In total knee arthroplasty, which is a common orthopaedic procedure, the posterior aspects of the implants are in close proximity to the popliteal artery. Popliteal artery injury has been estimated to occur in 0.03% to 0.17% of total knee arthroplasty procedures1,2. Direct injury of the popliteal artery, although infrequent, can have catastrophic consequences for the limb3,4 and occasionally requires further surgery, including revascularization or even amputation5.
Several authors have identified the location of the popliteal artery in osteoarthritic knees or cadaver knees with use of magnetic resonance imaging (MRI), arteriography, or ultrasonography6-12. Vernon et al. investigated changes in the position of the popliteal artery in the sagittal plane during flexion of the knee in cadavers with use of arteriography10. Smith et al. investigated changes in the position of the popliteal vascular bundle (the popliteal artery and associated veins) during knee flexion in vivo with use of MRI9. Recently, Shetty et al. investigated the position of the popliteal artery relative to the posterior aspect of the tibia at 0° and 90° of flexion in vivo in 100 knees with use of duplex ultrasonography11. Most authors have concluded that the popliteal artery is less likely to sustain an iatrogenic injury during knee arthroplasty if the knee is in 90° of flexion to allow the artery to move further back from the tibia6,7,13,14. Likewise, most total knee arthroplasties are performed with the knee flexed 90° to reduce the risk of popliteal artery injury.
This safe position has been challenged by the authors of another in vivo study involving ultrasonography11,15. The results of that study conflicted with the hypothesis that 90° of flexion represents the safe zone. Furthermore, Farrington et al. reported that the position of the safe zone for the popliteal artery after knee arthroplasty in a series of seventeen patients (twenty knees) was less conclusive7. Safety margins may differ in revision knee arthroplasty, when vessels may be more vulnerable to injury. The question of whether the popliteal artery in a knee that has previously undergone arthroplasty is in the safest position at 90° of flexion has remained unanswered. By elucidating the movement of the popliteal artery with knee flexion after total knee arthroplasty, popliteal artery injury in revision total knee arthroplasty may become more preventable. Moreover, there is little information available on the difference in popliteal artery position between knees with posterior cruciate ligament (PCL)-retaining and PCL-sacrificing prostheses.
We hypothesized that movement of the popliteal artery with knee flexion would be altered after total knee arthroplasty, and that resection of the PCL would affect the movement of the popliteal artery. To confirm these hypotheses, we evaluated the change in position of the popliteal artery with knee flexion in knees that had received a PCL-retaining prosthesis and in knees that had received a PCL-sacrificing prosthesis, and we compared these results with the results from age-matched controls. Measurement of the position of the popliteal artery was performed with use of a noninvasive technique involving color-flow duplex ultrasonographic imaging (which superimposes a color Doppler ultrasonographic image on a conventional grayscale ultrasonographic image).
The study included twenty-two patients (forty-four knees) who had received a PCL-retaining Low Contact Stress (LCS) prosthesis (DePuy, Warsaw, Indiana) in one limb and a PCL-sacrificing prosthesis in the contralateral limb at our institution between May 2002 and May 2004. These patients were selected because they had had no clinical complications and had achieved full passive knee extension and at least 90° of knee flexion. The mean time since the arthroplasty was sixty-two months (range, thirty-one to ninety months. Twenty-two age-matched patients who were awaiting primary total knee arthroplasty for the treatment of osteoarthritis served as controls. No patient in any of the groups had undergone previous knee surgery other than the arthroplasty or sustained an injury involving a knee ligament. Table I summarizes the demographic and clinical characteristics of the patients in each group. The study was approved by our local institutional human studies committee.
A noninvasive technique involving color-flow duplex ultrasonographic imaging (HI VISION Avius; Hitachi Medical, Tokyo, Japan) and a 12-MHz linear probe were used to image the popliteal artery in the sagittal plane within the popliteal fossa. Color-flow duplex imaging is considered to be the most effective imaging tool for noninvasively assessing the positions of peripheral arteries16. With the patient in the prone position, the distance between the anterior border of the popliteal artery and the posterior surface of the tibia 1 cm distal to the posterior tibial joint line was accurately measured with the knee fully extended and at 30°, 45°, 60°, and 90° of flexion. In >90° of knee flexion, accurate imaging became more difficult. The probe was placed gently on the popliteal fossa, and the popliteal artery and the posterior tibial cortex were identified on the image (Fig. 1). To minimize interobserver variation, one experienced surgeon (M.T.) performed all examinations and measurements. The maximum intraobserver error was 0.9 mm when some of the distance measurements were repeated.
Statistical Analysis
The three groups were compared with use of a nonparametric Kruskal-Wallis test (one-way repeated-measures analysis of variance), and a post-hoc Bonferroni-Dunn test was performed. Odds ratios with accompanying 95% confidence intervals were calculated. An odds ratio was considered significant when 1.0 was not included in the 95% confidence interval. SPSS software (version 12.0; SPSS, Chicago, Illinois) was used for all statistical analyses.
When a total knee arthroplasty is performed, the implants must be placed in close proximity to the popliteal artery. Primary total knee arthroplasty is usually performed with the knee in at least 90° of flexion, which has been believed to be the safest position to avoid injuring the popliteal artery. However, safety margins may differ in revision knee arthroplasty. In addition, resection of the PCL during primary total knee arthroplasty may affect subsequent movement of the popliteal artery. We hypothesized that the movement of the popliteal artery with knee flexion would be altered by total knee arthroplasty, and that resection of the PCL would also affect movement of the artery. To confirm these hypotheses, we compared the change in position of the popliteal artery during knee flexion in knees that had received a PCL-retaining or a PCL-substituting prosthesis with that in age-matched controls.
Other investigators have used cadaveric studies or in vivo imaging involving angiography, MRI, or ultrasound to describe the position of the popliteal artery during knee flexion9-11. We used in vivo color-flow duplex ultrasonographic imaging for the following reasons: (1) duplex ultrasonographic imaging, unlike arteriography, is a noninvasive technique; and (2) MRI scans of knees following arthroplasty are often affected by artifacts resulting from the metallic implants.
Our study had several limitations. First, it has been reported that excessive pressure can be applied to the popliteal structures by the probe during the test9. In our study, we placed the probe gently on the popliteal fossa of the knee, which should not have affected the movement of the popliteal artery. In addition, we tested the effect of pressing the ultrasound probe more firmly against the popliteal fossa and found that the increase in pressure did not substantially affect the distance measurement. Second, total knee arthroplasty is usually performed with the patient in the supine position, but the subjects in our study were tested in the prone position, which eliminates the potential effect of gravity on the position of the popliteal artery during surgery in the supine position. However, a change to the supine position would presumably have had a similar effect on each of the three groups in our study, and the movement of the popliteal artery in both total knee arthroplasty groups in our study clearly differed from that in the control group. Finally, the selection of points representing the bone and the artery on the ultrasonographic image was subjective and thus potentially prone to two types of error, interobserver and intraobserver error. The effect of interobserver bias was minimized by having a single, experienced surgeon perform all measurements. Furthermore, arterial identification was easily performed with use of the color Doppler ultrasonography. Any systematic intraobserver error may have skewed the distance measurements in a positive or a negative direction. However, the effect of intraobserver error could be minimized by comparing the difference between full extension and 90° of flexion rather than examining only the absolute distance at each angle. Despite the limitations mentioned above, ultrasonography can be considered to be an effective tool for in vivo analysis of the position of the popliteal artery during flexion and extension of the knee.
Most authors have concluded that the popliteal artery moves away from the posterior aspect of the tibia with progressive flexion of the knee if the knee has not undergone arthroplasty6,7,13,14. The present study also demonstrated a similar marked movement of the popliteal artery as the knees in the control group moved from full extension to 90° of flexion. However, there has been some doubt regarding the safest knee position during revision knee arthroplasty7,15. Although the revision rate following total knee arthroplasty (estimated to be between 3.8% and 8.0%) is relatively low, the number of revision arthroplasty procedures can be expected to increase in the future17,18. By elucidating the movement of the popliteal artery with knee flexion in knees that have undergone arthroplasty, it may be possible to reduce the risk of popliteal artery injury during revision total knee arthroplasty.
Although the popliteal artery in the control group moved away from the tibia as the knee was flexed, the position of the artery remained nearly constant in the knees that had previously undergone arthroplasty involving either prosthesis design. Furthermore, in the knees that had undergone arthroplasty, the popliteal artery was actually more likely to move closer to the posterior surface of the tibia rather than away from it during movement of the knee from full extension to 90° of flexion. These findings differed significantly from those in the control group.
Our study indicated that 90° of flexion is not always the safest position in which to position the knee during revision arthroplasty, regardless of whether a PCL-retaining or a PCL-sacrificing prosthesis had been used previously. Furthermore, we observed no significant difference between knees with a PCL-retaining and a PCL-sacrificing prosthesis with regard to the position of the popliteal artery at any knee flexion angle. This result indicates that the treatment of the PCL in total knee arthroplasty does not affect the subsequent movement of the popliteal artery.
The reason that the movement of the popliteal artery in knees that had undergone arthroplasty differed from that in untreated knees remains unknown. Fibrosis and thickening of the capsule and synovium, which are clearly visible during revision surgery, provide a feasible explanation for why the popliteal artery in a knee that has undergone arthroplasty may move away from the knee joint as the knee is flexed. The hypertrophied soft tissues can lead to tethering of the popliteal artery, bringing it closer to the posterior aspect of the tibia with flexion.
It is important for the surgeon to exercise extreme caution during placement of retractors and while using oscillating saws and osteotomes in the posterior knee region during revision knee arthroplasty. The present study provides objective data supporting the theory that movement of the popliteal artery with knee flexion differs between knees that have undergone arthroplasty and those that have not. On average, the popliteal artery remained distant from the tibia, and thus in a relatively safe position, in knees that had undergone arthroplasty. However, the behavior of the popliteal artery during knee flexion varied among individuals who had undergone knee arthroplasty. In more than half of the knees that had undergone arthroplasty, the popliteal artery was closer to the tibia at 90° of knee flexion than it was in full extension. Surgeons should be aware that the popliteal artery may be vulnerable to injury during revision knee arthroplasty performed with the knee in a flexed position.
Color-flow duplex ultrasonographic imaging is one of the most effective imaging tools for noninvasive assessment of the movement of the popliteal artery during flexion in knees that require either primary or revision knee arthroplasty. Therefore, the use of noninvasive color-flow duplex imaging to preoperatively evaluate flexion-induced changes in the position of the popliteal artery could help to determine the optimal flexion angle to prevent popliteal artery injury during knee arthroplasty, and particularly during a revision procedure. More widespread use of ultrasonography for this purpose can be expected to result in safer primary or revision arthroplasty. Future studies should also compare the movement of the popliteal artery in the same patients before and after knee arthroplasty.
Note: The authors thank Takako Ujiie, ultrasonographer, for her advice on measuring the position of the popliteal artery with use of color-flow Doppler ultrasonography.