Complex Primary Total Knee Arthroplasty: Long-Term Outcomes

Background: Total knee arthroplasty in patients with severe preoperative deformity, ligamentous instability, and/or marked bone loss occasionally requires a varus and valgus constrained or rotating-hinge design prosthesis. The purpose of this study was to compare patient populations that underwent primary total knee arthroplasty with constrained or unconstrained total knee arthroplasty implants to determine patient demographic characteristics, long-term survival, and reasons for reoperation and revision for each group.

Methods: We identified 28,667 primary total knee arthroplasties performed from 1979 to 2013 at our institution. A total of 427 knees had a varus and valgus constrained design and 246 knees underwent rotating-hinge total knee arthroplasties. Patient demographic information and preoperative diagnoses were analyzed by implant type. A multivariate analysis was performed to account for age, sex, and body mass index (BMI). Kaplan-Meier survival rates for each complication leading to reoperation or component revision were determined at 10 and 20 years. Adjusted hazard ratios were determined for the most common causes for reoperation and revision compared with a routine total knee arthroplasty control group.

Results: Patient demographic characteristics were significantly different (p < 0.05) between all groups for age, sex, and BMI. The varus and valgus constrained and rotating-hinge groups had decreased survival free of all-cause reoperation at 10 and 20 years compared with the unconstrained total knee arthroplasty group, with a hazard ratio of 1.74 (95% confidence interval [95% CI], 1.36 to 2.23) for the valgus and varus constrained group and 2.07 (95% CI, 1.58 to 2.70) for the rotating-hinge group. The adjusted hazard ratio for all-cause revision was significantly higher for the varus and valgus constrained group at 1.65 (p = 0.007) but not for the rotating-hinge group at 1.48 (p = 0.054) compared with the unconstrained total knee arthroplasty group. Wear and osteolysis, infection, and fracture were the most common reasons for component revision in both groups.

Conclusions: We found increased reoperation and revision rates associated with the use of constrained implants at the time of index total knee arthroplasty. The rate of component revision for any reason at 10 years was >2 times higher in the constrained total knee arthroplasty groups compared with the unconstrained total knee arthroplasty group. At 20 years postoperatively, the component revision rate was >3 times higher.

Level of Evidence: Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.