Unicompartimental knee arthroplasty outcome is sometimes compared to total knee arthroplasty but various implant parameters might greatly influence this outcome. The objectives of this study were to report the results of a consecutive series of 172 all-polyethylene unicompartmental knee arthroplasties (UKAs) and to detail possible factors of success and failure.

It is possible to outline implant and technique factors determining success or failure in unicompartimental knee arthroplasty.

One hundred seventy-two HLS-type cemented resurfacing UKAs, with the femoral implant made of chrome-cobalt and the tibial implant tibial entirely in polyethylene (without anchorage studs) were consecutively implanted between 1988 and 2004 in 134 patients (111 females and 23 males) in our center according to the indications established in 1988, using the same technique for each surgery. The patients’ mean age was 72.2 years (range, 25–90 years). The review rate was 83.7% (144 UKAs), with a mean follow-up of 62.3 months (range, 24–160 months). The series included 84 medial UKAs and 60 lateral UKAs. The clinical data were analyzed using the IKS criteria and the patients had a complete radiological evaluation before surgery and at the last follow-up.

The rate of satisfied or very satisfied patients was 97.2%. No pain or slight pain was found in 81% of the cases. The mean flexion was 133° (range, 85–150°). The mean knee score varied from 63.6 before surgery to 91.5 (90.4 for medial UKAs and 92.9 for lateral UKAs) and the function score from 63.6 to 83.8 (84.7 for medial UKAs and 82.6 for lateral UKAs). The mean range of motion was 133° (range, 85–150°), better than the medial UKAs for osteonecrosis. The mean residual deformity was 4° varus for the medial UKAs and 2° valgus for the lateral UKAs. A radiolucency was found in 23% of the cases (20% tibial and 3% femoral), nonprogressive in all cases. In 87.2% of the cases, the opposite femorotibial compartment remained radiologically normal. No progression to osteoarthritis in the femoropatellar joint required additional surgery. Sixteen patients required revision surgery: in six cases, the implant was removed and a total prosthesis implanted (one late infection, one case of involvement of the opposite compartment, and four cases of tibial component loosening). In the other cases, one tibial baseplate was changed, five arthroscopies were done, and four unicompartmental knee replacements were done on the opposite compartment. The Kaplan-Meier survival rate (taking into account the revisions with implant change) was 95.6. The results of this series were very satisfactory and were similar to recent series in the world literature that showed survival rates between 90 and 98% at 10 years, rates that are equivalent to those found for total knee replacements. The mean flexion range of motion found was higher than the majority of other recent series, probably because of the precise patient selection in the present study, a minimally invasive approach, and the femoral implant design with an ascending condylar posterior cut. The deterioration of the contralateral compartment is frequently reported, but was perhaps prevented by the absence of overcorrection and patient selection. In this series, none of the UKAs was revised for wear. We explain this by the systematic preservation of a moderate undercorrection, particularly for medial UKAs, the quality of the polyethylene, and a selection based on patient weight and age.

The option of an all-polyethylene tibial implant, with minimal bone cuts (femoral resurfacing), makes excellent long-term results possible.

Level of Evidence: Level IV. Therapeutic Study.