Postoperative knee motion after total knee arthroplasty has often been viewed as a surrogate for improved patient satisfaction11,22. Over the last decade, there have been numerous innovations in total knee arthroplasty design aimed at improving knee kinematics. One of these design changes has been the introduction of implants with the potential for a high degree of flexion. "High-flexion" designs often require removal of additional bone from the posterior femoral condyle and the addition of more metal at an increased radius of curvature, thus allowing for the theoretical increase in the motion arc. In addition to the high-flexion design, many authors have advocated use of mobile-bearing designs for younger or higher-demand patients because of the potential reduction in polyethylene wear and the restoration of more normal kinematics3. Investigators who have analyzed the short-term clinical outcomes of these design changes have found improvement in the range of motion.
In their study "Comparison Between Standard and High-Flexion Posterior-Stabilized Rotating-Platform Mobile-Bearing Total Knee Arthroplasties. A Randomized Controlled Study," Choi et al. evaluated patients with a total of 170 total knee arthroplasties who had been randomly assigned to receive either a standard (n = 85) or a high-flexion (n = 85) rotating-platform mobile-bearing knee prosthesis. All patients were followed for a minimum of two years. Data were collected in a blinded fashion. Of the original 170 knees, twenty were withdrawn from the study for various reasons. The remaining knees were subdivided into three subgroups, (1) eighty-six patients (forty-one in the standard group and forty-five in the high-flexion group) with a unilateral arthroplasty, (2) twenty patients (forty knees) with bilateral arthroplasty who had the same prosthesis implanted in each knee, and (3) twenty-two patients (forty-four knees) with bilateral arthroplasty who had a standard prosthesis in one knee and a high-flexion prosthesis in the other. The authors concluded that there was no significant difference between the standard and high-flexion prosthetic designs with regard to any of the parameters that they evaluated, including radiographic, functional, and patient satisfaction indices. The average range of knee motion was 130° with the standard design and 128° with the high-flexion design.
This is a well-written paper, and the authors have probably arrived at the correct conclusion. In general, the field needs well-done prospective, randomized studies to test various hypotheses.
However, we do have several fundamental concerns with the methodology utilized in the study. The study was prospective, and over 90% of the patients were women who had excellent preoperative motion (approximately a 126° to 128° arc in both groups). Obviously, this group does not reflect the typical population of patients undergoing total knee arthroplasty. Since the preoperative range of motion remains one of the most reliable contributors to the postoperative range of motion, it would seem that the window of opportunity to discriminate between the motion increases obtained with the two different designs was limited from the outset of the study. The authors chose 5° of flexion as the minimum clinically important difference and calculated that they would require sixty-four knees in each group to achieve adequate power to differentiate between the designs. The actual study cohort included only seventy-four knees in the standard group and seventy-six in the rotating-platform group. As previously noted, there were twenty bilateral cases with one of each type of prosthesis implanted in each knee. The reliability of the observations reported in the study is jeopardized because of questions regarding the non-independence in these bilateral cases, which could have weakened the power of the study to identify a minimum clinically important difference. Also, there were inconsistencies in the study population regarding treatment of the patellofemoral joint. Although the literature presents conflicting opinions regarding the necessity of patellar resurfacing, it would be advantageous to avoid this inconsistent treatment in a prospective study, where one would expect both groups to be treated similarly. It would also be informative to know satisfaction indices in the third group (a high-flexion prosthesis in one knee and a standard prosthesis in the other) to see if there were any patient preferences.
We laud the authors for their excellent report. Additional long-term follow-up studies will be necessary to better understand this subject.