Abstract
Rotator cuff repair (RCR) can be performed open or arthroscopically, with a recent dramatic increase in the latter. Despite controversy about the preferred technique, there has been an increase in the number of repairs performed arthroscopically. The purpose of this study was, therefore, to compare revision rates following open and arthroscopic RCR repair.
Adult patients undergoing first-time, primary rotator cuff repair in Ontario, Canada (April 2003-March 2014) were identified using physician billing and hospital databases. Patients were followed for a minimum of two and up to 13 years for the primary outcome, revision rotator cuff repair, and secondary outcome, surgical site infection. The intervention considered was open versus arthroscopic technique. Patient factors (age, gender, residence, socioeconomic status, medical comorbidities) and provider factors (surgical volume, hospital setting, worked night before, year of surgery) were recorded. Standardized mean differences were used for covariate comparison. A Cox Proportional Hazards model was used to compare RCR survivorship between the two groups after adjustment for patient and provider factors, generating hazard ratios with 95% confidence intervals (HR, 95% CIs). Censoring occurred on the first of the primary outcome, death, shoulder arthroplasty or arthrodesis, or the end of the follow-up period (March 2016).
A total of 37,255 patients were included. The overall revision RCR rate was 2.9% (1,096 patients) with a median time to revision of 23 months (IQR 12–52). Revision repair was more common in the arthroscopic group in comparison to the open group (3.2% vs 2.6%, NNT 166.7, p=0.004), with an adjusted HR of 0.72 (0.63–0.83 95% CI, p < 0 .0001). The surgical site infection rate was significantly higher in the open group compared with the arthroscopic group (0.5% vs 0.2%, NNT 333.3, p < 0 .001). Patient and provider covariates had no statistically significant effect on revision rates, aside from increasing age (per 10 year increase, HR 0.85, 0.81–0.90 95% CI, p < 0 .0001).
Revision rotator cuff repair is approximately 30% more common in patients undergoing arthroscopic repair, in comparison to open repair, after adjustment. Surgical site infection is uncommon regardless of surgical technique, however, it is slightly more common following open repair. In the setting of an economic healthcare crisis, trends of increasing arthroscopic RCR may demand scrutiny, as the technique is associated with higher revision rates and higher costs.
