Patellar resurfacing in primary total knee replacement: a meta-analysis.

BACKGROUND: Treatment of the patella during total knee replacement is an area of continuing debate. We performed a meta-analysis of randomized controlled trials to address the hypothesis that patellar resurfacing in primary total knee replacement improved patient outcome.
METHODS: Randomized controlled trials comparing patellar resurfacing with nonresurfacing in primary total knee replacement were included. The primary outcomes analyzed were knee scores, anterior knee pain, and patient satisfaction. We also investigated the prevalence of complications, revision surgery related specifically to the patellofemoral joint, the infection rate, operative time, and radiographic appearance.
RESULTS: Sixteen randomized controlled trials assessing 3465 knee replacements were eligible; 1710 procedures included patellar resurfacing and 1755 did not. The knee component of the Knee Society Score was significantly higher in the resurfacing group (p = 0.005); however, no significant difference was observed for the function component of the Knee Society Score or for any other reported knee score. Anterior knee pain was reported in 13% of resurfaced knees and in 24% of nonresurfaced knees; this difference was not significant (p = 0.1). Patients were satisfied with the outcome after 485 (90%) of 539 procedures that included patellar resurfacing compared with 488 (89%) of 548 that did not; this difference was not significant. There were ninety-three reported patellofemoral complications in the resurfacing group and 205 in the nonresurfacing group; this difference was significant (p = 0.02) in a random-effect model. The rate of reoperation because of anterior knee pain (p < 0.00001) and the rate of reoperation because of any patellofemoral complication (p = 0.002) were significantly higher in the nonresurfaced group. No differences were found in the analyses of infection rate, operative time, or radiographic appearance.
CONCLUSIONS: Patients who underwent patellar resurfacing experienced anterior knee pain and satisfaction with the arthroplasty procedure that were equivalent to those experienced by patients whose patella was not resurfaced; however, these patients underwent significantly fewer additional surgical procedures. Further long-term follow-up of modern prostheses in randomized studies measuring outcome with a patella-specific score is needed.