P; Dereymaeker, G; Broos, P: Severe Lisfrancs reconstruction. Clin. Orthop. Relat. Res. (243):86 91, 1989.
injuries: primary arthrodesis or ORIF? Foot Ankle Int. 23:902 905, 26. Stein, RE: Radiological aspects of the tarsometatarsal joints. Foot Ankle
2002. 3:286 289, 1983.
19. Mulier, T; Reynders, P; Sioen, W; et al.: The treatment of Lisfranc 27. Trevino, SG; Kodros, S: Controversies in tarsometatarsal injuries.
injuries. Acta Orthop. Belg. 63:82 90, 1997. Orthop. Clin. North Am. 26:229 238, 1995.
20. Myerson, M: The diagnosis and treatment of injuries to the Lisfranc 28. Ware, J; Kosinski, M; Gandek, B: SF-36 Health Survey: Manual and
joint complex. Orthop. Clin. North Am. 20:655 664, 1989. Interpretation Guide, Lincoln, RI., Quality Metric, 2000.

Copyright 2012 by the American Orthopaedic Foot & Ankle Society

View publication stats