High rate of implant loosening for uncemented resurfacing-type medial unicompartmental knee arthroplasty.

PURPOSE: To determine the medium-term implant survivorship, the clinical results and the failure mechanisms of a novel unicompartmental arthroplasty for uncemented resurfacing of the medial tibio-femoral compartment. METHODS: Seventy-six consecutive patients were prospectively evaluated with a mean final follow-up of 6 years (SD 5.3 months). In 44 patients, the diagnosis was osteoarthritis, and in 32 patients, it was avascular necrosis of the medial femoral condyle. The Hospital for Special Surgery Score was used for objective clinical evaluation, and a self-administered visual analogue scale was used to quantify residual pain at each observation point. Implant survivorship was determined assuming revision for any reason as endpoint. RESULTS: Nineteen patients were revised (6 with osteoarthritis and 13 with avascular necrosis of the medial femoral condyle). The mean interval time from index surgery to revision was 11.2 months (SD 4.66 months). Implant survivorship was higher in patients with osteoarthritis with respect to those with avascular necrosis of the medial femoral condyle (p = 0.018). Aseptic loosening was the most frequent failure mechanism. Femoral component loosening was reported in five patients and tibial component loosening was reported in other six patients. Assuming revision for any reason as endpoint, an implant survivorship of 74.3 % at 6-year follow-up was determined. In the remaining 57 patients, satisfactory clinical results were obtained. Hospital for Special Surgery Score and visual analogue scale for residual pain showed significant improvements (p < 0.03 and p < 0.045, respectively). CONCLUSIONS: At the present time, the standard cemented implants and the conventional designs for unicompartmental knee replacement still represent the optimal solution. The authors do not recommend the widespread use of this technique. LEVEL OF EVIDENCE: IV.