Abstract
Background
A recent Cochrane review has shown that total shoulder arthroplasty (TSA) seems to offer an advantage in terms of shoulder function over hemiarthroplasty, with no other obvious clinical benefits. This is the first study to compare complication rates on a national scale.
Methods
All patients (9804 patients) who underwent either TSA or shoulder hemiarthroplasty as a planned procedure between 2005 and 2008 in the English NHS were identified using the hospital episodes statistic database. Data was extracted on 30-day rates of readmission, wound complications, reoperation and medical complications (myocardial infarction (MI) and chest infection (LRTI)), and inpatient 90-day DVT, PE and mortality rates (MR). Revision rate at 18 months was analysed for the whole cohort and, for a subset of 939 patients, 5-year revision rate. Odds ratio (OR) was used to compare groups.
Results
Mean age was 70.6 years for the TSA group and 69.3 for hemiarthroaplasty. There were no significant differences in 30-day readmission (0.82% vs. 0.83%, OR 0.99 (95% CI 0.63–1.55)), wound complication (0.57% vs. 0.51%, OR 1.11 (0.63–1.94)), and reoperation rates (0.20% vs. 0.22%, OR 0.89 (0.36–2.19)). 18-month revision was significantly higher in the TSA groups (3.03% vs. 1.76%, OR 1.75 (1.33–2.28), but at 5-years no difference was seen (3.85% vs. 3.67%, OR 0.95 (0.46–1.91).
The overall 90-day MR was 0.43% (42 patients). 0.58% had a LRTI and 0.29% had an MI. There were 13 DVTs (0.13%) and 22 PEs (0.22%, 5 fatal).
Discussion
Despite concerns regarding osteolysis around the glenoid peg after TSA, revision rates at 5 years were not significantly different to hemiarthroplasty. Except for 18-month revision rates, there were no differences in any of the outcome measures in this study. Methods of national data analysis described in this study can provide benchmark values for future subspecialty revalidation.
