Effect of Anatomic Variations in the Anterior Tibial Artery on Risk of Injury During Orthopaedic Knee Surgeries

Abstract

Background:

Injury to the posterior vasculature is a potential complication in orthopaedic knee surgery that may be associated with variations in its anatomy, such as the type II-A2 variant, which places the anterior tibial artery (ATA) in closer proximity to the tibia. However, how close surgical instrumentation comes to injuring the ATA is not well described.

Purpose:

To determine how the type II-A2 variant of the popliteal vasculature affects proximity of the ATA to instrumentation for orthopaedic knee procedures.

Study Design:

Cross-sectional study; Level of evidence, 3.

Methods:

A total of 222 knee magnetic resonance imaging (MRI) scans from a single academic tertiary referral center were evaluated, and ATA branching patterns were characterized. The distances from the ATA to simulated instrumentation of high tibial osteotomy (HTO), posterolateral corner anatomic reconstruction, posterior cruciate ligament (PCL) reconstruction, lateral meniscus posterior horn repair, and lateral meniscus posterior root repair on axial plane MRI scans were measured by 2 authors independently using imaging software. Intrarater and interrater reliability of the measurements was calculated using the intraclass correlation coefficient.

Results:

ATAs with the type II-A2 pattern passed anterior to the popliteus on 3.15% (n = 7) of the 222 MRI scans. The distance between the ATA and the simulated instrumentation was significantly closer in type II-A2 compared with normal (type I) knees for the lateral meniscus posterior root repair tunnel (11.1 vs 15.7 mm; P = .014), HTO cuts (0.6 vs 8.2 mm; P < .001), and PCL reconstruction tunnel (4.1 vs 11.7 mm; P < .001). Interrater reliability was good to excellent for all measurements, and intrarater reliability ranged from moderate to excellent.

Conclusion:

HTO cut, PCL reconstruction tunnel, and lateral meniscus posterior root tunnel instrumentation were significantly closer to the ATA in knees with type II-A2 anatomy compared with normal (type I) anatomy. Careful analysis of vasculature using MRI may be of utility for select surgery about the knee to guide surgical technique.

Keywords

anterior tibial artery meniscal repair MRI high tibial osteotomy root repair vascular injury knee posterolateral corner reconstruction posterior cruciate ligament reconstruction

One of the potential complications of orthopaedic surgeries of the knee is vascular injury, with an estimated incidence of 0.017%.² This vascular injury can result in in acute intraoperative blood loss or more subacute findings such as poor perfusion. Despite undergoing open surgery to address the vascular injury, more than three-quarters of patients continue to experience residual symptoms.²

Orthopaedic knee procedures also pose a risk of vascular injuries. This includes the saw cut of a high tibial osteotomy (HTO), implant placement for lateral meniscus posterior horn repair, tunnel drilling for the posterior root of lateral meniscus repair, and proximal tibial tunnel placement for LaPrade posterolateral corner (PLC) and posterior cruciate ligament (PCL) reconstructions. With proper care, these techniques are safe with a low risk of arterial or peripheral nerve injury.^8,13,16,19

Knowledge of posterior knee anatomy with respect to surgical instrumentation is imperative in preventing injury in these procedures. However, specific patient anatomy is also important. Typically, the popliteal artery gives rise to the anterior tibial artery (ATA) after passing the joint line, becoming the short tibioperoneal trunk that then divides into the peroneal and posterior tibial arteries.⁸ However, there are well-recognized variations, including high proximal branching above the popliteus, classified as type II by Kim et al,⁷ which has a reported prevalence of 3% to 7%.^8,9,18 When the ATA courses posterior to the popliteal muscle, it is classified as type II-A1⁷; however, when it passes between the anterior surface of the popliteal muscle and the posterior tibial cortex, it is classified as type II-A2.⁶ The type II-A2 subvariant is found in about 2% of the population and effectively pins the vessel against the posterior aspect of the tibia at the level of the knee.⁸ Anatomic variants such as this can place the patient at particularly high risk for arterial injury during surgery, highlighting the importance of understanding patient-specific anatomy. Although various methods such as ultrasound and computed tomography arteriography can be used to determine arterial anatomy, noncontrast magnetic resonance imaging (MRI) of the knee is a commonly available method for orthopaedic sports surgeons.⁵

The purpose of this study was to utilize routine, noncontrast MRI of the knee to assess the prevalence of the type II-A2 subvariant of the ATA and to measure the distance from instrumentation in orthopaedic surgeries to the aberrant vessel. We hypothesized that the presence of the type II-A2 subvariant would result in closer proximity to the surgical instrumentation sites for HTO, implant placement for lateral meniscus posterior horn repair, and tibial tunnel placement for PLC reconstructions, lateral meniscus posterior root repair, and PCL reconstructions, when compared with the normal anatomic position of the popliteal artery or ATA.

Methods

This study included a retrospective series of 222 consecutive knee MRI scans obtained at a single academic tertiary referral center between March and May 2022. For all images, the patients lay supine with the knee extended. Exclusion criteria were substantial artifact impairing interpretation of the imaging, lesion with mass effect on posterior structures, or fracture with significant displacement of the distal femur or tibial plateau. The MRI scans consisted of sagittal, coronal, and axial cuts. The scans were reviewed by 2 authors (C.F. and G.B.), both of whom were residents specializing in orthopaedic surgery (postgraduate years 3 and 4).

The height of the branching of the ATA was measured (Figure 1), and the anatomy was classified according to Kim et al.⁷ Broadly, type I (normal) was defined as branching of the popliteal artery at or below the inferior border of the popliteus muscle, type II (including subvariants II-A1 and II-A2) as high branching of the popliteal artery, and type III as hypoplastic or aplastic. Measurements were obtained using Sectra imaging software.

Figure 1.

T2-weighted (A) axial view and (B) sagittal view magnetic resonance imaging scans showing high branching of the anterior tibial artery. The yellow line demarcates the level of the axial cut.

The lateral meniscus was measured in the axial plane at the level of the meniscus. Points representing the posterior-most aspects of the root and posterior horn were selected, and these were measured to the artery (Figures 2 and 3).

Figure 2.

T2-weighted magnetic resonance imaging scan of a left knee with high branching of the anterior tibial artery showing measurement from the artery to the posterior horn of the lateral meniscus. The orange double arrow represents the distance from the tunnel to the anterior tibial artery.

Figure 3.

T2-weighted magnetic resonance imaging scans of a left knee with high branching of the anterior tibial artery. (A) Axial view. The orange double arrow represents the distance from the tunnel to the anterior tibial artery. (B) Sagittal view.

The HTO measurement was performed using the axial plane. Using Sectra imaging software, we then reoriented the axial plane to mimic the oblique cut of the osteotomy (Figure 4). The cut was selected at the level of the proximal extent of the osteotomy just proximal to the fibular head 1.5 cm from the plateau. A point along the posterior tibial cortex closest to the vessel during the procedure was selected as long as it was medial to the lateral-most extent of the osteotomy (1 cm from the lateral cortex). The distance from this point to the nearest point of the popliteal artery or ATA was measured in this reoriented axial plane.

Figure 4.

T2-weighted magnetic resonance imaging scans of a left knee with high branching of the anterior tibial artery. (A) Axial reformat view. The orange bracket represents the distance from the tunnel to the anterior tibial artery. (B) Coronal reformat view. The yellow line represents the plane of the high tibial osteotomy cut. (C) Sagittal reformat view in the plane of the shortest distance between the artery and the posterior cortex. The orange arrow indicates the anterior tibial artery.

For the proximal tibial tunnel of the PLC reconstruction, we utilized the axial plane at an estimated 1 cm medial and proximal to where the anticipated fibular tunnel would be made (Figure 5). The fibular tunnel was estimated to be at a level 1 cm distal to the tip of the fibular styloid. The templated tibial tunnel was set to have a radius of 5 mm to approximate a 10-mm graft tunnel.

Figure 5.

T2-weighted magnetic resonance imaging scans of a left knee with high branching of the anterior tibial artery. The circle with a 5-mm radius represents the posterolateral corner tibial tunnel. The orange arrow represents the distance from the tunnel to the anterior tibial artery.

For the PCL reconstruction measurements, the distance from PCL pin placement was measured. In the sagittal plane, a trajectory was made 45° off the longitudinal axis of the tibia and 7 mm superior to the posterior cortex. Then, in the axial plane, the distance between the intersection of the modeled pin trajectory and posterior cortex and the artery was measured (Figure 6).

Figure 6.

T2-weighted magnetic resonance imaging scans of axial and sagittal planes with normal branching of the anterior tibial artery. (A) Axial view. The orange double arrow represents the distance from the posterior cruciate ligament pin exit to the artery. (B) Sagittal reformat view with projected trajectory of the pin. (C) Coronal reformat view. The yellow line represents the sagittal placement of the pin.

All measurements by both reviewers were used to calculate interrater reliability, while 25 randomly selected knees were used for repeat measurement to calculate intrarater reliability. The repeat measurements for intrarater reliability were performed by 1 author (C.F.) six months after the intial time period using the same approach as before, from the unmarked MRI scan.

Statistical Analysis

Results are reported as means with standard deviations. Two-sample t tests were used to assess the difference in distances between the normal division position compared with the type II variants.

The (2,1) intraclass correlation coefficient (ICC) was used to determine inter- and intrarater reliability, with ICCs >0.90 considered excellent, 0.75 to 0.90 considered good, 0.50 to 0.75 considered moderate, and <0.50 considered poor reliability. Statistical analyses were conducted using RStudio (Version 2023.03.1+446; Posit Software). The threshold for statistical significance was set at P < .05.

Results

The search identified 232 patients with 254 knee MRI scans. After screening, 32 knees were removed, yielding 200 patients with 222 knee MRI scans. The mean patient age was 31.5 years, and 51% were male, with 48% of the cases being the right knee. In total, 92 (41.4%) of the knees did not have branching of the division within the imaging study, which was presumed to indicate normal distal branching. There were 8 (3.6%) type II popliteal artery variants, of which 7 knees (3.15%) were type II-A2 and 1 knee was type II-A1 (0.45%) (Table 1).

Table 1

Descriptive Characteristics of Knee Imaging (N = 222)^a

	Type I (n = 214)	Type II-A1 (n = 1)	Type II-A2 (n = 7)
Age, y	31.6 ± 9.5	49	25.9 ± 9.4
Male sex	52	0	42.9
Right side affected	49	0	29
Branching distance, mm^b	NA	13.6	17.5 ± 9.0

Data are presented as mean ± SD or percentage. MRI, magnetic resonance image; NA, not applicable.

Branching distance indicates the distance from the branching of the anterior tibial artery from the joint line. Positive values indicate proximal to the joint.

The mean distance from the different surgical procedures to either the ATA or the popliteal artery (whichever was closer to the instrumentation of the type II anatomic variants) is shown in Table 2. The distance between the ATA and the simulated instrumentation was significantly closer in type II-A2 compared with normal (type I) knees with regard to posterior root repair of the lateral meniscus (11.1 ± 3.6 vs 15.7 ± 3.0 mm; P = .014), HTO cuts (0.6 ± 0.3 vs 8.2 ± 2.8 mm; P < .001), and the PCL reconstruction tunnel (4.1 ± 2.0 vs 11.7 ± 2.7 mm; P < .001). There were no significant differences for the posterior horn of the lateral meniscus (P = .127) and the tibial tunnel of the PLC reconstruction (P = .189).

Table 2

Distances From Instrumentation of Surgical Intervention to Either the ATA or Popliteal Artery^a

	Type I (n = 214)	Type II-A1 (n = 1)^b	Type II-A2 (n = 7)^b	P
Posterior horn of lateral meniscus, mm	9.7 ± 2.7	6.3	7.7 ± 3.0	.127
Posterior root of lateral meniscus, mm	15.7 ± 3.0	11.7	11.1 ± 3.6	.014
PLC reconstruction tunnel, mm	14.5 ± 4.0	7.8	11.8 ± 4.8	.189
HTO, mm	8.2 ± 2.8	3.6	0.6 ± 0.3	<.001
PCL reconstruction tunnel	11.7 ± 2.7	7.3	4.1 ± 2.0	<.001

Data are presented as mean ± SD. Boldface P values indicate a statistically significant difference between groups (P < .05). ATA, anterior tibial artery; HTO, high tibial osteotomy; PCL, posterior cruciate ligament; PLC, posterolateral corner.

In instances of high branching, the closer of the 2 distances (ATA or popliteal artery) was included.

The ICCs for interrater and intrarater reliability are shown in Table 3. The interrater reliability was excellent for the posterior horn of the lateral meniscus (ICC, 0.96) and good for the other surgeries (ICCs, 0.77-0.82). The intrarater reliability was excellent for the posterior horn of the lateral meniscus and PCL reconstruction tunnel (ICCs, 0.97 and 0.94, respectively), good for the posterior root of the lateral meniscus and HTO saw cuts (ICCs, 0.76 and 0.87, respectively), and moderate for PLC reconstruction (ICC, 0.68).

Table 3

ICCs for Interrater and Intrarater Reliability^a

	ICC (95% CI)
	Interrater Reliability	Intrarater Reliability
Posterior horn lateral meniscus	0.96 (0.94-0.97)	0.97 (0.94-0.99)
Posterior root lateral meniscus	0.79 (0.66-0.86)	0.76 (0.51-0.89)
PLC tibial tunnel	0.77 (0.68-0.84)	0.68 (0.40-0.85)
HTO	0.77 (0.65-0.84)	0.87 (0.74-0.94)
PCL	0.82 (0.75-0.87)	0.94 (0.87-0.97)

HTO, high tibial osteotomy; ICC, intraclass correlation coefficient; PCL, posterior cruciate ligament; PLC, posterolateral corner.

Discussion

Type II-A2 aberrant branching of the ATA is an important consideration in surgical planning for knee procedures. Inadvertent injury can cause arterial transection, pseudoaneurysms, and potentially even limb loss. The popliteus muscle typically provides protection to the posterior vascular structures during orthopaedic knee surgeries. However, in type II-A2 aberrant high branching of the popliteal artery, the ATA may course immediately posterior to the proximal tibial cortex, anterior to the popliteus. In our study, we found that the division could reliably be detected using routine noncontrast knee MRI scans. We found the type II-A2 branching to occur anterior to the popliteus in 3.15% of cases. This anatomic variation decreased the distance between instrumentation and the major artery to iatrogenic injury during soft tissue dissection, proximal tibial saw cuts, tibial tunnel drilling for PCL reconstruction, and implant placement for posterior horn meniscal repair.

The incidence of type II division in our study is consistent with previous reports in the literature that used MRI (2.1%-3.2%), angiography (2.9%), and a cadaveric dissection (2.4%).^8,17 Various studies in the United States, Europe, South Korea, and Japan have found the incidence of the type II-A2 variant to vary from 0.4% to 2.9%.^{3,4,6,10,12,15} Our study did yield a slightly higher prevalence than most studies, but it was not powered to detect significant differences in prevalence relative to the population. In our study, the division between normal (type I) and type II-A2 variants of the ATA was found closest to the area of surgical involvement in the HTO. The distances then increased for the following procedures, in order: PCL reconstruction, lateral meniscus posterior horn repair, PLC reconstruction, and lateral meniscus posterior root repair.

While previous studies have focused on evaluating the distance between the popliteal artery and the cortex or capsule,⁸ our study had the strength of examining specific instrumentation involved in different procedures for the presence of type II-A2 variants. We found that the mean distance from the aberrant type II-A2 tibial artery was significantly closer compared with normal type I anatomy for the cut of the HTO (0.6 vs 8.2 mm), PCL tunnel reconstruction (4.1 vs 11.7 mm), and lateral meniscus root repair (11.1 vs 15.7 mm). Considering the close proximity of the vessel to the posterior cortex of the tibia in patients with type II-A2 anatomy, it is important to appreciate this anatomic variant preoperatively and proceed carefully with certain procedures. For example, when performing an HTO, careful posterior dissection and placement of retractors to protect the artery from osteotomes or saws is crucial, as the osteotomy site was found to have the closest proximity to the ATA (type II-A2 vs type I: 0.6 vs 8.2 mm).

Our measurements of the distances from the aberrant type II-A1/type II-A2 tibial artery to the simulated posterior horn of the lateral meniscus or tibial tunnel of PLC reconstructions were not significantly closer to the popliteal artery when compared with normal type I anatomy. In the cases of the type II-A2 variants in this investigation, the ATA was anterior to the popliteus and coursed along the midline of the tibia rather than along the lateral aspect of the proximal tibia. As such, it was not brought that much closer to the more lateral structures measured for lateral meniscus posterior horn repair and PLC reconstruction.

While differences in type I and type II anatomy were significantly closer for the PCL reconstruction, HTO, and lateral meniscus root repair, it is important to also appreciate the overall amount of distance between the area of surgical involvement and the ATA. When performing lateral meniscus posterior horn repair, the artery is in close proximity (mean difference, 7.7 mm in type II-A2 vs 9.7 mm in type I). While there was not a statistical difference between the normal type I and type II positions, the overall distance is quite small and less than what we found for the lateral meniscus root repair (mean difference, 11.1 mm in type II-A2 vs 15.7 mm in type I). In fact, many patients with and without the high division had <10 mm between the artery wall and peripheral meniscal edge. Thus, the ATA may be injured with the use of most all-inside devices if used from the anterolateral portal.¹⁹

Interrater reliability was good or excellent for all the procedures studied. However, intrarater reliability was only moderate for the PLC reconstruction tunnel while good for HTO and posterior root of the lateral meniscus and excellent for posterior horn of the lateral meniscus and PCL tunnel pin. The boundaries of the posterior horn of the lateral meniscus were clearly delineated and had the highest ICCs across both parameters. In contradistinction, our simulated reconstruction of the PLC reconstruction tibial tunnel may not have been as obvious. Although it was simulated 1 cm medial and proximal to where the anticipated fibular tunnel would be made, having fewer concrete landmarks may have introduced some variation into our measurements.

When considering clinical applications, our investigation does have several limitations. Measurements were obtained in patients who underwent knee MRI in the extended position. Although not compared with a gold standard vascular study, we believe that MRI accurately detects anatomic variants. Frequently during procedures where the popliteal vessels are at risk, the knee can be flexed in an attempt to increase the distance between bone and vessels, although this is contested in more recent studies.^11,14 However, the artery likely would not change position much in type II-A2 variants. With the aberrant ATA held against the tibia by the popliteus muscle, we speculate that knee flexion would not significantly change the position of this artery. In addition, for the HTO procedure, which had the closest proximity and risk for injury, the osteotomy is generally made in the semiextended position to allow for posterior retractor placement. Measurements may differ from the actual surgical instrument trajectory performed in the operating room. For example, it is unlikely the device will be aimed directly at the artery in performing a lateral meniscus repair. Careful surgical techniques may increase the distance from the instrument to the artery. Additionally, the authors were not blinded to patient identification as medical record numbers (MRN) was used to find MRI scans. This or other remarkable features of the imaging may have allowed bias in interpretation. Authors were, however, blinded to each other's measurements. Lastly, differences in patient size were not accounted for and may be of interest in larger studies with more type II-A2 variants.

With the rise in popularity of outpatient surgery centers, it is even more important to understand patient-specific anatomy as resources such as blood and vascular surgeons may not be readily available in the event of a popliteal artery injury. Other factors such as scarring after knee injury or operations may further increase the risks of vascular injury during subsequent knee surgery.¹ Such factors should be incorporated into surgical planning and the preoperative checklist for at-risk surgeries. Preoperative MRI may be a valuable part of the planning even for procedures such as the HTO, where it may not always be ordered, if resources permit. In cases of aberrant anatomy, the orthopaedic surgeon may consider performing the case at a facility with adequate resources such as vascular surgery or blood or alter their surgical technique to minimize the risk of iatrogenic injury. Regardless of location, we recommend scrutiny of preoperative MRI scans to appreciate patient-specific anatomy and careful surgical technique to minimize the risk of vascular injury when an aberrant ATA is present.

Conclusion

In this study, we found type II-A2 branching to occur in 3.15% of knee MRI scans. Compared with type I anatomy, the artery in type II-A2 knees was significantly closer to hypothetical surgical instrumentation in repair of the posterior root of the lateral meniscus, PCL reconstruction, and HTO. The cut of the HTO was the shortest simulated distance with a mean separation of 0.6 mm. These results suggest that care should be taken to avoid vascular injury in patients with known type II-A2 variants undergoing the aforementioned procedures.

Footnotes

Acknowledgements

The authors thank their research coordinator, Monica Sri Vel, for her contributions to the publication.

Final revision submitted May 2, 2024; accepted June 11, 2024.

One or more of the authors has declared the following potential conflict of interest or source of funding: G.B. has received education payments from Evolution Surgical. J.Z. has received education payments from Evolution Surgical. S.S. has received education payments from Evolution Surgical; consulting fees from Flexion Therapeutics, Bioventus, Biorez, DJO, DePuy/Medical Device Business Services, Kinamed, Linvatec, Pacira Pharmaceuticals, Smith & Nephew, Vericel, JRF Ortho, LifeNet Health, and Olympus America; nonconsulting fees from Arthrex, Smith & Nephew, Synthes GmbH, Vericel, and Linvatec; royalties from Linvatec and Conmed; honoraria from Pacira Pharmaceuticals, Flexion Therapeutics, JRF Ortho, and Vericel; and hospitality payments from Aesculap Biologics. G.A. has received education payments from Arthrex and consulting fees from Bioventus. J.C. has received grant support from DJO and Arthrex, education payments from Arthrex and Smith & Nephew, and hospitality payments from Wright Medical and Zimmer Biomet. AOSSM checks author disclosures against the Open Payments Database (OPD). AOSSM has not conducted an independent investigation on the OPD and disclaims any liability or responsibility relating thereto.

Ethical approval for this study was obtained from Stanford University (reference No. 65440).

ORCID iDs

Geoffrey Abrams

Jonathan W. Cheah

References

Bartlett

Roberts

Wong

Risk to popliteal vessels in major knee surgery, an anatomical study and survey of vascular surgeons. JBJS Br Proc. 2004;86:468.

Bernhoff

Rudström

Gedeborg

Björck

Popliteal artery injury during knee replacement: a population-based nationwide study. Bone Joint J. 2013;95-B(12):1645-1649.

Celtikci

Ergun

Durmaz

Conkbayir

Hekimoglu

Evaluation of popliteal artery branching patterns and a new subclassification of the “usual” branching pattern. Surg Radiol Anat. 2017;39(9):1005-1015.

Day

Orme

Popliteal artery branching patterns—an angiographic study. Clin Radiol. 2006;61(8):696-699.

Helms

CA.

The impact of MR imaging in sports medicine. Radiology. 2002;224(3):631-635.

Kil

Jung

GS.

Anatomical variations of the popliteal artery and its tibial branches: analysis in 1242 extremities. Cardiovasc Intervent Radiol. 2009;32(2):233-240.

Kim

Orron

Skillman

JJ.

Surgical significance of popliteal arterial variants. A unified angiographic classification. Ann Surg. 1989;210(6):776-781.

Klecker

Winalski

Aliabadi

Minas

The aberrant anterior tibial artery: magnetic resonance appearance, prevalence, and surgical implications. Am J Sports Med. 2008;36(4):720-727.

Kropman

RHJ

Kiela

Moll

de Vries

JPPM

. Variations in anatomy of the popliteal artery and its side branches. Vasc Endovascular Surg. 2011;45(6):536-540.

10.

Mavili

Dönmez

Kahriman

Özaşlamacı

Özcan

Taşdemir

Popliteal artery branching patterns detected by digital subtraction angiography. Diagn Interv Radiol. 2011;17(1):80-83.

11.

Ninomiya

Dean

Goldberg

VM.

Injury to the popliteal artery and its anatomic location in total knee arthroplasty. J Arthroplasty. 1999;14(7):803-809.

12.

Prayer

Karnel

Schurawitzki

Angiographic demonstration of the branching of the popliteal artery in the arteries of the lower extremity—variations and frequency distribution. Article in German. Rontgenpraxis. 1990;43(3):100-103.

13.

Rubens

Wellington

Bouchard

AG.

Popliteal artery injury after tibial osteotomy: report of two cases. Can J Surg. 1990;33(4):294-297.

14.

Shetty

Tindall

Qureshi

Divekar

Fernando

KWK

. The effect of knee flexion on the popliteal artery and its surgical significance. J Bone Joint Surg Br. 2003;85(2):218-222.

15.

Szpinda

An angiographic study of the anterior tibial artery in patients with aortoiliac occlusive disease. Folia Morphol (Warsz). 2006;65(2):126-131.

16.

Tawes

Etheredge

Webb

Enloe

Stallone

RJ.

Popliteal artery injury complicating arthroscopic menisectomy. Am J Surg. 1988;156(2):136-138.

17.

Trotter

The level of termination of the popliteal artery in the White and the Negro. Am J Phys Anthropol. 1940;27(1):109-118.

18.

Yanik

Bulbul

Demirpolat

Variations of the popliteal artery branching with multidetector CT angiography. Surg Radiol Anat. 2015;37(3):223-230.

19.

Zaidi

Cobb

Bentley

Danger to the popliteal artery in high tibial osteotomy. J Bone Joint Surg Br. 1995;77(3):384-386.