- Email: [email protected]
The location of the popliteal artery in extension and 90 degree knee flexion measured on MRI
The Knee 16 (2009) 143–148
Contents lists available at ScienceDirect
The Knee
The location of the popliteal artery in extension and 90 degree knee flexion measured on MRI Jae Ho Yoo a,1, Chong Bum Chang b,⁎ a b
Department of Orthopaedic Surgery, Soonchunhyang University Hospital, Bucheon, 1174 Jung-Dong, Wonmi-Gu, Bucheon-Si, Gyeonggi-Do, 420-767, South Korea Joint Reconstruction Center, Department of Orthopaedic Surgery, Seoul National University Bundang Hospital, 300 Gumidong, Bundangu, Seongnamsi, Gyunggido (463-707), Korea
a r t i c l e
i n f o
Article history: Received 5 May 2008 Received in revised form 15 October 2008 Accepted 26 October 2008 Keywords: Knee Popliteal artery MRI
a b s t r a c t We measured the location of the popliteal artery (PA) in extension and 90 degree of knee flexion by magnetic resonance images (MRI) to provide practical information to avoid PA injury. The MRIs of 30 knees of Korean male subject whose mean age was 20.7 were acquired in knee extension and 90 degree flexion. The distance from the posterior aspect of knee joint to the PA was measured at three levels on the axial images and one sagittal image. At the joint line level, the PA was located lateral to the PCL 2.4 mm in extension and 3.2 mm in flexion (p = 0.247), and 3.9 mm in extension and 7.6 mm in flexion from the posterior capsule (p b 0.001). At 1 cm distal to the joint line, it is 2.7 mm in extension and 7.2 mm in flexion (p b 0.001), and at 2 cm distal to the joint line, 4.9 mm in extension and 9.7 mm in flexion from the posterior tibial cortex (p b 0.001). In sagittal plane, the nearest distance between PA and posterior tibial cortex was 1.8 mm in extension, and 6.2 mm in flexion (p b 0.001). The PA was located around 3 mm lateral to the PCL, and within 5 mm in extension and 10 mm in 90 degree flexion of the knee behind knee joint. It moves farther posteriorly in 90 degree flexion than in extension of the knee. The conventional wisdom of flexing the knee to prevent the PA injury was supported by this study. © 2008 Elsevier B.V. All rights reserved.
1. Introduction Although the injury to the popliteal artery (PA) during the surgeries around the knee joint has been regarded as a rare complication, the incidence has been reported to be as high as 1.8%[1–4]. Moreover, once it happens, the consequences are disastrous. The arthroscopy, the total knee arthroplasty (TKA), and the high tibial osteotomy are most commonly conducted operations, which also might jeopardize the PA. Tawes et al. reported that one of five arterial injuries during arthroscopic meniscectomy led to above knee amputation[5]. Arterial compromise after total knee arthroplasty has been reported to be between 0.03 and 0.2%, and resulted in amputation of the limb in up to 25% of the cases[6– 8]. There are also a few reports of direct injury to the PA during high tibial osteotomy (HTO)[9–11]. The course of the popliteal artery is also a concern in bicortical screw fixation of the tibial tubercle transfer[12]. With all the more increasing surgeries in the posterior aspect of the knee joint such as meniscal repair or transplantation, and the reconstruction of posterior cruciate ligament, and with the incremental incidence of total knee arthroplasty, the chance of injury rises accordingly. In addition, the current interest and application of the minimally invasive technique in TKA also could increase the possibility of the PA injury during tibial bone resection, all the more with limited surgical exposure.
The fact that injury does occur is not surprising as the vessel is in the worst position, within a few millimeters from the posterior aspect of the knee joint with only a small amount of fat being the only intervening tissue[13,14]. Although the location of the PA may change in different situations and vary among individuals, approximate allocation of the PA would obviously make the surgeon feel more secure. The conventional wisdom has it that flexing the knee joint puts the PA away from the posterior aspect of the knee joint when doing procedures around the posterior aspect of the knee, and this surgical aphorism has been supported by previous studies by MRI, angiography, or cadaveric dissection[12,14–20]. However, a study opposing to the conventional wisdom also exists that the PA paradoxically moved closer to the tibia by flexing the knee in 12 out of 20 knees[11]. The purpose of this study is to investigate the location of the PA in extension and 90 degrees of flexion at different levels of the knee joint to provide practical information to avoid popliteal artery injury. The conventional wisdom of flexing the knee so that the neurovascular bundles moves farther from the posterior aspect of the knee to prevent popliteal artery injury is also audited. 2. Materials and method 2.1. Subjects
⁎ Corresponding author. Tel.: +82 31 787 7201; fax: +82 31 787 4056. E-mail addresses: [email protected], [email protected] (J.H. Yoo), [email protected] (C.B. Chang). 1 Tel.: +82 32 621 5263; fax: +82 32 621 5016, Mobile: +82 11 671 2096. 0968-0160/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.knee.2008.10.009
The magnetic resonance images (MRI) of the 33 knees in 33 healthy Korean volunteers were obtained for the study. The inclusion criteria used for the volunteers were; (1) absence of knee pain, (2) no
144
J.H. Yoo, C.B. Chang / The Knee 16 (2009) 143–148
Fig. 1. The measurement was performed at three different levels of axial images; joint line level (I), 1 cm distal to joint level (II), 2 cm distal to joint level (III) in MRI of the knee in extension (a) and in 90 degree flexion (b).
history of knee injury, (3) normal range of lower limb alignment and no obvious knee deformity, (4) no instability of the knee joint, (5) none of other abnormal physical findings such as Baker's cyst. Three subjects were excluded from the study because of the inadequate quality of the images for exact measurement. Consequently, a total of 30 knee MRI scans from 30 male subjects were included, none of whom had bilateral scans. There were 15 right and 15 left knees. The mean age was 20.7 years ranging from 18 to 25. The mean weight, height, and body mass index (BMI) were, 66.1 kg (SD 10.8, range 51–80), 176.2 cm (SD 5.6, range 165–186), 21.8 kg/m2 (SD 2.0, range 16.7–27.3), respectively. 2.2. MRI acquisition All the MR images were obtained with the knee in full extension and 90 degrees of flexion using 1.5 Tesla scanner (Intera, Philips Medial Systems, The Netherlands). The axial and sagittal plane images were obtained using the following protocol; proton density-weighted images with fat suppression (repetition time 2378 ms and echo time 50 ms). Each section was 5 mm in thickness with a 0.5 mm intervening spacing and has a 512 × 512 matrix. The full extension of the knee was maintained in the standard surface coil with a corresponding height of pillow under the ankle to prevent the flexion of the knee. The 90 degree of the knee flexion simulated the surgical situations during various knee operations such as arthroscopy, arthroplasty, or osteotomy. To assume the actual position during operations, the subject lay on his back
allowing the neurovascular bundles behind the knee joint dangling down according to the gravity. 2.3. Measurement All MR images were digitally acquired through a picture-archiving and communication system (PACS), and the assessments were performed subsequently using PACS software (PiViewSTAR, INFINITT Ltd, Seoul, Korea) on a desk-top computer. We distinguished the arterial vasculature from the veins according to signal void as a result of faster flow velocity on the images[15]. The topographical relationship of neurovascular bundles including artery, vein, and nerve was also taken into consideration in that the popliteal artery is usually located in the most anterior side, followed by popliteal vein, and tibial nerve[16,21,22]. No anomalous duplication of the neurovascular bundle was detected. The measurement was conducted on the best images that show the concerning structures described below. In axial plane, the measurements were conducted in three different levels in both the extended and 90 degree flexed knee; the joint line, 1 cm distal to the joint line, and 2 cm distal to the joint line (Fig. 1). The orientation of the cutting slices was referenced by the joint line to standardize the measurement level. The aim of the measurements in the three different levels was to represent the procedures potentially involving knee surgeries. Most arthroscopic procedures in the knee joint are performed around the level of the joint line. The tibial bone cutting in knee arthroplasty or tibial tunnel creation in the reconstruction of the
Fig. 2. The measurement at joint line level (I in Fig. 1). The location of the PA is described by the distance between the posterior capsule (1) and the anterior wall of PA (2) in anteroposterior dimension, and that between the lateral border of posterior cruciate ligament (3) and medial wall of popliteal artery (4) in mediolateral dimension, on the MRIs of the knee in extension (a) and in 90 degree flexion (b).
J.H. Yoo, C.B. Chang / The Knee 16 (2009) 143–148
145
Fig. 3. The measurement at 1 cm distal to joint line (II in Fig. 1). The distance between the posterior cortex of proximal tibia (1) and the anterior wall of popliteal artery (2) is measured on the MRIs of the knee in extension (a) and in 90 degree flexion (b).
posterior cruciate ligament (PCL) would come around 1 cm distal to joint line. And the proximal tibial osteotomy in HTO takes place about 2 cm distal to the joint line. At each level we measured the distance from the germane anatomical landmarks to the PA. At the joint line level, the optimal image showing the meniscus or the tibial plateau was selected. Then, the distance between the posterior joint capsule and the anterior border of the PA was measured, by which we can assume the approximate distance to PA from the posterior capsule of the knee joint in arthroscopic surgery. In addition, the mediolateral location of PA with reference to the lateral border of PCL was also measured giving positive value if the PA lay lateral to the PCL and negative one if medial (Fig. 2). The lateral border of the PCL was selected because it could be more easily detected during most arthroscopic surgery and the positive or negative sign was given on the basis that the trajectory of PA is mostly known to be deviated to lateral side[15,23]. At 1 cm and 2 cm distal to the joint line, the shortest distance between the posterior cortex of tibia and the anterior border of the PA was measured (Figs. 3 and 4)[11,14]. The PA was not allocated in mediolateral direction due to the absence of relevant reference which is applicable to the actual operative field. In sagittal plane, the images that best showed the whole trajectory of PA was selected also in both the extended and 90 degree flexed knee. The closest distance between the posterior cortex of proximal tibia and the anterior wall of PA was measured regardless of the level (Fig. 5). In order to assess the reliability of the measurements, one orthopaedic surgeon and one orthopaedic resident performed the mea-
surements twice with an interval of 1 week in 10 knees randomly selected from the 30 knees. The degree of measurement reliabilities was assessed using the intraclass correlation coefficient (ICC). A typical interpretation of an ICC value is as follows: 0.00 to 0.20, poor agreement; 0.21 to 0.40, fair agreement; 0.41 to 0.60, moderate agreement; 0.61 to 0.80, substantial agreement; and 0.81 to 1.00, almost perfect agreement[24]. The ICCs for intra- and inter-rater agreement ranged from 0.81 to 0.86 for all measurements. These findings led us to rely on the validity of the measurements. As there were no significant differences among the measurements by two examiners, the measurement data by a single investigator were used in the following analyses.
2.4. Statistical analysis At the beginning of this study, the statistical power was assessed. Under the two-sided tail setting, the required sample size was 34, if the effect size, alpha error, and the power (1-beta) were 0.5, 0.05, and 0.8, respectively. On the other hand, with the alpha error 0.05, power (1-beta) 0.8, and the sample size 30 as is our study design, the effect size was calculated to be 0.53, which evidenced the relevance of the subject number of our study. All data were entered in a statistical program (SPSS, version 11.5, Chicago, Illinois, USA), and the results were given in mean, standard deviation (SD), and range at each level of measurement in both extension and 90 degree flexion of the knee. The 95% confidence interval (CI) was also presented. The effect of knee flexion was further
Fig. 4. The measurement at 2 cm distal to joint line (III inf Fig. 1). The distance between the posterior cortex of proximal tibia (1) and the anterior wall of popliteal artery (2) is measured on the MRIs of the knee in extension (a) and in 90 degree flexion (b). Note that the tip of the fibula is seen.
146
J.H. Yoo, C.B. Chang / The Knee 16 (2009) 143–148 Table 2 The location of popliteal artery at the joint line level, 1 cm and 2 cm distal to the joint line as indicated by mean (standard deviation, range) Plane
Dimension Extension (mm) 90 degree flexion (mm) p-value
Axial
2.4 (2.9, – 4–9.4) 3.2 (3.0, –5.5–10.3) 3.9 (1.5, 0.9–6.8) 7.6 (2.4, 3.7–14.2) 2.7 (1.1, 1.0–5.3) 7.2 (2.6, 0–11.4)
0.247 b 0.001 b 0.001
4.9 (2.6, 1.3–12.1) 9.7 (3.3, 3.2–16.2)
b 0.001
1.8 (1.8, 0–6.3)
b 0.001
Joint line level ML AP 1 cm distal to AP joint line 2 cm distal to AP joint line Sagittal the closest AP level
Fig. 5. The measurement in sagittal plane. The closest distance between the posterior cortex of proximal tibia and the anterior wall of popliteal artery (between the white arrows) was measured on the MRIs of the knee in extension (a) and in 90 degree flexion (b).
analyzed by paired t-test between the position of the PA in extension and flexion of the knee. The statistical significance was set at 0.05. 3. Results On axial plane, the PA was located lateral to the lateral border of PCL 2.4 mm [95% CI (1.3, 3.5)] in extension and 3.2 mm [95% CI (2.1, 4.3)] in 90 degree flexion at the joint line level, the difference of which was not statistically significance (p = 0.247). However, the PA was not entirely located lateral in all the subjects, but medial in four cases (13%) in extension, and in three (10%) in flexion. The PA moved laterally with knee flexion in 10 cases (33%) (Table 1). On axial, the PA lay 3.9 mm [95% CI (3.3, 4.4)] in extension and 7.6 mm [95% CI (6.7, 8.5)] in 90 degree flexion behind posterior capsule at the joint line level (p b 0.001). It is located 2.7 mm [95% CI (2.2, 3.1)] in extension and 7.2 mm [95% CI (6.3, 8.1)] in 90 degree flexion at 1 cm distal to joint line (p b 0.001), and 4.9 mm [95% CI (3.9, 5.8)] in extension and 9.7 mm [95% CI (8.5, 11.0)] in 90 degree flexion at 2 cm distal to joint line (p b 0.001) from the posterior cortex of proximal tibia. The intervening structures between them are loose connective tissues or the popliteus muscle and at 2 cm distal to joint line the popliteus muscle belly occupied most of the space between them. The paired t-test revealed the difference was statistically significant indicating the PA moves farther away from knee joint by flexion at each level except only one case in which the PA moved anteriorly 0.1 mm at 2 cm distal to joint line. In sagittal plane, the closest distance between posterior cortex of proximal tibia and the PA is 1.8 mm [95% CI (1.1, 2.5)] in extension and 6.2 mm [95% CI (5.4, 6.9)] in 90 degree flexion (p b 0.001) (Table 2). The PA moves farther away from the posterior aspect of the knee joint in all of the measured dimensions with statistical significance.
4. Discussion In the areas where much of the operations on the knee joint are performed between joint line and 2 cm distal to the joint line, the PA was located about 3 mm lateral to the PCL, and around 2–5 mm posterior to the posterior confinement of knee joint in extension and 6–10 mm in flexion, which almost doubled in extension. Most of the PA was located lateral to the lateral border of the PCL. The conventional wisdom of flexing the knee to avoid the injury of the PA was supported in the current study. The PA is at risk in many of modern orthopaedic surgeries[21,25]. The blind insertion of the needle tip loaded with small fixation device during the all-inside repair of the posterior horn of the lateral meniscus may jeopardize the PA[26]. Aneurysm[27] or pseudoaneurysm[5,13,28,29], occlusion[30], and arteriovenous fistula[29,31] after meniscectomy have been reported. It is also at stake during the establishment of tibial tunnel during the reconstruction of the PCL Table 1 The distribution of location of popliteal artery in mediolateral direction with reference to the lateral border of posterior cruciate ligament
Lateral Medial
Extension
90 degree flexion
87% (26/30) 13% (4/30)
90% (27/30) 10% (3/30)
6.2 (2.0, 1.5–11.6)
At the joint line level, it is allocated with reference to lateral border of posterior cruciate ligament mediolaterally, and to the posterior capsule anteroposteriorly, while the posterior cortex of proximal tibia serves as a reference in anteroposterior aspect at 1 cm and 2 cm distal to joint line, where no relevant anatomical reference existed. Note. ML: mediolateral, the negative value means that popliteal artery is located medial to the lateral border of the posterior cruciate ligament and the positive value lateral to it, AP: anteroposterior.
[16,23] and the anterior cruciate ligament[32,33]. The transplantation of the meniscus both in the creation of the bone trough[34] and in the meniscocapsular suture[35] poses the PA at stake. The institution of the posterior trans-septal portal also takes place just in front of the PA [36]. The surgeon performing the proximal tibial bone cutting in total knee arthroplasty or the high tibial osteotomy always feel uneasy of the potential injury of the PA. Acute arterial occlusion[6,37–40] and aneurysm[41] after TKA, direct injury to PA in HTO also have been reported[11]. Even though these are rare, it would lead to a disastrous outcome once it occurs, and these complications have always haunted the surgeons doing knee operations. However, they would feel much more secure with the understanding of the spatial relationship between the knee joint and the popliteal neurovascular bundles. Moreover, the practical guideline by metric dimension would be an additional benefit as is provided in our study. The PA traverses the popliteal fossa from the opening of the adductor magnus and then descends laterally to the intercondylar fossa, inclining obliquity to the distal border of the popliteus, where it divides into the anterior and posterior tibial arteries[42,43]. The anterior relations to the PA proximomedially are fat covering the femoral popliteal surface, then the capsule of the knee, and finally the fascia of the popliteus[18]. Utmost care should be taken behind the knee joint capsule. The PA is separated from the posterior cortex of proximal tibia by popliteus, and the lower half of the PA is held forward against the popliteus muscle by the geniculate vessels, the soleus, and the lateral head of gastrocnemius[11]. At the HTO level, the PA tended to be farther from the posterior tibial cortex than the other levels, due to the intervening bulk of the popliteus muscle. However, although most of surgical procedures are performed in flexed knee, the conventional anatomical description of the PA usually refers only to the extended knee[11]. It is generally believed that the popliteal artery is less likely to be injured when the knee is flexed to 90 degrees since this allows the artery to fall away from tibia[14,16,19,44,45]. Posterior movement of the PA occurs during flexion, with the creation of flexures behind the posterior articular structures of the knee[18,20,46]. The middle genicular artery is coiled in knee extension and progressively uncoils until it becomes straight during flexion of the knee, which explains the extent of movement[18]. Several methods of assessing the location of PA have been utilized including cadaveric dissection[16,21], angiography[12,17,18,23,46], ultrasonography[11,18], and MRI[14,15,18,25]. The stiff cadaveric knee with post-mortem changes is a poor substitute for in vivo investigation. The angiographic radiographs only provide 2-dimensional images difficult to determine spatial relationships. The reduced distance of the PA from the tibial surface in flexion of the knee in the study of Zaidi et al. was probably due to the excessive pressure placed upon the popliteal structures by the ultrasonographic probe[11,14].
J.H. Yoo, C.B. Chang / The Knee 16 (2009) 143–148
The MRI study of the living subject with supine position in obtaining 90 degree flexion images as is our study design best simulates the in vivo condition, which obviates the above foibles. Vernon et al., however, reported that the tibial section of PA varied very little with the flexion of the knee joint maintaining a constant curve closely applied to the tibia[20]. Zaidi et al. said that the PA was closer to the tibia in 90 degree of knee flexion than in full extension in more than half of the knee examined, or 12 in 20[11], which was later criticized in that they performed with the subjects in lateral position, negating the effect of gravity, and that the pressure of the ultrasound probe against the popliteal fossa may have prevented posterior vessel motion and spurious anterior displacement[14]. Ninomiya et al. illustrated the risk of popliteal artery injury by arteriography in 4 cadaveric knees and verified that it is located lateral to the midline in 95% of volunteer knee by MRI[17]. Smith et al. allocated the popliteal artery in 9 volunteer knees at 1 and 2 cm distal from the knee joint in extension and 90 degree flexion by MRI, which is similar to the current study, albeit with smaller study population and less comprehensive coverage of measurement field[14]. Matava et al. evaluated the proximity of the PCL insertion to the popliteal artery as a function of the knee flexion angle on 14 cadaveric knees by MRI and concluded that increasing knee flexion angle reduces the risk of arterial injury during arthroscopic PCL reconstruction[16]. Shiomi et al. demonstrated that flexion of the knee increases the distance between the PA and the proximal tibia in 15 volunteer knees by MRI study. But the MRI was acquired in lateral position, which might deviate the soft tissue anatomy around the knee joint[19]. Shetty et al. investigated the effect of the knee flexion on the popliteal artery by ultrasonography, cadaveric dissection, MRI and arteriography. However, they focused on the level of the bone cutting in case of knee arthroplasty or osteotomy and did not address the relationship at the level of the knee joint[18]. Shetty et al. also suggested that the safe zone for the bicortical screw purchase through the posterior tibial cortex in tibial tubercle transfer was the medial aspect of the proximal tibial metaphysis[12]. Although it proposed a useful practical guidance, the analysis was limited to coronal plane alone. Keser et al. allocated the popliteal artery at the level of the knee joint in 334 knees, but the axial MR images were obtained in knee extension, which is different from the actual knee position of 90 degrees of knee flexion in arthroscopy[15]. We believe that our study present the most comprehensive anatomical information in the living subject on the relationship between the knee joint and the popliteal artery with 3 levels of evaluation (joint line, 1 and 2 cm distal to joint line) encompassing most of the operative fields around the knee joint including arthroscopy, joint replacement arthroplasty and proximal tibial osteotomy. The supine position during MRI acquisition simulating gravity also renders it relevant. Our study has some limitations. First, the actual operative filed is not exactly the same to the models of our study. In arthroscopy, the knee joint is infused with fluid and swells with it. In TKA, the incised soft tissue during exposure might have an effect to the neurovascular bundles behind the knee joint[21], However, the experiment setting in our study is the best replica to the actual operative situation without invasive measures, we suppose. Second, all the included subjects were young male. It is, however, reported that the location of the popliteal artery was not significantly influenced by sex[15] or side[14]. Even though TKA or HTO operations are not usually done in young persons, our data could provide rough guide not to injure the popliteal artery in old patients. It would be helpful to remind that the popliteal artery has a tendency to assume a more tortuous course in the popliteal fossa with increasing age[23,47]. Moreover, the feasible popliteal cysts, secondary fibrosis and posterior osteophyte might distort the position of the popliteal artery in arthritic knees. In conclusion, the PA is located approximately 3 mm lateral to the PCL and within 5 mm in extension and within 10 mm in flexion behind knee joint. We have also shown that the flexion is a safer position for
147
the operations around the knee to avoid the injury to the popliteal artery. The surgeon, of course, should still exert extreme caution to avoid the injury to the popliteal vasculature.
References [1] Bernard M, Grothues-Spork M, Georgoulis A, Hertel P. Neural and vascular complications of arthroscopic meniscal surgery. Knee Surg Sports Traumatol Arthrosc 1994;2: 14–8. [2] DeLee JC. Complications of arthroscopy and arthroscopic surgery: results of a national survey. Committee on Complications of Arthroscopy Association of North America. Arthroscopy 1985;1:214–20. [3] Small NC. Complications in arthrosocpy: the knee and other joints. Arthroscopy 1986;2:253–8. [4] Small NC. Complications in arthroscopic meniscal surgery. Clin Sports Med 1990;9: 609–17. [5] Tawes Jr RL, Etheredge SN, Webb RL, Enloe LJ, Stallone RJ. Popliteal artery injury complicating arthroscopic menisectomy. Am J Surg 1988;156:136–8. [6] Bellemans J, Stockx L, Peerlinck K, Vermylen J, Lacroix H, Suy R. Arterial occlusion and thrombus aspiration after total knee arthroplasty. Clin Orthop Relat Res 1999: 164–8. [7] Calligaro KD, DeLaurentis DA, Booth RE, Rothman RH, Savarese RP, Dougherty MJ. Acute arterial thrombosis associated with total knee arthroplasty. J Vasc Surg 1994;20:927–30 discussion 30–32. [8] Rand JA. Vascular complications of total knee arthroplasty. Report of three cases. J Arthroplasty 1987;2:89–93. [9] Insall JN, Joseph DM, Msika C. High tibial osteotomy for varus gonarthrosis. A longterm follow-up study. J Bone Joint Surg Am 1984;66:1040–8. [10] Rubens F, Wellington JL, Bouchard AG. Popliteal artery injury after tibial osteotomy: report of two cases. Can J Surg 1990;33:294–7. [11] Zaidi SH, Cobb AG, Bentley G. Danger to the popliteal artery in high tibial osteotomy. J Bone Joint Surg Br 1995;77:384–6. [12] Shetty AA, Tindall AJ, Nickolaou N, James KD, Ignotus P. A safe zone for the passage of screws through the posterior tibial cortex in tibial tubercle transfer. The Knee 2005;12:99–101. [13] Jeffries JT, Gainor BJ, Allen WC, Cikrit D. Injury to the popliteal artery as a complication of arthroscopic surgery. A report of two cases. J Bone Joint Surg Am 1987;69: 783–5. [14] Smith PN, Gelinas J, Kennedy K, Thain L, Rorabeck CH, Bourne RB. Popliteal vessels in knee surgery. A magnetic resonance imaging study. Clin Orthop Relat Res 1999: 158–64. [15] Keser S, Savranlar A, Bayar A, Ulukent SC, Ozer T, Tuncay I. Anatomic localization of the popliteal artery at the level of the knee joint: a magnetic resonance imaging study. Arthroscopy 2006;22:656–9. [16] Matava MJ, Sethi NS, Totty WG. Proximity of the posterior cruciate ligament insertion to the popliteal artery as a function of the knee flexion angle: implications for posterior cruciate ligament reconstruction. Arthroscopy 2000;16:796–804. [17] Ninomiya JT, Dean JC, Goldberg VM. Injury to the popliteal artery and its anatomic location in total knee arthroplasty. J Arthroplasty 1999;14:803–9. [18] Shetty AA, Tindall AJ, Qureshi F, Divekar M, Fernando KW. The effect of knee flexion on the popliteal artery and its surgical significance. J Bone Joint Surg Br 2003;85:218–22. [19] Shiomi J, Takahashi T, Imazato S, Yamamoto H. Flexion of the knee increases the distance between the popliteal artery and the proximal tibia: MRI measurements in 15 volunteers. Acta Orthop Scand 2001;72:626–8. [20] Vernon P, Delattre JF, Johnson EJ, Palot JP, Clement C. Dynamic modifications of the popliteal arterial axis in the sagittal plane during flexion of the knee. Surg Radiol Anat 1987;9:37–41. [21] Cosgarea AJ, Kramer DE, Bahk MS, Totty WG, Matava MJ. Proximity of the popliteal artery to the PCL during simulated knee arthroscopy: implications for establishing the posterior trans-septal portal. J Knee Surg 2006;19:181–5. [22] Jackson DW, Proctor CS, Simon TM. Arthroscopic assisted PCL reconstruction: a technical note on potential neurovascular injury related to drill bit configuration. Arthroscopy 1993;9:224–7. [23] Cohen SB, Boyd L, Miller MD. Vascular risk associated with posterior cruciate ligament reconstruction using the arthroscopic transtibial tunnel technique. J Knee Surg 2004;17:211–3. [24] Bravo G, Potvin L. Estimating the reliability of continuous measures with Cronbach's alpha or the intraclass correlation coefficient: toward the integration of two traditions. J Clin Epidemiol 1991;44:381–90. [25] Kramer DE, Bahk MS, Cascio BM, Cosgarea AJ. Posterior knee arthroscopy: anatomy, technique, application. J Bone Joint Surg Am 2006;88(Suppl 4):110–21. [26] Miller MD, Hart JA. All-inside meniscal repair. Instr Course Lect 2005;54:337–40. [27] Kiss H, Drekonja T, Grethen C, Dorn U. Postoperative aneurysm of the popliteal artery after arthroscopic meniscectomy. Arthroscopy 2001;17:203–5. [28] Aldrich D, Anschuetz R, LoPresti C, Fumich M, Pitluk H, O'Brien W. Pseudoaneurysm complicating knee arthroscopy. Arthroscopy 1995;11:229–30. [29] Vassallo P, Reiser MF, Strobel M, Peters PE. Popliteal pseudoaneurysm and arteriovenous shunt following arthroscopic meniscectomy: case report. Cardiovasc Intervent Radiol 1989;12:142–4. [30] Potter D, Morris-Jones W. Popliteal artery injury complicating arthroscopic meniscectomy. Arthroscopy 1995;11:723–6. [31] Mullen DJ, Jabaji GJ. Popliteal pseudoaneurysm and arteriovenous fistula after arthroscopic meniscectomy. Arthroscopy 2001;17:E1.
148
J.H. Yoo, C.B. Chang / The Knee 16 (2009) 143–148
[32] Audenaert E, Vuylsteke M, Lissens P, Verhelst M, Verdonk R. Pseudoaneurysm complicating knee arthroscopy. A case report. Acta Orthop Belg 2003;69:382–4. [33] Janssen RP, Scheltinga MR, Sala HA. Pseudoaneurysm of the popliteal artery after anterior cruciate ligament reconstruction with bicortical tibial screw fixation. Arthroscopy 2004;20:E4–6. [34] Cole BJ, Carter TR, Rodeo SA. Allograft meniscal transplantation: background, techniques, and results. Instr Course Lect 2003;52:383–96. [35] Verdonk PC, Demurie A, Almqvist KF, Veys EM, Verbruggen G, Verdonk R. Transplantation of viable meniscal allograft. Surgical technique. J Bone Joint Surg Am 2006;88(Suppl 1 Pt 1):109–18. [36] Ahn JH, Ha CW. Posterior trans-septal portal for arthroscopic surgery of the knee joint. Arthroscopy 2000;16:774–9. [37] Holmberg A, Milbrink J, Bergqvist D. Arterial complications after knee arthroplasty: 4 cases and a review of the literature. Acta Orthop Scand 1996;67:75–8. [38] Kobayashi S, Isobe K, Koike T, Saitoh S, Takaoka K. Acute arterial occlusion associated with total knee arthroplasty. Arch Orthop Trauma Surg 1999;119:223–4. [39] Ohira T, Fujimoto T, Taniwaki K. Acute popliteal artery occlusion after total knee arthroplasty. Arch Orthop Trauma Surg 1997;116:429–30. [40] Parfenchuck TA, Young TR. Intraoperative arterial occlusion in total joint arthroplasty. J Arthroplasty 1994;9:217–20.
[41] Hozack WJ, Cole PA, Gardner R, Corces A. Popliteal aneurysm after total knee arthroplasty. Case reports and review of the literature. J Arthroplasty 1990;5:301–5. [42] Colborn GL, Lumsden AB, Taylor BS, Skandalakis JE. The surgical anatomy of the popliteal artery. Am Surg 1994;60:238–46. [43] Kim TK, Savino RM, McFarland EG, Cosgarea AJ. Neurovascular complications of knee arthroscopy. Am J Sports Med 2002;30:619–29. [44] Adamson GJ, Wiss DA, Lowery GL, Peters CL. Type II floating knee: ipsilateral femoral and tibial fractures with intraarticular extension into the knee joint. J Orthop Trauma 1992;6:333–9. [45] Coventry MB. Osteotomy about the knee for degenerative and rheumatoid arthritis. J Bone Joint Surg Am 1973;55:23–48. [46] Avisse C, Marcus C, Ouedraogo T, Delattre JF, Menanteau B, Flament JB. Anatomoradiological study of the popliteal artery during knee flexion. Surg Radiol Anat 1995;17:255–62. [47] Wensing PJ, Scholten FG, Buijs PC, Hartkamp MJ, Mali WP, Hillen B. Arterial tortuosity in the femoropopliteal region during knee flexion: a magnetic resonance angiographic study. J Anat 1995;187(Pt 1):133–9.
Contents lists available at ScienceDirect
The Knee
The location of the popliteal artery in extension and 90 degree knee flexion measured on MRI Jae Ho Yoo a,1, Chong Bum Chang b,⁎ a b
Department of Orthopaedic Surgery, Soonchunhyang University Hospital, Bucheon, 1174 Jung-Dong, Wonmi-Gu, Bucheon-Si, Gyeonggi-Do, 420-767, South Korea Joint Reconstruction Center, Department of Orthopaedic Surgery, Seoul National University Bundang Hospital, 300 Gumidong, Bundangu, Seongnamsi, Gyunggido (463-707), Korea
a r t i c l e
i n f o
Article history: Received 5 May 2008 Received in revised form 15 October 2008 Accepted 26 October 2008 Keywords: Knee Popliteal artery MRI
a b s t r a c t We measured the location of the popliteal artery (PA) in extension and 90 degree of knee flexion by magnetic resonance images (MRI) to provide practical information to avoid PA injury. The MRIs of 30 knees of Korean male subject whose mean age was 20.7 were acquired in knee extension and 90 degree flexion. The distance from the posterior aspect of knee joint to the PA was measured at three levels on the axial images and one sagittal image. At the joint line level, the PA was located lateral to the PCL 2.4 mm in extension and 3.2 mm in flexion (p = 0.247), and 3.9 mm in extension and 7.6 mm in flexion from the posterior capsule (p b 0.001). At 1 cm distal to the joint line, it is 2.7 mm in extension and 7.2 mm in flexion (p b 0.001), and at 2 cm distal to the joint line, 4.9 mm in extension and 9.7 mm in flexion from the posterior tibial cortex (p b 0.001). In sagittal plane, the nearest distance between PA and posterior tibial cortex was 1.8 mm in extension, and 6.2 mm in flexion (p b 0.001). The PA was located around 3 mm lateral to the PCL, and within 5 mm in extension and 10 mm in 90 degree flexion of the knee behind knee joint. It moves farther posteriorly in 90 degree flexion than in extension of the knee. The conventional wisdom of flexing the knee to prevent the PA injury was supported by this study. © 2008 Elsevier B.V. All rights reserved.
1. Introduction Although the injury to the popliteal artery (PA) during the surgeries around the knee joint has been regarded as a rare complication, the incidence has been reported to be as high as 1.8%[1–4]. Moreover, once it happens, the consequences are disastrous. The arthroscopy, the total knee arthroplasty (TKA), and the high tibial osteotomy are most commonly conducted operations, which also might jeopardize the PA. Tawes et al. reported that one of five arterial injuries during arthroscopic meniscectomy led to above knee amputation[5]. Arterial compromise after total knee arthroplasty has been reported to be between 0.03 and 0.2%, and resulted in amputation of the limb in up to 25% of the cases[6– 8]. There are also a few reports of direct injury to the PA during high tibial osteotomy (HTO)[9–11]. The course of the popliteal artery is also a concern in bicortical screw fixation of the tibial tubercle transfer[12]. With all the more increasing surgeries in the posterior aspect of the knee joint such as meniscal repair or transplantation, and the reconstruction of posterior cruciate ligament, and with the incremental incidence of total knee arthroplasty, the chance of injury rises accordingly. In addition, the current interest and application of the minimally invasive technique in TKA also could increase the possibility of the PA injury during tibial bone resection, all the more with limited surgical exposure.
The fact that injury does occur is not surprising as the vessel is in the worst position, within a few millimeters from the posterior aspect of the knee joint with only a small amount of fat being the only intervening tissue[13,14]. Although the location of the PA may change in different situations and vary among individuals, approximate allocation of the PA would obviously make the surgeon feel more secure. The conventional wisdom has it that flexing the knee joint puts the PA away from the posterior aspect of the knee joint when doing procedures around the posterior aspect of the knee, and this surgical aphorism has been supported by previous studies by MRI, angiography, or cadaveric dissection[12,14–20]. However, a study opposing to the conventional wisdom also exists that the PA paradoxically moved closer to the tibia by flexing the knee in 12 out of 20 knees[11]. The purpose of this study is to investigate the location of the PA in extension and 90 degrees of flexion at different levels of the knee joint to provide practical information to avoid popliteal artery injury. The conventional wisdom of flexing the knee so that the neurovascular bundles moves farther from the posterior aspect of the knee to prevent popliteal artery injury is also audited. 2. Materials and method 2.1. Subjects
⁎ Corresponding author. Tel.: +82 31 787 7201; fax: +82 31 787 4056. E-mail addresses: [email protected], [email protected] (J.H. Yoo), [email protected] (C.B. Chang). 1 Tel.: +82 32 621 5263; fax: +82 32 621 5016, Mobile: +82 11 671 2096. 0968-0160/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.knee.2008.10.009
The magnetic resonance images (MRI) of the 33 knees in 33 healthy Korean volunteers were obtained for the study. The inclusion criteria used for the volunteers were; (1) absence of knee pain, (2) no
144
J.H. Yoo, C.B. Chang / The Knee 16 (2009) 143–148
Fig. 1. The measurement was performed at three different levels of axial images; joint line level (I), 1 cm distal to joint level (II), 2 cm distal to joint level (III) in MRI of the knee in extension (a) and in 90 degree flexion (b).
history of knee injury, (3) normal range of lower limb alignment and no obvious knee deformity, (4) no instability of the knee joint, (5) none of other abnormal physical findings such as Baker's cyst. Three subjects were excluded from the study because of the inadequate quality of the images for exact measurement. Consequently, a total of 30 knee MRI scans from 30 male subjects were included, none of whom had bilateral scans. There were 15 right and 15 left knees. The mean age was 20.7 years ranging from 18 to 25. The mean weight, height, and body mass index (BMI) were, 66.1 kg (SD 10.8, range 51–80), 176.2 cm (SD 5.6, range 165–186), 21.8 kg/m2 (SD 2.0, range 16.7–27.3), respectively. 2.2. MRI acquisition All the MR images were obtained with the knee in full extension and 90 degrees of flexion using 1.5 Tesla scanner (Intera, Philips Medial Systems, The Netherlands). The axial and sagittal plane images were obtained using the following protocol; proton density-weighted images with fat suppression (repetition time 2378 ms and echo time 50 ms). Each section was 5 mm in thickness with a 0.5 mm intervening spacing and has a 512 × 512 matrix. The full extension of the knee was maintained in the standard surface coil with a corresponding height of pillow under the ankle to prevent the flexion of the knee. The 90 degree of the knee flexion simulated the surgical situations during various knee operations such as arthroscopy, arthroplasty, or osteotomy. To assume the actual position during operations, the subject lay on his back
allowing the neurovascular bundles behind the knee joint dangling down according to the gravity. 2.3. Measurement All MR images were digitally acquired through a picture-archiving and communication system (PACS), and the assessments were performed subsequently using PACS software (PiViewSTAR, INFINITT Ltd, Seoul, Korea) on a desk-top computer. We distinguished the arterial vasculature from the veins according to signal void as a result of faster flow velocity on the images[15]. The topographical relationship of neurovascular bundles including artery, vein, and nerve was also taken into consideration in that the popliteal artery is usually located in the most anterior side, followed by popliteal vein, and tibial nerve[16,21,22]. No anomalous duplication of the neurovascular bundle was detected. The measurement was conducted on the best images that show the concerning structures described below. In axial plane, the measurements were conducted in three different levels in both the extended and 90 degree flexed knee; the joint line, 1 cm distal to the joint line, and 2 cm distal to the joint line (Fig. 1). The orientation of the cutting slices was referenced by the joint line to standardize the measurement level. The aim of the measurements in the three different levels was to represent the procedures potentially involving knee surgeries. Most arthroscopic procedures in the knee joint are performed around the level of the joint line. The tibial bone cutting in knee arthroplasty or tibial tunnel creation in the reconstruction of the
Fig. 2. The measurement at joint line level (I in Fig. 1). The location of the PA is described by the distance between the posterior capsule (1) and the anterior wall of PA (2) in anteroposterior dimension, and that between the lateral border of posterior cruciate ligament (3) and medial wall of popliteal artery (4) in mediolateral dimension, on the MRIs of the knee in extension (a) and in 90 degree flexion (b).
J.H. Yoo, C.B. Chang / The Knee 16 (2009) 143–148
145
Fig. 3. The measurement at 1 cm distal to joint line (II in Fig. 1). The distance between the posterior cortex of proximal tibia (1) and the anterior wall of popliteal artery (2) is measured on the MRIs of the knee in extension (a) and in 90 degree flexion (b).
posterior cruciate ligament (PCL) would come around 1 cm distal to joint line. And the proximal tibial osteotomy in HTO takes place about 2 cm distal to the joint line. At each level we measured the distance from the germane anatomical landmarks to the PA. At the joint line level, the optimal image showing the meniscus or the tibial plateau was selected. Then, the distance between the posterior joint capsule and the anterior border of the PA was measured, by which we can assume the approximate distance to PA from the posterior capsule of the knee joint in arthroscopic surgery. In addition, the mediolateral location of PA with reference to the lateral border of PCL was also measured giving positive value if the PA lay lateral to the PCL and negative one if medial (Fig. 2). The lateral border of the PCL was selected because it could be more easily detected during most arthroscopic surgery and the positive or negative sign was given on the basis that the trajectory of PA is mostly known to be deviated to lateral side[15,23]. At 1 cm and 2 cm distal to the joint line, the shortest distance between the posterior cortex of tibia and the anterior border of the PA was measured (Figs. 3 and 4)[11,14]. The PA was not allocated in mediolateral direction due to the absence of relevant reference which is applicable to the actual operative field. In sagittal plane, the images that best showed the whole trajectory of PA was selected also in both the extended and 90 degree flexed knee. The closest distance between the posterior cortex of proximal tibia and the anterior wall of PA was measured regardless of the level (Fig. 5). In order to assess the reliability of the measurements, one orthopaedic surgeon and one orthopaedic resident performed the mea-
surements twice with an interval of 1 week in 10 knees randomly selected from the 30 knees. The degree of measurement reliabilities was assessed using the intraclass correlation coefficient (ICC). A typical interpretation of an ICC value is as follows: 0.00 to 0.20, poor agreement; 0.21 to 0.40, fair agreement; 0.41 to 0.60, moderate agreement; 0.61 to 0.80, substantial agreement; and 0.81 to 1.00, almost perfect agreement[24]. The ICCs for intra- and inter-rater agreement ranged from 0.81 to 0.86 for all measurements. These findings led us to rely on the validity of the measurements. As there were no significant differences among the measurements by two examiners, the measurement data by a single investigator were used in the following analyses.
2.4. Statistical analysis At the beginning of this study, the statistical power was assessed. Under the two-sided tail setting, the required sample size was 34, if the effect size, alpha error, and the power (1-beta) were 0.5, 0.05, and 0.8, respectively. On the other hand, with the alpha error 0.05, power (1-beta) 0.8, and the sample size 30 as is our study design, the effect size was calculated to be 0.53, which evidenced the relevance of the subject number of our study. All data were entered in a statistical program (SPSS, version 11.5, Chicago, Illinois, USA), and the results were given in mean, standard deviation (SD), and range at each level of measurement in both extension and 90 degree flexion of the knee. The 95% confidence interval (CI) was also presented. The effect of knee flexion was further
Fig. 4. The measurement at 2 cm distal to joint line (III inf Fig. 1). The distance between the posterior cortex of proximal tibia (1) and the anterior wall of popliteal artery (2) is measured on the MRIs of the knee in extension (a) and in 90 degree flexion (b). Note that the tip of the fibula is seen.
146
J.H. Yoo, C.B. Chang / The Knee 16 (2009) 143–148 Table 2 The location of popliteal artery at the joint line level, 1 cm and 2 cm distal to the joint line as indicated by mean (standard deviation, range) Plane
Dimension Extension (mm) 90 degree flexion (mm) p-value
Axial
2.4 (2.9, – 4–9.4) 3.2 (3.0, –5.5–10.3) 3.9 (1.5, 0.9–6.8) 7.6 (2.4, 3.7–14.2) 2.7 (1.1, 1.0–5.3) 7.2 (2.6, 0–11.4)
0.247 b 0.001 b 0.001
4.9 (2.6, 1.3–12.1) 9.7 (3.3, 3.2–16.2)
b 0.001
1.8 (1.8, 0–6.3)
b 0.001
Joint line level ML AP 1 cm distal to AP joint line 2 cm distal to AP joint line Sagittal the closest AP level
Fig. 5. The measurement in sagittal plane. The closest distance between the posterior cortex of proximal tibia and the anterior wall of popliteal artery (between the white arrows) was measured on the MRIs of the knee in extension (a) and in 90 degree flexion (b).
analyzed by paired t-test between the position of the PA in extension and flexion of the knee. The statistical significance was set at 0.05. 3. Results On axial plane, the PA was located lateral to the lateral border of PCL 2.4 mm [95% CI (1.3, 3.5)] in extension and 3.2 mm [95% CI (2.1, 4.3)] in 90 degree flexion at the joint line level, the difference of which was not statistically significance (p = 0.247). However, the PA was not entirely located lateral in all the subjects, but medial in four cases (13%) in extension, and in three (10%) in flexion. The PA moved laterally with knee flexion in 10 cases (33%) (Table 1). On axial, the PA lay 3.9 mm [95% CI (3.3, 4.4)] in extension and 7.6 mm [95% CI (6.7, 8.5)] in 90 degree flexion behind posterior capsule at the joint line level (p b 0.001). It is located 2.7 mm [95% CI (2.2, 3.1)] in extension and 7.2 mm [95% CI (6.3, 8.1)] in 90 degree flexion at 1 cm distal to joint line (p b 0.001), and 4.9 mm [95% CI (3.9, 5.8)] in extension and 9.7 mm [95% CI (8.5, 11.0)] in 90 degree flexion at 2 cm distal to joint line (p b 0.001) from the posterior cortex of proximal tibia. The intervening structures between them are loose connective tissues or the popliteus muscle and at 2 cm distal to joint line the popliteus muscle belly occupied most of the space between them. The paired t-test revealed the difference was statistically significant indicating the PA moves farther away from knee joint by flexion at each level except only one case in which the PA moved anteriorly 0.1 mm at 2 cm distal to joint line. In sagittal plane, the closest distance between posterior cortex of proximal tibia and the PA is 1.8 mm [95% CI (1.1, 2.5)] in extension and 6.2 mm [95% CI (5.4, 6.9)] in 90 degree flexion (p b 0.001) (Table 2). The PA moves farther away from the posterior aspect of the knee joint in all of the measured dimensions with statistical significance.
4. Discussion In the areas where much of the operations on the knee joint are performed between joint line and 2 cm distal to the joint line, the PA was located about 3 mm lateral to the PCL, and around 2–5 mm posterior to the posterior confinement of knee joint in extension and 6–10 mm in flexion, which almost doubled in extension. Most of the PA was located lateral to the lateral border of the PCL. The conventional wisdom of flexing the knee to avoid the injury of the PA was supported in the current study. The PA is at risk in many of modern orthopaedic surgeries[21,25]. The blind insertion of the needle tip loaded with small fixation device during the all-inside repair of the posterior horn of the lateral meniscus may jeopardize the PA[26]. Aneurysm[27] or pseudoaneurysm[5,13,28,29], occlusion[30], and arteriovenous fistula[29,31] after meniscectomy have been reported. It is also at stake during the establishment of tibial tunnel during the reconstruction of the PCL Table 1 The distribution of location of popliteal artery in mediolateral direction with reference to the lateral border of posterior cruciate ligament
Lateral Medial
Extension
90 degree flexion
87% (26/30) 13% (4/30)
90% (27/30) 10% (3/30)
6.2 (2.0, 1.5–11.6)
At the joint line level, it is allocated with reference to lateral border of posterior cruciate ligament mediolaterally, and to the posterior capsule anteroposteriorly, while the posterior cortex of proximal tibia serves as a reference in anteroposterior aspect at 1 cm and 2 cm distal to joint line, where no relevant anatomical reference existed. Note. ML: mediolateral, the negative value means that popliteal artery is located medial to the lateral border of the posterior cruciate ligament and the positive value lateral to it, AP: anteroposterior.
[16,23] and the anterior cruciate ligament[32,33]. The transplantation of the meniscus both in the creation of the bone trough[34] and in the meniscocapsular suture[35] poses the PA at stake. The institution of the posterior trans-septal portal also takes place just in front of the PA [36]. The surgeon performing the proximal tibial bone cutting in total knee arthroplasty or the high tibial osteotomy always feel uneasy of the potential injury of the PA. Acute arterial occlusion[6,37–40] and aneurysm[41] after TKA, direct injury to PA in HTO also have been reported[11]. Even though these are rare, it would lead to a disastrous outcome once it occurs, and these complications have always haunted the surgeons doing knee operations. However, they would feel much more secure with the understanding of the spatial relationship between the knee joint and the popliteal neurovascular bundles. Moreover, the practical guideline by metric dimension would be an additional benefit as is provided in our study. The PA traverses the popliteal fossa from the opening of the adductor magnus and then descends laterally to the intercondylar fossa, inclining obliquity to the distal border of the popliteus, where it divides into the anterior and posterior tibial arteries[42,43]. The anterior relations to the PA proximomedially are fat covering the femoral popliteal surface, then the capsule of the knee, and finally the fascia of the popliteus[18]. Utmost care should be taken behind the knee joint capsule. The PA is separated from the posterior cortex of proximal tibia by popliteus, and the lower half of the PA is held forward against the popliteus muscle by the geniculate vessels, the soleus, and the lateral head of gastrocnemius[11]. At the HTO level, the PA tended to be farther from the posterior tibial cortex than the other levels, due to the intervening bulk of the popliteus muscle. However, although most of surgical procedures are performed in flexed knee, the conventional anatomical description of the PA usually refers only to the extended knee[11]. It is generally believed that the popliteal artery is less likely to be injured when the knee is flexed to 90 degrees since this allows the artery to fall away from tibia[14,16,19,44,45]. Posterior movement of the PA occurs during flexion, with the creation of flexures behind the posterior articular structures of the knee[18,20,46]. The middle genicular artery is coiled in knee extension and progressively uncoils until it becomes straight during flexion of the knee, which explains the extent of movement[18]. Several methods of assessing the location of PA have been utilized including cadaveric dissection[16,21], angiography[12,17,18,23,46], ultrasonography[11,18], and MRI[14,15,18,25]. The stiff cadaveric knee with post-mortem changes is a poor substitute for in vivo investigation. The angiographic radiographs only provide 2-dimensional images difficult to determine spatial relationships. The reduced distance of the PA from the tibial surface in flexion of the knee in the study of Zaidi et al. was probably due to the excessive pressure placed upon the popliteal structures by the ultrasonographic probe[11,14].
J.H. Yoo, C.B. Chang / The Knee 16 (2009) 143–148
The MRI study of the living subject with supine position in obtaining 90 degree flexion images as is our study design best simulates the in vivo condition, which obviates the above foibles. Vernon et al., however, reported that the tibial section of PA varied very little with the flexion of the knee joint maintaining a constant curve closely applied to the tibia[20]. Zaidi et al. said that the PA was closer to the tibia in 90 degree of knee flexion than in full extension in more than half of the knee examined, or 12 in 20[11], which was later criticized in that they performed with the subjects in lateral position, negating the effect of gravity, and that the pressure of the ultrasound probe against the popliteal fossa may have prevented posterior vessel motion and spurious anterior displacement[14]. Ninomiya et al. illustrated the risk of popliteal artery injury by arteriography in 4 cadaveric knees and verified that it is located lateral to the midline in 95% of volunteer knee by MRI[17]. Smith et al. allocated the popliteal artery in 9 volunteer knees at 1 and 2 cm distal from the knee joint in extension and 90 degree flexion by MRI, which is similar to the current study, albeit with smaller study population and less comprehensive coverage of measurement field[14]. Matava et al. evaluated the proximity of the PCL insertion to the popliteal artery as a function of the knee flexion angle on 14 cadaveric knees by MRI and concluded that increasing knee flexion angle reduces the risk of arterial injury during arthroscopic PCL reconstruction[16]. Shiomi et al. demonstrated that flexion of the knee increases the distance between the PA and the proximal tibia in 15 volunteer knees by MRI study. But the MRI was acquired in lateral position, which might deviate the soft tissue anatomy around the knee joint[19]. Shetty et al. investigated the effect of the knee flexion on the popliteal artery by ultrasonography, cadaveric dissection, MRI and arteriography. However, they focused on the level of the bone cutting in case of knee arthroplasty or osteotomy and did not address the relationship at the level of the knee joint[18]. Shetty et al. also suggested that the safe zone for the bicortical screw purchase through the posterior tibial cortex in tibial tubercle transfer was the medial aspect of the proximal tibial metaphysis[12]. Although it proposed a useful practical guidance, the analysis was limited to coronal plane alone. Keser et al. allocated the popliteal artery at the level of the knee joint in 334 knees, but the axial MR images were obtained in knee extension, which is different from the actual knee position of 90 degrees of knee flexion in arthroscopy[15]. We believe that our study present the most comprehensive anatomical information in the living subject on the relationship between the knee joint and the popliteal artery with 3 levels of evaluation (joint line, 1 and 2 cm distal to joint line) encompassing most of the operative fields around the knee joint including arthroscopy, joint replacement arthroplasty and proximal tibial osteotomy. The supine position during MRI acquisition simulating gravity also renders it relevant. Our study has some limitations. First, the actual operative filed is not exactly the same to the models of our study. In arthroscopy, the knee joint is infused with fluid and swells with it. In TKA, the incised soft tissue during exposure might have an effect to the neurovascular bundles behind the knee joint[21], However, the experiment setting in our study is the best replica to the actual operative situation without invasive measures, we suppose. Second, all the included subjects were young male. It is, however, reported that the location of the popliteal artery was not significantly influenced by sex[15] or side[14]. Even though TKA or HTO operations are not usually done in young persons, our data could provide rough guide not to injure the popliteal artery in old patients. It would be helpful to remind that the popliteal artery has a tendency to assume a more tortuous course in the popliteal fossa with increasing age[23,47]. Moreover, the feasible popliteal cysts, secondary fibrosis and posterior osteophyte might distort the position of the popliteal artery in arthritic knees. In conclusion, the PA is located approximately 3 mm lateral to the PCL and within 5 mm in extension and within 10 mm in flexion behind knee joint. We have also shown that the flexion is a safer position for
147
the operations around the knee to avoid the injury to the popliteal artery. The surgeon, of course, should still exert extreme caution to avoid the injury to the popliteal vasculature.
References [1] Bernard M, Grothues-Spork M, Georgoulis A, Hertel P. Neural and vascular complications of arthroscopic meniscal surgery. Knee Surg Sports Traumatol Arthrosc 1994;2: 14–8. [2] DeLee JC. Complications of arthroscopy and arthroscopic surgery: results of a national survey. Committee on Complications of Arthroscopy Association of North America. Arthroscopy 1985;1:214–20. [3] Small NC. Complications in arthrosocpy: the knee and other joints. Arthroscopy 1986;2:253–8. [4] Small NC. Complications in arthroscopic meniscal surgery. Clin Sports Med 1990;9: 609–17. [5] Tawes Jr RL, Etheredge SN, Webb RL, Enloe LJ, Stallone RJ. Popliteal artery injury complicating arthroscopic menisectomy. Am J Surg 1988;156:136–8. [6] Bellemans J, Stockx L, Peerlinck K, Vermylen J, Lacroix H, Suy R. Arterial occlusion and thrombus aspiration after total knee arthroplasty. Clin Orthop Relat Res 1999: 164–8. [7] Calligaro KD, DeLaurentis DA, Booth RE, Rothman RH, Savarese RP, Dougherty MJ. Acute arterial thrombosis associated with total knee arthroplasty. J Vasc Surg 1994;20:927–30 discussion 30–32. [8] Rand JA. Vascular complications of total knee arthroplasty. Report of three cases. J Arthroplasty 1987;2:89–93. [9] Insall JN, Joseph DM, Msika C. High tibial osteotomy for varus gonarthrosis. A longterm follow-up study. J Bone Joint Surg Am 1984;66:1040–8. [10] Rubens F, Wellington JL, Bouchard AG. Popliteal artery injury after tibial osteotomy: report of two cases. Can J Surg 1990;33:294–7. [11] Zaidi SH, Cobb AG, Bentley G. Danger to the popliteal artery in high tibial osteotomy. J Bone Joint Surg Br 1995;77:384–6. [12] Shetty AA, Tindall AJ, Nickolaou N, James KD, Ignotus P. A safe zone for the passage of screws through the posterior tibial cortex in tibial tubercle transfer. The Knee 2005;12:99–101. [13] Jeffries JT, Gainor BJ, Allen WC, Cikrit D. Injury to the popliteal artery as a complication of arthroscopic surgery. A report of two cases. J Bone Joint Surg Am 1987;69: 783–5. [14] Smith PN, Gelinas J, Kennedy K, Thain L, Rorabeck CH, Bourne RB. Popliteal vessels in knee surgery. A magnetic resonance imaging study. Clin Orthop Relat Res 1999: 158–64. [15] Keser S, Savranlar A, Bayar A, Ulukent SC, Ozer T, Tuncay I. Anatomic localization of the popliteal artery at the level of the knee joint: a magnetic resonance imaging study. Arthroscopy 2006;22:656–9. [16] Matava MJ, Sethi NS, Totty WG. Proximity of the posterior cruciate ligament insertion to the popliteal artery as a function of the knee flexion angle: implications for posterior cruciate ligament reconstruction. Arthroscopy 2000;16:796–804. [17] Ninomiya JT, Dean JC, Goldberg VM. Injury to the popliteal artery and its anatomic location in total knee arthroplasty. J Arthroplasty 1999;14:803–9. [18] Shetty AA, Tindall AJ, Qureshi F, Divekar M, Fernando KW. The effect of knee flexion on the popliteal artery and its surgical significance. J Bone Joint Surg Br 2003;85:218–22. [19] Shiomi J, Takahashi T, Imazato S, Yamamoto H. Flexion of the knee increases the distance between the popliteal artery and the proximal tibia: MRI measurements in 15 volunteers. Acta Orthop Scand 2001;72:626–8. [20] Vernon P, Delattre JF, Johnson EJ, Palot JP, Clement C. Dynamic modifications of the popliteal arterial axis in the sagittal plane during flexion of the knee. Surg Radiol Anat 1987;9:37–41. [21] Cosgarea AJ, Kramer DE, Bahk MS, Totty WG, Matava MJ. Proximity of the popliteal artery to the PCL during simulated knee arthroscopy: implications for establishing the posterior trans-septal portal. J Knee Surg 2006;19:181–5. [22] Jackson DW, Proctor CS, Simon TM. Arthroscopic assisted PCL reconstruction: a technical note on potential neurovascular injury related to drill bit configuration. Arthroscopy 1993;9:224–7. [23] Cohen SB, Boyd L, Miller MD. Vascular risk associated with posterior cruciate ligament reconstruction using the arthroscopic transtibial tunnel technique. J Knee Surg 2004;17:211–3. [24] Bravo G, Potvin L. Estimating the reliability of continuous measures with Cronbach's alpha or the intraclass correlation coefficient: toward the integration of two traditions. J Clin Epidemiol 1991;44:381–90. [25] Kramer DE, Bahk MS, Cascio BM, Cosgarea AJ. Posterior knee arthroscopy: anatomy, technique, application. J Bone Joint Surg Am 2006;88(Suppl 4):110–21. [26] Miller MD, Hart JA. All-inside meniscal repair. Instr Course Lect 2005;54:337–40. [27] Kiss H, Drekonja T, Grethen C, Dorn U. Postoperative aneurysm of the popliteal artery after arthroscopic meniscectomy. Arthroscopy 2001;17:203–5. [28] Aldrich D, Anschuetz R, LoPresti C, Fumich M, Pitluk H, O'Brien W. Pseudoaneurysm complicating knee arthroscopy. Arthroscopy 1995;11:229–30. [29] Vassallo P, Reiser MF, Strobel M, Peters PE. Popliteal pseudoaneurysm and arteriovenous shunt following arthroscopic meniscectomy: case report. Cardiovasc Intervent Radiol 1989;12:142–4. [30] Potter D, Morris-Jones W. Popliteal artery injury complicating arthroscopic meniscectomy. Arthroscopy 1995;11:723–6. [31] Mullen DJ, Jabaji GJ. Popliteal pseudoaneurysm and arteriovenous fistula after arthroscopic meniscectomy. Arthroscopy 2001;17:E1.
148
J.H. Yoo, C.B. Chang / The Knee 16 (2009) 143–148
[32] Audenaert E, Vuylsteke M, Lissens P, Verhelst M, Verdonk R. Pseudoaneurysm complicating knee arthroscopy. A case report. Acta Orthop Belg 2003;69:382–4. [33] Janssen RP, Scheltinga MR, Sala HA. Pseudoaneurysm of the popliteal artery after anterior cruciate ligament reconstruction with bicortical tibial screw fixation. Arthroscopy 2004;20:E4–6. [34] Cole BJ, Carter TR, Rodeo SA. Allograft meniscal transplantation: background, techniques, and results. Instr Course Lect 2003;52:383–96. [35] Verdonk PC, Demurie A, Almqvist KF, Veys EM, Verbruggen G, Verdonk R. Transplantation of viable meniscal allograft. Surgical technique. J Bone Joint Surg Am 2006;88(Suppl 1 Pt 1):109–18. [36] Ahn JH, Ha CW. Posterior trans-septal portal for arthroscopic surgery of the knee joint. Arthroscopy 2000;16:774–9. [37] Holmberg A, Milbrink J, Bergqvist D. Arterial complications after knee arthroplasty: 4 cases and a review of the literature. Acta Orthop Scand 1996;67:75–8. [38] Kobayashi S, Isobe K, Koike T, Saitoh S, Takaoka K. Acute arterial occlusion associated with total knee arthroplasty. Arch Orthop Trauma Surg 1999;119:223–4. [39] Ohira T, Fujimoto T, Taniwaki K. Acute popliteal artery occlusion after total knee arthroplasty. Arch Orthop Trauma Surg 1997;116:429–30. [40] Parfenchuck TA, Young TR. Intraoperative arterial occlusion in total joint arthroplasty. J Arthroplasty 1994;9:217–20.
[41] Hozack WJ, Cole PA, Gardner R, Corces A. Popliteal aneurysm after total knee arthroplasty. Case reports and review of the literature. J Arthroplasty 1990;5:301–5. [42] Colborn GL, Lumsden AB, Taylor BS, Skandalakis JE. The surgical anatomy of the popliteal artery. Am Surg 1994;60:238–46. [43] Kim TK, Savino RM, McFarland EG, Cosgarea AJ. Neurovascular complications of knee arthroscopy. Am J Sports Med 2002;30:619–29. [44] Adamson GJ, Wiss DA, Lowery GL, Peters CL. Type II floating knee: ipsilateral femoral and tibial fractures with intraarticular extension into the knee joint. J Orthop Trauma 1992;6:333–9. [45] Coventry MB. Osteotomy about the knee for degenerative and rheumatoid arthritis. J Bone Joint Surg Am 1973;55:23–48. [46] Avisse C, Marcus C, Ouedraogo T, Delattre JF, Menanteau B, Flament JB. Anatomoradiological study of the popliteal artery during knee flexion. Surg Radiol Anat 1995;17:255–62. [47] Wensing PJ, Scholten FG, Buijs PC, Hartkamp MJ, Mali WP, Hillen B. Arterial tortuosity in the femoropopliteal region during knee flexion: a magnetic resonance angiographic study. J Anat 1995;187(Pt 1):133–9.