Patellar Resurfacing in Primary Total Knee Arthroplasty: A Meta‐analysis and Trial Sequential Analysis of 50 Randomized Controlled Trials

OBJECTIVE: During total knee arthroplasty, femur and tibia parts are regularly replaced, while resurfacing the patellar or not is an ongoing discussion. To compare revision rate, anterior knee pain rate, patient‐reported outcome measures, complication, radiographic, and clinical outcomes after patellar resurfacing versus non‐resurfacing in total knee arthroplasty. METHODS: PubMed, Medline, EMBASE, CENTRAL, and CINAHL databases were searched on 25 April 2021 to enroll randomized controlled trials that compared patellar resurfacing versus non‐resurfacing. We used the grading of recommendations assessment, development and evaluation (GRADE) framework to assess the certainty of evidence. Our primary outcome was revision rate and secondary outcomes was anterior knee pain rate. Outcomes were pooled using the random‐effect model and presented as risk ratio (RR), or mean difference (MD), with 95% confidence interval (CI). RESULTS: Fifty studies (5586 knees) were included. Significant reductions in patellar revision rate (RR 0.41, 95% CI [0.19, 0.88]; P = 0.02; I (2) = 24.20%) and non‐patellar revision rate (RR 0.64, 95% CI [0.55, 0.75]; P < 0.001; I (2) = 0%) were seen after patellar resurfacing. Patellar resurfacing significantly reduced the anterior knee pain rate than nonresurfacing (RR 0.72, 95% CI [0.57, 0.91]; P = 0.006; I (2) = 69.5%). Significant differences in patient‐reported outcome measures were found. However, these differences were inconsistent and lacked clinical importance. Patellar resurfacing resulted in a significant lower rate of patellar clunk (RR 0.58, 95% CI [0.38, 0.88]; P = 0.01; I (2) = 0%), a higher patellar score (MD 1.24, 95% CI [0.67, 0.81]; P < 0.001; I (2) = 73.8%), but prolonged surgical time (MD 8.59, 95% CI [5.27, 11.91]; P < 0.001; I (2) = 88.8%). CONCLUSIONS: The clear relationship is that patellar resurfacing reduces revisions, anterior knee pain, and patellar clunk. It will be interesting to compare the initial cost with the revision cost when required and cost‐utility analysis with long‐term results in future studies.