Ten-year survival and clinical results of constrained components in primary total knee arthroplasty.

The use of constrained prostheses in primary total knee arthroplasty (TKA) is unusual. This is a study of the indications, clinical results, and 10-year survival of these components. Fifty-four knees (44 patients) were prospectively followed, and 42 knees (34 patients) had 5 to 16 years (mean, 9 years) of follow-up. The indications for the components were valgus deformity with incompetent medial collateral ligament in 27 knees, severe flexion contracture with inability to balance the knee in 12, and others in 3. Knees were evaluated by the Hospital for Special Surgery and Knee Society score systems. Statistical analysis included paired Student t test and survivorship analysis. Of the entire cohort of 54 knees, there were only 2 failures: a tibial loosening revised at 3 years and a femoral loosening (no stem) revised at 1 year. Of the 42 knees with minimum 5-year follow-up, 12 knees were rated as excellent, 24 good, 3 as fair, and 3 as poor. There was a significant improvement in postoperative knee score, but not in the function score. The mean flexion contracture preoperatively was 17 degrees; postoperatively, 1.7 degrees; and mean flexion preoperatively, 93 degrees; postoperatively, 97 degrees. The 10-year survival with failure, defined as component revision for loosening, was 96% (confidence interval, 90.6%-100%). In difficult primary TKAs, there were 86% good or excellent results and a 10-year survival of 96% despite the increased constraint. The constrained condylar TKA remains indicated for knees with severe valgus deformity, incompetent medial collateral ligament, or severe flexion contracture in which the knee cannot be properly balanced.