Archives of Orthopaedic and Trauma Surgery

https://doi.org/10.1007/s00402-018-3022-x

TRAUMA SURGERY

Relationships between Hoffa fragment size and surgical approach

selection: a cadaveric study
Wich Orapiriyakul1 · Theerachai Apivatthakakul1,2 · Chanakarn Phornphutkul1,2

Received: 16 March 2018

© Springer-Verlag GmbH Germany, part of Springer Nature 2018

Abstract
Introduction  Fixation of a small Hoffa fragment requires a selection of the proper surgical approach for reduction and pos-
terior to anterior screws fixation. However, currently there are no guidelines regarding how to select the best approach for
small posterior Hoffa fractures.
Objectives  To compare the size of Hoffa fractures that are appropriate for reduction and fixation with the medial parapatellar
approach (MPPA) and those which require the direct medial approach (DMA), and to make a similar comparison between
the lateral parapatellar approach (LPPA) and the posterolateral approach (PLA).
Materials and methods  Twenty extremities of fresh cadavers were included. After completion of each approach, the articular
surface boundaries were marked and soft tissue was removed. On the medial condyle, an imaginary line was drawn from the
most anterior (A) to the most posterior (B) point, representing the AP diameter (d3). The most posterior boundary of MPPA
(C) and the most anterior boundary of DMA (D) were similarly marked. Distances between B and C (d1) and between B and
D (d2) were measured as well as the anterior–posterior diameter of the condyle (d3). The same measurements were made
for the lateral condyle.
Results  On the medial condyle, the average values of d1, d2, and d3 were 10.8 mm ± 3.8, 17.3 mm ± 3.3, and 60.1 mm ± 3.2,
while percentages of d1/d3 and d2/d3 were 18.3% ± 6.4 and 28.7% ± 4.7. In lateral condyle, the averages for d1, d2, d3 were
6.1 mm ± 1.4, 12.1 mm ± 2.8 and 60.9 mm ± 3.3 mm and the percentages of d1/d3 and d2/d3 were 10.1% ± 2.3 and 19.9% ±
4.9.
Conclusions  When the Hoffa fragment is less than 18.3% of the AP diameter of medial condyle or 10.1% of lateral condyle,
the fracture is invisible with the PPA. When the Hoffa fragment is more than 28.7% of the medial condyle or 19.9% of the
lateral condyle, the PPA should be selected. If the Hoffa fragment is less than 28.7% of the medial condyle or 19.9% of
the lateral condyle, the DMA or PLA with posterior-to-anterior screws is recommended. Combined approaches should be
considered in some complex cases with articular comminution.

Keywords  Hoffa fracture · Femoral condyle fracture · Surgical approach · Posterior-to-anterior screw fixation

Introduction

Coronal plane fractures of the femoral condyle, known as

* Theerachai Apivatthakakul
tapivath@gmail.com Hoffa fractures, are not uncommon, but selection of an
improper surgical approach can lead to poor fracture reduc-
Wich Orapiriyakul
wich.orapir@gmail.com tion and fixation resulting in complications, e.g., fracture
displacement, screws backing out, avascular necrosis, or
Chanakarn Phornphutkul
berkbann@hotmail.com post-traumatic arthritis [1, 2]. The parapatellar approach
(PPA) is an effective approach for intraarticular distal fem-
1
Department of Orthopaedics, Faculty of Medicine, Chiang oral condyle fracture reduction and fixation using anterior-
Mai University, Chiang Mai, Thailand to-posterior screws (A–P), but with that approach it is dif-
2
Excellence Center in Osteology Research and Training ficult to access and reduce small Hoffa fragments (Letenneur
Center (ORTC), Chiang Mai University, Chiang Mai, classification Type IIb and IIc) using the knee hyperflexion
Thailand

13
Vol.:(0123456789)

technique. Alternative approaches have been proposed, e.g.,
direct medial [3] and extensile subvastus approaches [4], for
isolated medial Hoffa fractures and direct lateral [5] and pos-
terolateral [6] approaches for lateral Hoffa fractures. These
approaches allow posterior-to-anterior (P–A) screw fixation
which provides better stability than A–P screws [7]. Cur-
rently, no reports have been published comparing the surgi-
cal visualization with the anterior and posterior approaches
for treating Hoffa fractures. This study compares the surgical
exposure using the medial parapatellar approach (MPPA)
and the direct medial approach (DMA) for small medial
Hoffa fragments and for lateral Hoffa fragments using both
the lateral parapatellar approach (LPPA) and the posterolat-
eral approach (PLA). The percentage of the condylar width
is used as the unit of comparison. We also investigated an
alternative posterolateral approach to expose the posterior
aspect of the lateral femoral condyle.
Materials and methods
Twenty extremities from ten fresh frozen cadavers were
included in this study. The mean age of the donors was
69.7 years. The average circumference of all extremities
measured at the distal femoral epicondyle was 32.4 cm.
(SD = 2.9; min, max = 28.5, 37.1). Ten extremities were
selected to compare MPPA and DMA for the medial condyle
approach; another ten extremities were selected to compare
LPPA and PLA for the lateral condyle approach. The articu-
lar boundaries for each approach were marked with metal
pins, after which soft tissue was removed and the femur was
transected at the distal one-third level. The study protocol
was approved by the Institutional Ethical Committee Board,
Faculty of Medicine, Chiang Mai University.
Surgical techniques
Medial parapatellar approach (MPPA)
The torso was laid in the supine position with slight knee
flexion using a bolster. A 13-cm incision was started from
tibial tubercle, curved medially, and extended to a point
2.5  cm above the superior pole of the patella. A 13-cm
medial arthrotomy was made from the proximal through the
medial retinaculum and vastus medialis distally toward the
tibial tubercle. The patella was pushed laterally and 120°
of knee flexion was induced to expose the medial femoral
condyle (Fig. 1a).
Direct medial approach (DMA)
This approach was performed in the supine position with slight
knee flexion and external rotation of the hip. A 10-cm incision
13
Archives of Orthopaedic and Trauma Surgery
was started from the adductor tubercle 8 cm proximal to the
joint level, extending distally along the proximal tibia to a
point 2 cm distal to the joint level. The medial collateral liga-
ment was identified and protected. A longitudinal capsulotomy
was then performed along the posterior border of the medial
collateral ligament to expose the posterior part of medial femo-
ral condyle (Fig. 1b–d).
Lateral parapatellar approach (LPPA)
The torso was laid in the supine position with slight knee
flexion using a bolster. A 13-cm incision was begun at the
tibial tubercle, curving along the lateral side of the patella, and
reaching a point 2.5 cm above the superior pole of the patella.
A lateral arthrotomy was performed via the lateral retinaculum
and was proximally extended through the vastus lateralis. The
knee was flexed 120° and the patella was dislocated medially
(Fig. 2a).
Direct posterolateral approach (PLA)
This approach was performed in the prone position. A 10-cm
incision was started 6  cm above the popliteal crease and
extended to 4 cm below the crease medial to the fibula head.
The common peroneal nerve (CPN), which lies posterior to
biceps femoris, was identified and retracted together later-
ally. The lateral head of the gastrocnemius muscle was also
identified and retracted laterally. A vertical capsulotomy was
preformed to provide access to the posterior femoral condyle
(Fig. 2b–d).
Reference points and measurement methods
As shown in Fig. 3, the distance from the most anterior point
(A) to the most posterior point (B) of the condyle, i.e., the
anterior–posterior diameter of the condyle (d3), was used as
the reference plane. Metal pins were used to mark the most
posterior boundaries with the anterior approaches (MPPA and
LPPA) (C) and the most anterior boundaries with the pos-
terior approaches (DMA and PLA) (D). The perpendicular
distance from the reference plane (B) to selected metal pins
(C and D) are labeled d1 and d2, respectively. d1 represents
the blind area with the PPA and d2 represents the visible area
with the posterior approach. A digital Vernier caliper was used
to measure distances. Data were analyzed using STATA 12.1
software (College Station, Texas, USA) and Microsoft Excel.
Means, standard deviations, and 95% confidential intervals
were calculated.
Archives of Orthopaedic and Trauma Surgery

Fig. 1  Medial approach. a Medial parapatellar approach (MPPA). b–d Direct medial approach (DMA). b Skin incision. c Dissect and identify
the posterior border of the medial collateral ligament (MCL). d Capsulotomy posterior to MCL to expose the posterior medial condyle

Fig. 2  Lateral approach. a
Lateral parapatellar approach
(LPPA). b–d Direct postero-
lateral approach (PLA). b Skin
incision at the lateral border
of the lateral gastrocnemius
muscle. c Common peroneal
nerve (CPN) was identified. The
lateral head of the gastrocne-
mius muscle, CPN, and the
biceps muscle were retracted
laterally. d Capsulotomy of
the posterolateral capsule to
expose the lateral condyle (LC)
posteriorly

13
 Archives of Orthopaedic and Trauma Surgery

Table 1  Distance measurements made on the distal femoral condyle

to selected points as shown in Fig. 3

Distance Mean (mm) Standard (Min, max) 95% confi-

deviation (mm) dential inter-
(mm) val (mm)

d1 10.8 3.8 (4.9, 17.8) 8.0–13.5

d2 17.3 3.3 (12.3, 22.5) 14.9–19.6
d3 60.1 3.2 (56.1, 65.0) 57.8–62.4

Discussion

Hoffa fractures involving the articular surface of the knee

Fig. 3  Distance measurements made on the distal femoral condyle.
A: Most anterior point of the condyle. B: Most posterior point of the
condyle. C: Most posterior boundary of MPPA or LPPA. D: Most
anterior boundary of DMA or PLA. d1: Perpendicular distance from
C to reference plane. d2: Perpendicular distance from D to reference
plane. d3: A–P condylar width (distance between A and B)
Results
On the medial condyle, the average distance d 1 was
10.8 ± 3.8 mm (95% CI 8.0–13.5), d2 was 17.3 ± 3.3 mm
(95% CI 14.9–19.6), and d3 was 60.1 ± 3.2 mm (95% CI
57.8–62.4). The averages of d1/d3 and d2/d3 expressed as
a percentage were 18.3% ± 6.4 (95% CI 13.7–22.9%) and
28.7% ± 4.9 (95% CI 25.2–32.2%), respectively (Fig. 4;
Tables 1, 2). On the lateral condyle, the average distance d1
was 6.1 ± 1.4 mm (95% CI 5.1–7.2), d2 was 12.1 ± 2.8 mm
(95% CI 10.1–14.0), and d3 was 60.9 ± 3.3 mm (95% CI
58.6–63.3). The averages of d1/d3 and d2/d3 expressed as a
percentage were 10.1% ± 2.3 (95% CI 8.4–11.7) and 19.9%
±4.9 (95%CI 16.4–23.4%), respectively (Fig. 5; Tables 3, 4).
require anatomical reduction. The prevalence of Hoffa
fractures, in particular isolated small Hoffa fractures, was
recently highlighted, emphasizing the importance of optimal
preoperative planning in the selection of surgical approach
and fixation construct. Hoffa fractures occur more frequently
in the lateral femoral condyle. In the axial plane, fractures
commonly extend from anterolateral to posteromedial in the
lateral condyle and from anteromedial to posterolateral in
the medial femoral condyle. In the sagittal plane, fractures
are transverse, from anteroinferior to posterosuperior. Artic-
ular comminution is more commonly seen in lateral condylar
fractures and is concentrated in the weight-bearing zone of
the articular surface [8].
Selection of an approach which is appropriate for the
Hoffa fragment size is crucial to the success of fragment
reduction and fixation. The parapatellar approach is com-
monly used for large Hoffa fragment reductions and A–P
screw fixation, but treatment of small posterior Hoffa frag-
ments is difficult due to poor visualization and fracture
reduction problems. There is no consensus regarding surgi-
cal exposure for treatment of small Hoffa fractures (Leten-
neur classification type II). With some small Hoffa fractures,

Fig. 4  Boxplot showing the width of the DMA (green highlight) and the MPPA (yellow highlight) as a percentage of the width of the medial
femoral condyle (green box and red box, respectively)

13
Archives of Orthopaedic and Trauma Surgery

Table 2  Percentage of MPPA (d1) and DMA (d2) to A–P medial con- Table 3  Comparison of the average distances from reference plane on
dylar width A–P lateral condylar width

Ratio Mean (%) Standard (Min, max) (%) 95% confiden- Distance Mean (mm) Standard (Min, max) 95% confi-
deviation tial interval deviation (mm) dential inter-
(%) (%) (mm) val (mm)

d1/d3 18.3 6.4 (8.8, 29.7) 13.7–22.9 d1 6.1 1.4 (4.5, 8.3) 5.1–7.2
d2/d3 28.7 4.9 (21.9, 37.5) 25.2–32.2 d2 12.1 2.8 (7.4, 15.3) 10.1–14.0
d3 60.9 3.3 (57.0, 68.0) 58.6–63.3

access by the parapatellar approach is not possible, making

reduction and fixation either very difficult or impossible. Table 4  Percentage of LPPA ­(d1) and PLA ­(d2) to lateral A–P condy-
Alternatives, e.g., the posterior, medial, and posterolateral lar width
approaches, allow P–A screw or plate fixation which pro- Ratio Mean (%) Standard (Min, max) (%) 95% confiden-
vide mechanical advantages [7, 9], but these approaches do deviation tial interval
not address the fracture line of a large anterior fragment. (%) (%)
Jarit et al. [7] reported that fixation with posterior to ante- d1/d3 10.1 2.3 (7.3, 14.0) 8.4–11.7
riorly placed cancellous lag screws was significantly more d2/d3 19.9 4.9 (13.0, 25.9) 16.4–23.4
stable than anterior to posteriorly placed screws with 0.67
and 1.36 mm displacement, respectively, after 10 loading
cycles (P = 0.05). This technique also requires that the screw Hoffa fractures whether with or without comminution. A
heads be recessed beneath the articular surface. Available few studies have suggested that the direct medial approach
approaches and treatment outcomes of medial and lateral allows not only small medial Hoffa fragment reduction but
Hoffa fractures are shown in Tables 5, 6 and 7. also P–A screw or plate fixation, providing more stable fixa-
Isolated medial Hoffa fractures are less common and the tion than A–P screws [3, 4]. Karim et al. [3] reported using
fractures are more simple than lateral Hoffa fractures [8]. the medial approach to fix medial Hoffa fractures; however,
Viskontas et al. [4] proposed using the extensile medial sub- they did not mention the plane of dissection used in their
vastus approach for isolated medial Hoffa fractures. This approach. Gao et al. [21] reported on 13 cases which used
approach provides greater visualization of both the anterior the modified posteromedial approach between the sartorius
and posterior parts of the medial condyle through two sur- and semimembranosus/semitendinosus muscles to fix medial
gical windows, but extensive skin incision is required. A Hoffa fractures, of which 6 were small fragment fractures
study by Xie et al. [8] reported that the prevalence of frac- (Letenneur Type II). That approach is appropriate for frac-
tures line of medial Hoffa fracture were less concentration tures that extend to the metaphysis which also need a medial
than lateral Hoffa fractures, with more than half running buttress plate; however, direct insertion of P–A screws into
across the middle-third. Fortunately, the zone of comminu- the intra-articular fragment is difficult. In the present study,
tion was usually identified in this area. For these reasons, we used the direct medial approach which can be used to
MPPA is still the standard approach to treating most medial access posterior Hoffa fractures with an average articular

Fig. 5  Boxplot showed the percentage of distance of the PLA (green highlight; green box) and the LPPA (yellow highlight; red box) correlating
to lateral femoral condyle

13
 Archives of Orthopaedic and Trauma Surgery

Table 5  Surgical approaches and treatment outcomes of lateral Hoffa fracture treatment

Author Approach Letenneur Type of study No. cases Incision Intermuscular Screw trajec- Outcome
classification plane tory, fixation

Lewis et al. MPPA I = 2, IIb = 1, Case reports 7 Anterior NM A–P + condy- ROM 0–130°
[10] IIc = 1, midline lar fragment Fair–good
III = 3 to shaft
screw
Liebergall LPPA with – Technical – Straight LPPA Reflect IT A–P –
et al. [11] Gerdy’s tricks band and
tubercle Gerdy’s
osteotomy tubercle
Kumar et al. Lateral NM Case series 3 NM subvastus A–P 2 patients had
[12] approach excellent
results
Papadopoulos Lateral I Case report 1 NM IT tract and A–P Impaired knee
et al. [13] approach biceps flexion
femoris (80° of flexion)
Heterotopic
ossification
Mounasamy Swashbuckler 33C3 with Technical note 1 Anterior LPPA A–P + lateral Loss follow-up
et al. [14] approach PCL injury Case report midline locking plate
Shi et al. [5] Direct lateral I = 3, IIa = 1 Case series 12 Curved lateral IT band and P–A = 9 (all Excellent
approach IIb = 1, IIc = 3 incision biceps type II) (KSS = 80–92)
III = 4 femoris A–P = 3
Tan et al. [6] Posterolateral IIc Case report 1 Vertical pos- Medial to Surgical NM
approach and letter to terolateral biceps suture
editor incision femoris
Egol et al. Lateral Comminuted Supplement 1 Straight inci- IT band and P–A (HS) Union and return
[15] approach LC article sion biceps to full activities
femoris with at 9 months
IT band
elevation
Lian et al. Direct lateral III = 12 Case series 12 Curved lateral IT band and Meta Union at 11–32
[16] approach incision biceps plate + can- weeks
femoris nulated
screws

CS cortical screw, HS headless screw, KSS Knee Society score, LC lateral condyle, LPPA lateral parapatellar approach, MPPA medial parapatel-
lar approach, MC medial condyle, NM not mentioned, PTCS partial thread cancellous screw

fragment of 28.7% ± 4.9 of the AP diameter of the medial
condyle. Reduction using a joystick attached to this fragment
can be easily achieved and fixation with headless screws or
countersink screws provides the absolute stability necessary
for cartilage healing. This is a simple approach which does
not damage any medial knee structures, with the exception
of the capsulotomy. The limitation of this approach is that it
does not expose the metaphyseal fracture or allow fixation of
a buttress plate from the posterior. If the fragment is larger
than 28.7% ± 4.9 of the AP diameter of medial condylar
width, MPPA should be considered.
A lateral Hoffa fracture is more common and more com-
plex with comminution than a medial Hoffa fracture. LPPA
is the standard approach for displaced intra-articular frac-
tures of the distal femur, including lateral Hoffa fractures.
Small lateral Hoffa fractures are not common, but they
are challenging in terms of selection of surgical approach,
fracture reduction, and internal fixation. Shi et al. [5] used
a direct lateral approach (DLA) through the intermuscular
plane between the biceps femoris and the iliotibial band to
treat 12 lateral Hoffa fractures, 5 of which were Letenneur
Type II. Recently, Lian et al. [16] treated 12 patients with
Letenneur Type III fracture via posterolateral approach using
meta plate and cannulated screw. All the fracture healed
without complications. With that approach, the fracture line
can be visualized from the lateral side as some of the poste-
rior condyle, but not directly posterior. In addition, the P–A
screw direction has to be slightly oblique. A Letenneur Type
IIc, which involves a small intraarticular fragment, may not
be appropriate for treatment with this approach [6]. Tan et al.
[6] proposed treating Letenneur Type IIc fractures using a
plane of approach between the biceps femoris and the CPN,
reporting satisfactory results. In our study, we used the pos-
terolateral approach to directly access the posterior aspect of

13
Archives of Orthopaedic and Trauma Surgery

Table 6  Surgical approaches and treatment outcomes of medial Hoffa fracture treatment

Authors Approach Letenneur Type of study No. cases Incision Intermuscular Screw trajec- Outcome
classification plane tory, fixation

Karim et al. Medial NM Case report 1 NM NM P–M, A–L lag NM

[3] approach 1–4 screws
Miyamoto NM IIB Case report 1 NM NM A–P Full ROM, no
et al. [17] articular step
off
Ocguder et al. MPPA I with tibial Case report 1 NM MPPA A–P (HS) ROM 35 at 5
[18] eminence months
fx and PCL
injury
Yücel et al. Posteromedial I Case report 1 PM NM P–A ROM 0–145°
[19] approach
Chang et al. NM I = 2 Case report 2 NM NM A–P (HS), CS, Poor reduc-
[20] DCP plate tion via
percutaneous
technique
Gao et al. [21] Modified pos- I = 3, IIa = 1, Retrospective 13 Straight Between Sar- MP–AL and Good
teromedial IIb = 3, cohort incision torius and locking plate (KSS > 80)
approach IIc = 2, over vastus semimem-
III = 4 medialis, branosus/
Sartorius semitendi-
and MC nosus
Kapoor et al. Posterior IIc Case report 1 Lazy S-shape NM P–A (HS) ROM 0–125°
[22] approach Fracture union

CS cortical screw, HS headless screw, KSS Knee society score, LC lateral condyle, LPPA lateral parapatellar approach, MPPA medial parapatel-
lar approach, MC medial condyle, NM not mentioned, PTCS partial thread cancellous screw

the lateral condyle. After identifying the lateral head of the
gastrocnemius, the muscle was retracted laterally to expose
the posterior aspect of the condyle. The plane of dissection
was loose, wide and provided direct exposure of the poste-
rior capsule on the lateral femoral condyle. The retractor was
applied gently to protect the popliteal vessels and the tibial
nerve on the medial side. The posterolateral capsule was
incised longitudinally to expose the posterior aspect of the
lateral femoral condyle. This allows direct visualization of
Letenneur Type II fractures and permits perpendicular screw
fixation of the fracture in the posterior-to-anterior direction.
LPPA allows visualization of almost 90% of the anterior-to-
posterior lateral condylar width with the exception of 10.1%
± 2.3 of the area. However, fixation of a very small fragment
in the anterior-to-posterior direction is very difficult and is
usually less stable. When the lateral Hoffa fragment is less
than 19.9% of the d3 A–P diameter of the lateral condylar
width, or in the other hand the fracture is visible and be
able to fix from posterior, we recommend using this direct
posterolateral approach.
In complex cases, e.g., comminuted Hoffa fractures,
fractures involving the borderline zone (between the PPA
and the posterior approaches) [8], an oblique fracture plane
in three dimensions, and patients with thick muscle mass,
the single approach may be inadequate for fracture reduc-
tion and fixation. In those cases, the use of a combination
of PPA together with alternative approaches such as DMA
for medial condyle or PLA for lateral condyle is recom-
mended. For the lateral decubitus in the floating position,
the combination of LPPA and PLA is optional, while in the
supine position with the contralateral leg lowered is suit-
able for MPPA and DMA in medial Hoffa fractures. These
techniques allow the surgeon to use both approaches within
a single draping. Additionally, knee position can be manipu-
lated to facilitate fracture visualization. In complex medial
Hoffa fractures requiring combined approaches, we recom-
mended an anterior midline incision for MPPA to avoid dis-
turbing the cutaneous blood supply by maintaining a small
skin bridge between MPPA and DMA.
This study is intended to help surgeons in clinical prac-
tice to select the appropriate approaches by evaluating three-
dimensional CT scans. We described a case that elucidates
the importance of surgical approach selection. The patient
had sustained a complex medial Hoffa fracture. The Hoffa
fragment–condylar width ratio was approximately 20–30%.
He had been initially treated with P–A partial thread screw
fixation via MPPA at another hospital, but the fracture was
not anatomically reduced. That failure of fragment reduc-
tion could have been due to the fracture being located at the
borderline zone and further complicated by articular commi-
nution and bone loss. The fixation was revised and re-fixed
with three P–A headless screws through a combination of

13


Table 7  Surgical approaches and treatment outcomes of medial and lateral Hoffa fracture treatment
Authors Approach Letenneur classifi- Type of study No. cases Incision Intermuscular plane Screw trajectory, Outcome

13
cation fixation

Holmes et al. [23] LPPA, MPPA NM Case report 5 Midline incision PPA A–P, 3.5 mm ROM 0–115°, mean
LC = 2 KSS = 173
MC = 3
Viskontas et al. [4] Two windows NM Technical note + LC = 8 Extensile anterior LPPA A–P and P–A ROM ~ 0–120°
(MPPA windows Case series MC = 5 skin incision Subvastus plane
and subvastus 25 cm
windows)
Gavaskar et al. [24] MPPA I = 8, II = 4 Case report LC = 11 Midline incision MPPA A–P(PTCS) Mean KSS 78.8 ± 5.9
III = 6 MC = 7 P–A (HS)
Dhillon et al. [25] LPPA, LPPA + sub- NM Case report 6 1 Midline PPA A–P (PTCS) ROM 0-120°
vastus, MPPA Subvastus approach Mean healing time
4.6 months
Beltran et al. [26] Swashbuckler (S) NM Cadaveric study 20 fresh (S)- Midline inci- LPPA – –
Mini-swashbuckler frozen sion 30 cm
(MS) hemipelvis (MS)-12 cm from
TT –LPP plane to
vastus lateralis
Xu et al. [27] LPPA, MPPA I = 7, II = 1 Case report LC = 7 Anterolateral or PPA Intercondylar Mean healing time
III = 3 MC = 4 Anteromedial inci- screw + cross A–P 11.6 weeks
sion Mean KSS 174.6
Sahu et al. [28] Small/ lateral inci- NM Prospective, case 22 Small medial / NM NM 90.9% union
sion series lateral incision
Bel et al [29] 78% AL or AM AO 33B3 = 18 Pro- and retrospec- 18 NM NM 85% A–P 12% articular step off
22% PL or PM tive review 15% P–A
Xu et al. [30] LPPA, MPPA I = 17, III = 10 Case study 27 Anterolateral or PPA Intercondylar New technique is
anteromedial screw + cross A–P effective as A–P
incision vs A–P
Onay et al. [2] MPPA, direct lat- I = 7, IIa = 1, IIb = 1, Retrospective 13 MPPA NM 77% A–P Mean healing time 10
eral, direct medial IIc = 1 III = 3 review 23% P–A weeks
Mean KKS = 78.4
54% OA, 15.4% AVN
Trikha et al. [31] Direct lateral or NM Retrospective LC = 21 NM IT band and bicep- A–P (stab incision) Mean healing time
direct medial review MC = 11 femoris 11.56 weeks, mean
approach KSS = 83.19
Singh et al. [32] Swashbuckler for NM Retrospective LC = 5 Midline incision LPPA, MPPA A–P ROM 0–110°
LC review MC = 3 and MPPA 87.5% excellent
MPPA for MC results

AL anterolateral, AM anteromedial, CS cortical screw, HS headless screw, KSS Knee society score, LC lateral condyle, LPPA lateral parapatellar approach, MPPA medial parapatellar approach,
MC medial condyle, NM not mentioned, PTCS partial thread cancellous screw
Archives of Orthopaedic and Trauma Surgery
Archives of Orthopaedic and Trauma Surgery

MPPA–DMA in a supine–contralateral leg lowered posi- Our study has some limitations. First, the study was
tion for articular reduction achieving a favorable outcome done on intact cadaveric bone, while real fractures are com-
(Fig. 6). plex in three dimensions and can occur with or without

Fig. 6  Case demonstration for

combined approaches. a 3D-CT
reconstruction of initial P–A
screw fixation via MPPA. b
Comminution of articular sur-
face of medial condyle reduced
via MPPA. c DMA window
for fixation. d Intra operative
fluoroscopic images showed
successful anatomical reduc-
tion and headless screw fixation
from posterior to anterior. e
3 months post operation X-rays
showed complete healing

13

comminution. For that reason, other factors potentially
affecting fixation outcomes such as fracture plane and screw
trajectory require further evaluation. Second, it is difficult
to postulate the working area solely on the basis of the sur-
gical approach. However, by marking surgical boundaries
with metal pins, it was possible to establish boundaries that
establish the functional area. Lastly, most of the cadavers in
this study were of an advanced age, therefore, soft tissue ten-
sion was less than it would be in young, muscular patients.
Conclusions
When the Hoffa fragment is less than 18.3% of the AP diam-
eter of the medial condyle or 10.1% of the lateral condyle,
the fracture is not visible with the parapatellar approach.
When the Hoffa fragment is more than 28.7% of the medial
condyle diameter or 19.9% of the lateral condyle diameter,
the parapatellar approach should be selected. If the Hoffa
fragment is less than 28.7% of the medial condyle diameter
or 19.8% of the lateral condyle diameter, the direct medial
or posterolateral approach with posterior-to-anterior screws
is recommended. Combined approaches should be consid-
ered in complex cases which include comminution. These
ratios can help guide surgeons in determining the appropri-
ate approach to achieve good reduction and fixation.
Funding  The authors receive financial support from the Endowment
Fund, Faculty of Medicine, Chiang Mai University and Excellence
Center in Osteology Research and Training Center (ORTC), Chiang
Mai University, Thailand for preparation of this manuscript. They did
not receive payments or other benefits or commitments or agreement
to provide such benefits from commercial entity.
Compliance with ethical standards  16. Lian X, Zeng Y-J (2018) Meta plate and cannulated screw fixation
Conflict of interest  The authors declare that they have no conflict of
interest.
Ethical approval  This article does not contain any studies with human
participants or animals performed by any of the authors. It was
approved by Ethical committee of Faculty of Medicine, Chiang Mai
University.
References
1. White EA, Matcuk GR, Schein A et al (2015) Erratum to: Coronal
plane fracture of the femoral condyles: anatomy, injury patterns,
and approach to management of the Hoffa fragment. Skelet Radiol
44:45. https​://doi.org/10.1007/s0025​6-014-2038-8
2. Onay T, Gülabi D, Çolak I et al (2017) Surgically treated Hoffa
fractures with poor long-term functional results. Injury. https​://
doi.org/10.1016/j.injur​y.2017.11.026
3. Karim A, Rossiter N (2006) Isolated medial uni-condylar hoffa
fracture following traumatic knee dislocation. Inj Extra 37:12–14.
https​://doi.org/10.1016/j.injur​y.2005.06.045
13
Archives of Orthopaedic and Trauma Surgery
4. Viskontas DG, Nork SE, Barei DP, Dunbar R (2010) Technique of
reduction and fixation of unicondylar medial Hoffa fracture. Am
J Orthop (Belle Mead NJ) 39:424–428
5. Shi J, Tao J, Zhou Z, Gao M (2014) Surgical treatment of lateral
Hoffa fracture with a locking plate through the lateral approach.
Eur J Orthop Surg Traumatol 24:587–592. https:​ //doi.org/10.1007/
s0059​0-013-1224-z
6. Tan Y, Li H, Zheng Q et  al (2014) A modified posterolat-
eral approach for Hoffa fracture. Eur J Orthop Surg Traumatol
24:1321–1323. https​://doi.org/10.1007/s0059​0-013-1310-2
7. Jarit GJ, Kummer FJ, Gibber MJ, Egol KA (2006) A mechanical
evaluation of two fixation methods using cancellous screws for
coronal fractures of the lateral condyle of the distal femur (OTA
type 33B). J Orthop Trauma 20:273–276
8. Xie X, Zhan Y, Dong M et al (2017) Two and three-dimensional
CT Mapping of Hoffa fractures. J Bone Jt Surg Am 99:1866–
1874. https​://doi.org/10.2106/JBJS.17.00473​
9. Sun H, He QF, Huang YG et al (2017) Plate fixation for Leten-
neur type I Hoffa fracture: a biomechanical study. Injury 48:1492–
1498. https​://doi.org/10.1016/j.injur​y.2017.03.044
10. Lewis SL, Pozo JL, Muirhead-Allwood WF (1989) Coronal frac-
tures of the lateral femoral condyle. J Bone Jt Surg Br 71:118–120
11. Liebergall M, Wilber JH, Mosheiff R, Segal D (2000) Gerdy’s
tubercle osteotomy for the treatment of coronal fractures of the
lateral femoral condyle. J Orthop Trauma 14:214–215. https:​ //doi.
org/10.1097/00005​131-20000​3000-00013​
12. Kumar R, Malhotra R (2001) The Hoffa fracture: three case
reports. J Orthop Surg (Hong Kong) 9:47–51
13. Papadopoulos AX, Panagopoulos A, Karageorgos A, Tylli-
anakis M (2004) Operative treatment of unilateral bicondy-
lar Hoffa fractures. J Orthop Trauma 18:119–122. https​://doi.
org/10.1097/00005​131-20040​2000-00012​
14. Mounasamy V, Desai P, Mallu S et al (2012) A novel method
of removal of a broken drill bit in the femoral medullary canal
during internal fixation of a type C distal femoral fracture: a case
report. Chin J Traumatol 15:315–316. https:​ //doi.org/10.3760/cma
.j.issn.1008-1275.2012.05.014
15. Egol KA, Broder K, Fisher N, Konda SR (2017) Repair of dis-
placed partial articular fracture of the distal femur. J Orthop
Trauma 31:S10–S11. https​://doi.org/10.1097/BOT.00000​00000​
00089​6
for treatment of type Letenneur III lateral Hoffa fracture through
posterolateral approach. Zhongguo Gu Shang 31:267–271
17. Miyamoto R, Fornari E, Tejwani NC (2006) Hoffa fragment asso-
ciated with a femoral shaft fracture. A case report. J Bone Jt Surg
Am 88:2270–2274. https​://doi.org/10.2106/JBJS.E.01003​
18. Ocguder A, Bozkurt M, Kalkan T et al (2008) Hoffa fracture,
eminentia fracture and posterior cruciate ligament damage: an
unusual knee injury. Inj Extra 39:88–91. https:​ //doi.org/10.1016/j.
injur​y.2007.08.027
19. Yücel İ, Degirmenci E, Özturan K (2008) Hoffa fracture: a case
report. Düzce Tıp Fakültesi Derg 2:37–40
20. Chang JJ, Fan JC, Lam HY et al (2010) Treatment of an osteo-
porotic Hoffa fracture. Knee Surg Sport Traumatol Arthrosc
18:784–786. https​://doi.org/10.1007/s0016​7-009-0960-4
21. Gao M, Tao J, Zhou Z et al (2015) Surgical treatment and reha-
bilitation of medial Hoffa fracture fixed by locking plate and addi-
tional screws: a retrospective cohort study. Int J Surg 19:95–102.
https​://doi.org/10.1016/j.ijsu.2015.05.027
22. Kapoor C, Merh A, Shah M, Golwala P (2016) A case of dis-
tal femur medial Condyle Hoffa Type II(C) fracture treated with
headless screws. Cureus. https​://doi.org/10.7759/cureu​s.802
23. Holmes SM, Bomback D, Baumgaertner MR (2004) Coronal frac-
tures of the femoral condyle: a brief report of five cases. J Orthop
Trauma 18:316–319
Archives of Orthopaedic and Trauma Surgery
24. Gavaskar AS, Tummala NC, Krishnamurthy M (2011) Opera-
tive management of Hoffa fractures–a prospective review of 18
patients. Injury 42:1495–1498. https​://doi.org/10.1016/j.injur​
y.2011.09.005
25. Dhillon MS, Mootha AK, Bali K et al (2012) Coronal fractures
of the medial femoral condyle: a series of 6 cases and review of
literature. Musculoskelet Surg 96:49–54. https​://doi.org/10.1007/
s1230​6-011-0165-0
26. Beltran MJ, Blair JA, Huh J et al (2013) Articular exposure with
the swashbuckler versus a “mini- swashbuckler” approach. Injury
44:189–193. https​://doi.org/10.1016/j.injur​y.2012.10.021
27. Xu Y, Li H, Yang HH (2013) A new fixation method for Hoffa
fracture. Eur J Trauma Emerg Surg 39:87–91. https​ : //doi.
org/10.1007/s0006​8-012-0238-2
28. Sahu RL, Gupta P (2014) Operative management of Hoffa fracture
of the femoral condyle. Acta Med Iran 52:443–447
29. Bel J-C, Court C, Cogan A et al (2014) Unicondylar fractures of
the distal femur. Orthop Traumatol Surg Res 100:873–877. https​
://doi.org/10.1016/j.otsr.2014.10.005
30. Xu Y, Li H, Yang HH, Pan ZJ(2016) A comparison of the clinical
effect of two fixation methods on Hoffa fractures. Springerplus.
https​://doi.org/10.1186/s4006​4-016-2861-6
31. Trikha V, Das S, Gaba S, Agrawal P (2017) Analysis of functional
outcome of Hoffa fractures: a retrospective review of 32 patients.
J Orthop Surg (Hong Kong) 25:2309499017718928. https​://doi.
org/10.1177/23094​99017​71892​8
32. Singh R, Singh RB, Mahendra M (2017) Functional outcome of
isolated Hoffa fractures treated with cannulated cancellous screw.
Malays Orthop J 11:20–24. https:​ //doi.org/10.5704/MOJ.1707.016
13