Risk factors associated with revision of reverse total shoulder replacement: a national registry analysis

Introduction Reverse Total Shoulder Replacement (reverse TSR) use is increasing. There is however limited information on the use and outcome of this procedure and how this varies according to the type of reverse TSR prosthesis used. Methods An analysis of all primary reverse TSRs reported to AOA NJRR from 16 April 2004 to 31 December 2014 was undertaken. Factors analysed included age, gender, prosthesis fixation and type and diagnosis. Outcome was time to first revision, diagnosis and type of revision were also assessed. Results There were 9682 (66.2% female) reverse TSRs reported to the AOANJRR during the study period. It is the most common class of TSR used in Australia accounting for 60.3% of all TSR procedures. This has increased from 43.3% in 2008. The median age is 77yrs females and 74 yrs males. Osteoarthritis is the most common diagnosis (46.0%) followed by rotator cuff arthropathy (33.6%) and fracture 14.1%. Over 70% of procedures use cementless fixation. Compared to conventional TSR, reverse TSR has an increased rate of revision in the first 3 months but at 7 years the CPR is 5.0% compared to 5.7% for conventional TSR. Age, primary diagnosis and humeral fixation are not risk factors. However, gender and prosthesis type are. There was a significantly higher revision rate in males in the first three months when used for OA (HR=2.37, 95% CI 1.45-3.88, p<0.001). The SMR/SMR L1 combination reverse TSR has a higher rate of revision compared to other reverse TSR prostheses. (HR=1.51, CI1.19-1.91, p<0.001). The main difference is an increased rate of revision for both instability/dislocation and loosening/lysis. Conclusions Identifying factors associated with increased risk of revision surgery is crucial to minimising revision rates nationally. The utilisation of reverse TSR is increasing. The AOA NJRR has identified that gender and prosthesis choice are important risk factors.