ORIGINAL ARTICLE
Medial Knee Approach: An Anatomical Study of
Minimally Invasive Plate Osteosynthesis in Medial
Femoral Condylar Fracture
Norachart Sirisreetreerux, MD,*† Babar Shafiq, MD,* Greg M. Osgood, MD,*
and Erik A. Hasenboehler, MD*
Objectives: To determine the location of distal medial neurovascular
structures, identifying a medial “safe zone” for minimally invasive
plate osteosynthesis to treat displaced femoral condylar fractures.
Methods: Eleven uninjured lower-half torsos were dissected on the
bilateral medial lower thigh. A longitudinal incision was made at the
midsagittal plane of the medial thigh starting 1 cm proximal to the
knee joint and extending to the proximal one-third of the femur.
Superficial and deep neurovascular structures were dissected. Dis-
tances to the medial vastus and adductor compartment were measured.
Results: Mean distances were 160 6 31.4 mm from the adductor
tubercle to Hunter canal; 94 6 18.3 mm from adductor tubercle to
adductor hiatus; 31.8 6 9.21 mm from Hunter canal to the femoral
shaft; and 31.7 6 7.78 mm from adductor hiatus to femoral shaft. All
specimens had a descending genicular artery (DGA) with a mean
distance to the adductor tubercle of 98.4 6 16.0 mm. The muscular
branch of the DGA crossed the femoral shaft at approximately
50 mm from the adductor tubercle; the osteoarticular branch ran
along the adductor magnus tendon. The nerve to the vastus medialis
was at the posterior border of the vastus medialis, entering at a mean
143 6 63.0 mm from the adductor tubercle.
Conclusions: Minor neurovascular branches of the DGA may be
vulnerable during medial femoral condyle plating. Careful blunt
dissection, proper instrumentation, and plate length within 160 mm
allow distal medial femur fixation without additional proximal dissection.
Key Words: adductor tubercle, descending genicular artery, Hunter
canal, medial femoral condylar fracture
(J Orthop Trauma 2016;30:e357–e361)
Accepted for publication July 15, 2016.
From the *Department of Orthopaedic Surgery, The Johns Hopkins University,
Baltimore, MD; and †Department of Orthopaedics, Faculty of Medicine,
Ramathibodi Hospital, Mahidol University, Bangkok, Thailand.
The authors report no conflict of interest.
Supplemental digital content is available for this article. Direct URL
citations appear in the printed text and are provided in the HTML
and PDF versions of this article on the journal’s Web site (www.
jorthotrauma.com).
Reprints: Erik A. Hasenboehler, MD, Department of Orthopaedic Surgery,
Johns Hopkins Bayview Medical Center, The Johns Hopkins University,
4940 Eastern Avenue, #A667, Baltimore, MD 21224-2780 (e-mail:
editorialservices@jhmi.edu).
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
DOI: 10.1097/BOT.0000000000000659
J Orthop Trauma  Volume 30, Number 11, November 2016
Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
INTRODUCTION
Isolated femoral condylar fractures (AO type 33-B1 to
B3) are less common than interbicondylar fractures (AO
types 33-C1 to C3). Displaced type-B medial femoral
condylar fractures usually require interfragmentary lag
screw fixation, buttress plating, or both. In type-C fractures
with severe medial metaphyseal comminution, combined
medial and lateral exposure may provide maximum stability
with dual plating. However, extensive soft-tissue dissection
is required to achieve proper exposure for medial plate
placement.1
The anatomy of the medial lower thigh and knee is
complex, and several structures are vulnerable during surgical
exposure: the superficial femoral artery at Hunter canal, the
deep femoral artery, and the genicular artery with its 7 variable
branches.2,3 These vessels may be cut or develop a pseudoa-
neurysm, as has been described after total knee arthroplasty.4,5
The femoral nerve is also at risk with 2 patterns of nerve
branches, as described by Jojima et al.6 Both branches supply
the vastus medialis longus and obliquus, and injury to these
nerves can cause vastus muscle atrophy. The typical locations
and distances of these structures relative to the surrounding
anatomic landmarks have not been described in detail.
Exposure of the distal medial femur is challenging, and
the anatomy has not been clearly described. Surgeons are
most familiar with the medial parapatellar approach, which is
the most common approach to the medial distal femur.6–11
This involves a midline arthrotomy, creating a large anterior
scar and capsular scarring. Less invasive approaches include
the subvastus approach10,11 and the approach of Visser et al,12
used for a medial closing wedge osteotomy of the distal
femur. However, the neurovascular structures of the vastus
medialis, including the musculo-articular branch of the de-
scending genicular artery (DGA), femoral nerve, and saphe-
nous nerve, are at risk, especially during the traditional
subvastus approach.12
An anterio-medial distal femoral “safe zone” was pro-
posed by Kim et al,13 who used computed tomography to
create a detailed map of the femoral artery and its distance
from the femur.13 Jiamton and Apivatthakakul9 evaluated the
feasibility of minimally invasive plate osteosynthesis (MIPO)
in the distal medial femur by studying the anatomical relation-
ships and structures at risk between the artery and the implant
in cadavers. Neither study investigated the smaller neurovas-
cular structures at risk during distal medial MIPO.
www.jorthotrauma.com | e357
Sirisreetreerux et al
Avoiding extensive soft-tissue dissection may reduce
morbidity. A description of all anatomic structures and
surgical references of the medial distal thigh is needed to
establish a true safe zone for MIPO of the medial femoral
condyle. To our knowledge, other than the studies mentioned
herein,6,10–12,14 no authors have described the smaller anatom-
ical structures at risk during a distal medial femur MIPO
approach.
Our aims were to (1) describe all relevant small
anatomic structures of the distal medial thigh at risk during
an MIPO approach to the distal medial thigh; (2) determine
their locations in relation to the adductor tubercle; and (3)
identify a safe zone for an MIPO approach to the medial distal
femur for fracture management.
MATERIALS AND METHODS
Eleven fresh-frozen cadaveric lower-half torsos (22
limbs) without history of trauma or surgery to the medial
thigh or knee were obtained from the local anatomy board.
Specimens were randomly assigned on availability. Dissec-
tions were performed by 3 authors. Each specimen was
dissected similarly on both lower medial thighs. Detailed
dissection of the neurovascular structures of the medial vastus
and adductor compartment was performed. The average
distances of these neurovascular structures to the palpable
prominence of the adductor tubercle were used to describe
a safe zone for a medial distal femur MIPO approach.
Measurements were recorded in millimeters using an elec-
tronic caliper (Mitutoyo ABSOLUTE Digimatic Series 500;
MSC Industrial Supply Co, Melville, NY).
Surgical Dissection
Each specimen was placed supine on a dissection table.
The landmark for incision was the middle portion of the medial
femoral condyle, which was defined by the halfway point from
the posteromedial border of the patella and the most posterior
aspect of the medial femoral condyle. A longitudinal incision
was made at the midsagittal plane of the medial thigh starting 1
cm proximal to the knee joint and extending to the proximal
one-third of the femur. When necessary, the incision was
performed more proximally for better exposure and dissection to
a maximum of 35 cm (see Figure, Supplemental Digital Con-
tent 1, http://links.lww.com/BOT/A753). The subcutaneous
layer was incised and elevated to expose the deep fascia. The
adductor tubercle was identified by palpation, and a Kirschner
(K)-wire was inserted into its center and most prominent part
from directly medial and perpendicular to the femur. This was
used as a reference point for our measurements. Subsequently,
we dissected the femoral triangle in the proximal thigh to iden-
tify the femoral vessels, the last branch of the femoral nerve
(Fig. 1A), adductor longus, and sartorius muscles. The sartorius
was retracted posteromedially, and the femoral artery and vein
were traced distally to Hunter canal, which was defined as the
point where the femoral vessels run deep to the sartorius and
through a channel of fascia between the vastus medialis
anterolaterally and adductor longus posteromedially. Sharp
and blunt dissections were performed through the vastus
medialis muscle anterior to Hunter canal down to the
e358 | www.jorthotrauma.com
Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
J Orthop Trauma  Volume 30, Number 11, November 2016
midpoint of the femoral shaft, exactly halfway between the
anterior and posterior cortex. A K-wire was inserted at this
point as a reference (see Figure, Supplemental Digital
Content 2, http://links.lww.com/BOT/A754). The distances
from Hunter canal to the midshaft of the femur and adductor
tubercle were measured (see Figures, Supplemental Digital
Contents 3 and 4, http://links.lww.com/BOT/A755 and
http://links.lww.com/BOT/A756). Hunter canal was then
incised, and the vastus medialis was dissected from the
adductor magnus tendon and the intermuscular septum.
The vastus medialis was then split longitudinally at its mid-
portion along the medial femur and retracted anteriorly away
from the adductor magnus, allowing better visualization of
the femoral vessels. The femoral vessels were traced from
Hunter canal to the adductor hiatus, where they pass through
the adductor magnus posterior and distal to the upper part of
the popliteal fossa. The superior medial genicular artery
(SMGA), which is situated close to the adductor tubercle,
was identified and its distance to the adductor tubercle mea-
sured. The DGA, which stems from the femoral artery, and
its branches were identified. At the midshaft of the femur at
the level of the adductor hiatus, a K-wire was positioned to
measure the distances of the DGA to the adductor tubercle,
the adductor hiatus to the femoral shaft, and the adductor
hiatus to the adductor tubercle (Fig. 1B). We recorded all
paths of the DGA and classified them according to the sys-
tem of Dubois et al2 Then, an incision was made through the
adductor magnus to approach the popliteal fossa, and we
measured the distances from the sciatic nerve and popliteal
artery and vein to the adductor tubercle.
Finally, the last branch of the femoral nerve, which
originated from the femoral nerve proximally, was identified
and traced distally to the vastus medialis (Fig. 1A). The point
where the nerve penetrated the muscle was identified and
used as a reference to measure the distance to the adductor
tubercle.6
Analysis
Statistical analysis was performed to determine
weighted means, medians, and standard deviations using
Microsoft Excel (Microsoft Corp, Redmond, WA).
RESULTS
Ten specimens (20 of the 22 lower limbs) were men.
The mean age was 77 years (range, 52–93 years). No speci-
mens had evidence of previous injury or surgery.
The mean distances of Hunter canal from the adductor
tubercle and femoral shaft were 160 6 31.4 mm proximal and
31.8 6 9.21 mm posterior, respectively. The mean distances
of the adductor hiatus from the adductor tubercle and femoral
shaft were 94 6 18.3 mm proximal and 31.7 6 7.78 mm
posterior, respectively. DGAs were found in all specimens,
with a mean distance of 98.4 6 16.0 mm proximal to the
adductor tubercle. Nineteen of the 22 DGAs arose from the
femoral artery before the femoral artery entered the adductor
hiatus. DGAs were classified using the system of Dubois et al2
as 14 type Ia, 2 each of type Ib and Ic, 1 type IIc, and 3 type III.
The osteoarticular branch of the DGA ran along the adductor
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
J Orthop Trauma  Volume 30, Number 11, November 2016
FIGURE 1. A, Dissection demon-
strating the last branch of the fem-
oral nerve to vastus medialis (white
arrow). The retractor is positioned
proximal toward the pelvis. B,
Measurement of distance from DGA
(thick white arrow) to adductor
tubercle (thin white arrow).
magnus tendon, anastomosing with the SMGA, whereas the
muscular branch traveled along the posterior border of the
vastus medialis. The saphenous branch of the DGA ran poste-
riorly along with the saphenous nerve (Fig. 2). Distances are
summarized in Supplemental Digital Content 5 (see Figure,
http://links.lww.com/BOT/A757).
The SMGA was found in 20 of the 22 specimens, with
a mean distance of 25.4 6 8.6 mm anterior to the adductor
tubercle. The popliteal vessels were situated posterior to the
adductor magnus, at a mean distance of 58.6 6 7.87 mm
posterior to the adductor tubercle.
The most distal branch of the femoral nerve that supplies
the vastus medialis before entering into the muscle was found
at a mean of 143 6 63.0 mm proximal to the adductor tubercle.
In 19 limbs, the femoral nerve branches ran along the poster-
omedial border of the vastus medialis (Table 1 and Fig. 1A).
DISCUSSION
In contrast to previous studies,9,13,14 our results demon-
strate a safe zone on the medial side of the distal femur up to
approximately 160 mm proximal and 32 mm posterior to the
adductor tubercle. In this region, a percutaneous plate may be
safely applied.
Lateral plating alone for complex fractures of the distal
femur (AO types C2 to C3) may be insufficient to prevent
varus collapse, increasing the risk of fixation failure. In such
FIGURE 2. Dissection showing the femoral artery, superior
medial genicular artery (SMGA), adductor hiatus (3 thin black
arrows), type Ic DGA (asterisk) with muscular branch (thick
black arrow), osteoarticular branch (triangle), and saphenous
branch (diamond).
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
MIPO in Medial Femoral Condylar Fracture
cases, combined medial and lateral exposure may be
required.1 Jiamton et al9 reported that the medial side of the
thigh was safe up to 60% of the length of the femur from the
adductor tubercle. However, they used a second, more prox-
imal incision and dissection to prevent injuries to the proxi-
mal vascular structures (ie, the descending branch of the
lateral femoral circumflex artery). Hence, the suggested safe
zone cannot be considered truly safe in the setting of a classic
MIPO approach.
Distally, a Von Langenbeck15 or Insall8 medial para-
patellar approach provides excellent exposure to the distal
medial femur. However, splitting the common quadriceps
tendon can lead to patellofemoral instability16–18 and compro-
mise blood supply to the medial and lateral distal femur and
patella, increasing the risk for avascular necrosis and non-
union.19–22
Kim et al13 studied the safe zone on the distal medial
aspect of the thigh in 30 patients. They found the last perfo-
rating artery to be an average of 80–150 mm from the lower
margin of the lesser trochanter, passing distally and posteri-
orly behind the adductor longus and ending at the distal third
of the thigh in a small muscular branch. They concluded the
safe zone for the distal medial femur to be anterior-medial and
that medial plating could be safely performed up to approx-
imately 150 mm from the lesser trochanter.13 However, they
did not investigate risk to other vascular structures or nerves
commonly found in this anatomical area, and they
www.jorthotrauma.com | e359
Sirisreetreerux et al J Orthop Trauma  Volume 30, Number 11, November 2016

who described the DGA as originating from the femoral artery

TABLE 1. Distances Between Major Neurovascular Structures
at an average angle of 35 degree, with the muscular branch
in 22 Fresh-Frozen Cadaveric Lower-Limb Specimens
running oblique into the vastus medialis muscle and anasto-
Parameters Mean 6 SD, mm Median, mm mosing at the level of the patella. They commented that this
Hunter canal to adductor tubercle 160 6 31.4 155 vessel was vulnerable during a traditional subvastus approach,
Hunter canal to femoral shaft 31.8 6 9.21 32 recommending a 50-degree safety angle to avoid injuring the
Adductor hiatus to adductor tubercle 94 6 18.3 93 DGA to the vastus medialis during a midvastus knee approach.
Adductor hiatus to femoral shaft 31.7 6 7.78 31.5 This contradicted the results of Jojima et al,6 who emphasized
Branch of femoral nerve piercing 143 6 63.0 147 potential injuries to the femoral nerve branches supplying the
vastus medialis to adductor vastus medialis rather than the DGA. We found a predomi-
tubercle
nantly posterior nerve pattern with a variable entry point into
DGA to adductor tubercle 98.4 6 16.0 102
the vastus medialis, which contradicts the findings of Jojima
SMGA to adductor tubercle 25.4 6 8.6 25.5
et al.6 For intraoperative application, our findings correlate with
Popliteal vessels to adductor tubercle 58.6 6 7.87 58.5
a greater safe zone to major nerve and vascular structures but
highlight the possibility of injury to smaller arteries such as
recommend plate positioning be anterior-medial, which may branches of the DGA (Fig. 3).
not allow for adequate buttress of the medial condyle. Based on our results, when using a medial midincision,
Compared with that of Kim et al,13 our safe zone is the estimated crossing point of the muscular branch of the
larger from anterior to posterior, which might be explained vastus medialis would be approximately 50 mm from the
by the laxity of cadaveric tissue or because of specimen age adductor tubercle. This is a major difference from the findings
and sex distribution. We used 1 female and 10 male speci- of Kim et al,13 who did not consider the DGA pattern and
mens, from 52 to 93 year old, whereas they used 12 female nerves to the vastus medialis.
and 18 male specimens, from 24 to 73 year old. We selected the adductor tubercle as the reference point
Most of the DGAs found in our specimens were type for our dissection, in contrast to Basarir et al14 and Visser
Ia.2 In all specimens, the osteoarticular branch ran along the et al,12 who used the superior patella pole or the midportion
adductor magnus tendon, anastomosing with the SMGA, if it of the femur, which was thought to be the most reliable
was present, whereas the muscular branch ran into the vastus anatomical landmark. Although our measurements seem sim-
medialis after reaching its posterior border. The saphenous or ilar to those of these authors, our study offers measurements
cutaneous branch traveled with the saphenous nerve (Fig. 2). from more reproducible landmarks.
One structure at risk during MIPO is the muscular branch Our results suggest avoiding a midvastus incision. Care
of the DGA, which is situated in the vastus medialis, crossing is needed during a distal medial femur MIPO approach to
50 mm proximal to the adductor tubercle at an approximate 35- avoid injuring the DGA and the nerve to the vastus medialis.
degree angle of projection in relation to the longitudinal axis of The risk of injury to these structures can be reduced using
the femur.12 Another structure at risk is the osteoarticular branch blunt dissection, protective sleeves, and careful surgical
along the adductor magnus, which can be injured if the vastus technique.
medialis is dissected and elevated directly anterior from the The distal medial aspect of the thigh is at low risk for
adductor magnus or intermuscular septum. Anterior dissection injury to major vascular structures, as well as smaller neuro-
of the vastus medialis puts at risk the distal branch of the vascular structures when approached with care. A safe zone of
femoral nerve to the vastus medialis, which usually runs pos- 160 mm for plate application extends from the adductor
terior to the muscle. With proper soft-tissue handling, the nerve tubercle up to Hunter canal. Anterior-medial plating13 should
may be protected by the muscle during percutaneous screw not be performed to avoid unnecessary mobilization of the
fixation. vastus medialis and possible injury to the DGA and femoral
Injury to the distal branch of the femoral nerve to the nerve branch to the vastus medialis.
vastus medialis is a risk of medial knee approaches.6,12 Our We recommend an MIPO incision be performed 1 cm
findings are similar to those of Visser et al12 and Basarir et al,14 anterior to the adductor tubercle, starting 1 cm proximal to the

FIGURE 3. Illustration showing the ideal MIPO

incision with the crossing point with the DGA at
approximately 50 mm (green dot). (Adductor
tubercle, pink dot; AVG Fem a., average path of
the femoral artery) (Used with permission 2015
from The Johns Hopkins University School of
Medicine.)

e360 | www.jorthotrauma.com Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.

Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
J Orthop Trauma  Volume 30, Number 11, November 2016
joint line and not exceeding 5 cm proximal to the adductor
tubercle to avoid injury to the osteoarticular and muscular
branches of the DGA. The vastus medialis should be split
longitudinally, approximately 1 cm anterior to the adductor
magnus, and not elevated. This way the osteoarticular branch
of the DGA remains protected by the vastus medialis.
Limitations of our study include the lack of specimen
dissection and analysis after an MIPO trial. Although our
safe zone is limited up to Hunter canal, we believe that for
a classic MIPO approach for distal medial femur fixation,
this safe zone is sufficient, without the need for a second,
more proximal incision. Injectable material such as neo-
prene instead of regular dye, as used in a previous study,3
may have helped to better visualize smaller intramuscular
vessels and localize possible vascular injuries when per-
forming MIPO in nondissected specimens. However, this
technique is expensive and requires a specialized labora-
tory. There may be interobserver variability in the tech-
nique used (palpation only) to insert our reference K-wire
into the most prominent point of the adductor tubercle.
Measuring its diameter first to determine the exact center
point would have been an alternative. However, anatomic
variability across specimens (ie, different widths of the
adductor tubercle) may result in inaccurate measurements.
Despite these weaknesses, we believe we have accurately
identified and recorded all extramuscular vessels and
defined the location of the distal branch of the femoral
nerve to the vastus medialis.
Strengths of our study include a thorough description of
major and minor structures of the distal medial femur.
Anatomical description of the minor structures, including
branches of the DGA and the nerve to the vastus medialis,
complements previous MIPO studies of the medial side of the
femur. Our results emphasize that an MIPO approach to the
distal medial knee presents low risk of injury to neurovascular
structures when proper soft-tissue handling and careful
surgical technique about the vastus medialis are used. Plates
can be positioned on the medial side from the adductor
tubercle to approximately Hunter canal.
We must consider that in an actual fracture, the
distances to neurovascular structures might be altered, and
some of the minor structures such as the DGA branches
may be disrupted by fracture displacement or the trauma
itself. Further injury to these structures may be preventable
by appropriate soft-tissue handling and protective sleeves
during MIPO. Surgeons should be aware of these changes
and consider them during MIPO. Posttraumatic studies
using computed tomography angiography before and after
surgery might also provide reliable information about
possible associated vascular injuries with distal femur
fractures.
CONCLUSIONS
This study shows that an MIPO approach to the distal
femur seems to be safe for major neurovascular structures
when approaching the medial distal femur up to Hunter canal.
Lesser vascular structures, including branches of the DGA,
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright Ó 2016 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
MIPO in Medial Femoral Condylar Fracture
might be at risk, especially during percutaneous screw
insertion. This risk may be reduced by minimally invasive
instrumentation and blunt exposure to bone.
REFERENCES
1. Sanders R, Swiontkowski M, Rosen H, et al. Double-plating of commi-
nuted, unstable fractures of the distal part of the femur. J Bone Joint Surg
Am. 1991;73:341–346.
2. Dubois G, Lopez R, Puwanarajah P, et al. The corticoperiosteal
medial femoral supracondylar flap: anatomical study for clinical eval-
uation in mandibular osteoradionecrosis. Surg Radiol Anat. 2010;32:
971–977.
3. Garcia-Pumarino R, Franco JM. Anatomical variability of descending
genicular artery. Ann Plast Surg. 2014;73:607–611.
4. Julien TP, Gravereaux E, Martin S. Superior medial geniculate artery
pseudoaneurysm after primary total knee arthroplasty. J Arthroplasty.
2012;27:323. e313–326.
5. Shaw A, Stephen AB, Lund JN, et al. Geniculate arterial pseudoaneur-
ysm formation following trauma and elective orthopaedic surgery to the
knee: 2 case reports and a review of the literature. J Radiol Case Rep.
2009;3:12–16.
6. Jojima H, Whiteside LA, Ogata K. Anatomic consideration of nerve
supply to the vastus medialis in knee surgery. Clin Orthop Relat Res.
2004;423:157–160.
7. Hoppenfeld S, deBoer P, Buckley R. Surgical Exposures in Orthopae-
dics: The Anatomic Approach. 5th ed. Philadelphia, PA: Wolters Kluwer;
2016.
8. Insall J. A midline approach to the knee. J Bone Joint Surg Am. 1971;53:
1584–1586.
9. Jiamton C, Apivatthakakul T. The safety and feasibility of minimally
invasive plate osteosynthesis (MIPO) on the medial side of the femur:
a cadaveric injection study. Injury. 2015;46:2170–2176.
10. Kim JM. Quadriceps dislocation medial approach for intraarticular and
medial structures of the knee. Orthopedics. 1991;14:1147–1151.
11. Roysam GS, Oakley MJ. Subvastus approach for total knee arthroplasty:
a prospective, randomized, and observer-blinded trial. J Arthroplasty.
2001;16:454–457.
12. Visser J, Brinkman JM, Bleys RL, et al. The safety and feasibility of
a less invasive distal femur closing wedge osteotomy technique: a cadav-
eric dissection study of the medial aspect of the distal femur. Knee Surg
Sports Traumatol Arthrosc. 2013;21:220–227.
13. Kim JJ, Oh HK, Bae JY, et al. Radiological assessment of the safe
zone for medial minimally invasive plate osteosynthesis in the distal
femur with computed tomography angiography. Injury. 2014;45:
1964–1969.
14. Basarir K, Erdemli B, Tuccar E, et al. Safe zone for the descending
genicular artery in the midvastus approach to the knee. Clin Orthop Relat
Res. 2006;451:96–100.
15. Von Langenbeck B. Zur resection des kniegelenks. Verhandl d Deutschen
Gesellsch f Chir VII. 1878;34:36.
16. Engh GA, Parks NL, Ammeen DJ. Influence of surgical approach on
lateral retinacular releases in total knee arthroplasty. Clin Orthop Relat
Res. 1996;331:56–63.
17. Parentis MA, Rumi MN, Deol GS, et al. A comparison of the vastus
splitting and median parapatellar approaches in total knee arthroplasty.
Clin Orthop Relat Res. 1999;367:107–116.
18. White RE Jr, Allman JK, Trauger JA, et al. Clinical comparison of the
midvastus and medial parapatellar surgical approaches. Clin Orthop Re-
lat Res. 1999;367:117–122.
19. Bonutti PM, Miller BG, Cremens MJ. Intraosseous patellar blood supply
after medial parapatellar arthrotomy. Clin Orthop Relat Res. 1998;352:
202–214.
20. Tria AJ Jr, Harwood DA, Alicea JA, et al. Patellar fractures in posterior
stabilized knee arthroplasties. Clin Orthop Relat Res. 1994;299:131–138.
21. Wetzner SM, Bezreh JS, Scott RD, et al. Bone scanning in the assess-
ment of patellar viability following knee replacement. Clin Orthop Relat
Res. 1985;199:215–219.
22. Windsor RE, Scuderi GR, Insall JN. Patellar fractures in total knee
arthroplasty. J Arthroplasty. 1989;4(Suppl):S63–S67.
www.jorthotrauma.com | e361