Contemporary patellofemoral arthroplasty is associated with lower revision rates compared with the previous decade: A...

The aim of this study was to assess global data on patellofemoral arthroplasty (PFA) survival a decade after a prior investigation by Reihs et al. using the same endpoints and data sources, aiming to quantify potential changes in implant survival over this period. The hypothesis was that the pooled revision rate of contemporary PFA, quantified as revisions per 100 component-years (CY), would be lower than prior decade estimates derived from identical methodology and registry sources.

A systematic search in Embase, MEDLINE and CENTRAL through October 2025 using a predefined search strategy was conducted using Ovid. Two reviewers screened and extracted data. The primary effect measure was revisions per 100 CY. Pooled estimates were obtained with fixed- and DerSimonian-Laird random-effects. Risk of bias was assessed with ROBINS-I. National arthroplasty registries meeting EAR/EUPHORIC Type A.1.1.1.1. criteria were analyzed.

Fifteen primary studies were included, with most series using contemporary onlay designs. Per-study rates ranged from 0.355 to 3.400/100 CY. The pooled revision rate was 1.59/100 CY (95% confidence interval [CI] = 1.24-2.04) with random effects, corresponding to a 10-year revision rate of 15.9%, representing an absolute improvement of 6.3% in survivorship compared to the benchmark analysis by Reihs et al. heterogeneity was moderate (I² = 52.7%). ROBINS-I indicated predominantly critical risk of bias. Registry synthesis (NJR, NZJR, AOANJRR, 117.524 CY) yielded 1.85/100 CY (95% CI = 1.775-1.931), corresponding to a 10-year revision rate of 18.5% (95% CI = 17.75-19.3) and an improvement of 0.4%-8.5% in the last decade if compared to Reihs et al. CONCLUSIONS: This meta-analysis demonstrates that modern PFA has a more favourable survivorship profile over the past decade, particularly with onlay designs. Proper patient selection and registry-based surveillance remain key to further optimization.

Level IV.