Kneeling and standing up from a chair as performance-based tests to evaluate knee function in the high-flexion range: a randomized controlled trial comparing a conventional and a high-flexion TKA design

BACKGROUND: We compared the functional outcome between conventional and high-flexion total knee arthroplasty (TKA) using kneeling and sit-to-stand tests at 1 year post-operative. In addition, the patient's daily functioning, pain and satisfaction were quantified using questionnaires. METHODS: We randomly assigned 56 patients to receive either a conventional or a high-flexion TKA. Primary outcomes were maximum flexion angle and maximum thigh-calf contact measured during kneeling at 1 year post operatively. Secondary outcomes were the angular knee velocity and ground reaction force ratio measured during sit-to-stand performance tests, and questionnaires. RESULTS: At one year post-operative, maximum knee flexion during kneeling was higher for the high-flexion TKA group (median 128.02° (range 108-146)) compared to the conventional TKA group (119.13° (range 72-135)) (p = 0.03). Maximum thigh-calf contact force was higher for the high flexion TKA group (median 17.82 N (range 2.98-114.64)) compared to the conventional TKA group (median 9.37 N (range 0.33-46.58))(p = 0.04). The sit-to-stand tests showed a significantly higher angular knee velocity in the conventional TKA group (12.12 rad/s (95%CI 0.34-23.91); p = 0.04). There were no significant differences between groups in ground reaction force ratios and patient-reported outcome scores. CONCLUSION: Although no differences were found in patient-reported outcome scores, differences in performance-based tests were clearly apparent. Standing up from a chair at 90° of knee flexion appeared to be easier for the conventional group. The kneeling test revealed significantly higher weight-bearing knee flexion for the high-flex group. Hence, if kneeling is an important activity for a patient a high-flex design may be recommendable. TRIAL REGISTRATION: The study was retrospectively registered in ClinicalTrials.gov under identifier NCT00899041 (date of registration: May 11, 2009).