EMERGING TRENDS, PATIENT CHARACTERISTICS AND SUBSEQUENT OUTCOMES OF SHOULDER ARTHROPLASTY FOR ACUTE TRAUMA: AN ANALYSIS OF 1,524 PATIENTS FROM THE NATIONAL JOINT REGISTRY OF ENGLAND, WALES, NORTHERN IRELAND AND THE ISLE OF MAN

摘要

Purpose: Procedure selection following proximal humeral fractures remains an area of uncertainty. Our aim was to use NJR data including patient reported outcome scores to describe current practice and identify factors predictive of better outcomes. Methods: We analysed observational data for all primary procedures recorded in the NJR for acute trauma between April 2012 and March 2017. Monthly case volumes were estimated using locally weighted regression. OSS and revision risk were estimated using linear regression with penalised maximum likelihood and an elastic net regularised Cox proportional hazards model. Missing data were handled with multiple imputation. Results: 1,523 acute trauma cases were available for analysis, including 661 hemiarthroplasties (HA) and 820 reverse total shoulder arthroplas-ties (RTSA). 79% of patients were female and the median age was 74.5 years (IQR: 67.8-80.1). Over the study period, rates of HA increased from 5 per month to a peak of 14 in 2014 before falling to 8. Over the same period the monthly rate of RTSA increased from 2 to 26. 571 cases were available for the OSS analysis. Predictors of a better OSS were use of RTSA, lower ASA grade and male gender. The unadjusted mean OSS for RTSA was 32.7 (95% Cl 31.4-34.0) and for HA was 26.9 (95% Cl 25.6-28.1). Use of RTSA was associated with a lower revision rate at two years. Conclusion: There has been a rapid recent change in UK practice towards reverse arthroplasty for acute proximal humeral fractures. While the results of this analysis need careful interpretation due to potential missing confounding variables, the observed change is currently supported by better patient reported outcomes and lower early revision rates.