The changing effect of surgeon volume on revision rates in shoulder arthroplasty with time in Australia.

Abstract

The concept that surgical volume affects rates of total primary shoulder arthroplasty (TSA) revision is widely accepted. The aim of this study was to determine if surgical volume is confounded by patient, implant, and institutional factors and time since TSA. Using data from a large national arthroplasty registry for the period January 1, 2008, to December 31, 2022, all interactions with orthopedic surgeon volume (SV) undertaking primary TSA procedures for all diagnoses were examined. A subanalysis from January 1, 2017, provided a contemporary analysis with additional patient demographics. The primary outcome measure was the cumulative percentage revision (CPR), which was defined using Kaplan-Meier estimates of survivorship and hazard ratios from Cox proportional hazard models with a multivariable adjustment. The incidence of orthopedic surgeons completing TSA increased nationally from 0.9 per 100,000 population in 2008 to 1.8 in 2022. Of 55,605 TSA procedures, the CPR at 11 years varied from 8.8% (95% confidence interval [CI] 8.0, 9.6) to 5.6% (95% CI 4.0, 7.7) for implants by surgeons with SV <10 per annum and for prostheses by SV >35 per annum, respectively. Instability or dislocation was the most common reason for revision. SV had a significant effect on the revision rate for all diagnoses (P = .001), instability or dislocation (P = .018), and revision within 5 years of primary arthroplasty (P = .011). In contrast, at subanalysis (last 5 years), there was no interaction both with main effects (SV effect adjusted by age and gender) and with an extended adjustment for both instability or dislocation and all-cause revision. Nationally there is an increasing incidence of primary TSA procedures and surgeons performing them. The association between SV and revision rate is complex. Although there is an association with main effects, there was no association in the past 5 years. Reduction in instability or dislocation revision rates contributed to this change. Patient, implant, or institutional factors did not confound analysis of SV on primary TSA revision rates but recency did.