Association of preoperative testosterone replacement therapy with postoperative complications following rotator cuff repair

Rotator cuff repair (RCR) is increasingly performed due to advancements in surgical techniques and an aging population. While generally successful, complications like re-tear, stiffness, infection, and thromboembolic events remain concerns. The rising use of testosterone replacement therapy (TRT) in middle-aged and older men raises questions about its impact on surgical outcomes, as its effect on RCR complications remains unclear. A retrospective cohort study was conducted using the PearlDiver Database. Patients who underwent arthroscopic RCR between January 2010 and April 2023 were identified using Current Procedural Terminology code 29827. Those with at least two years of continuous follow-up were included, while patients under 21 or with unknown procedural laterality were excluded. Patients were categorized into 2 cohorts: those who received TRT within three months preoperatively and a control group who did not. Propensity score matching (1:1, caliper = 0.001) was performed to control for age, gender, Charlson Comorbidity Index, obesity, tobacco use, and hypogonadism. The primary outcomes included 2-year postoperative complications and reoperations. Secondary outcomes included 90-day major medical complications such as surgical site infection, pneumonia, pulmonary embolism, deep vein thrombosis, urinary tract infection, wound dehiscence, sepsis, acute kidney injury, and readmissions. Statistical analyses were performed using chi-square tests, and odds ratios with 95% confidence intervals were calculated. A total of 8,241 TRT users and 673,982 control patients were identified before matching. After propensity score matching, 5,109 patients remained in each cohort with no significant baseline differences. No significant differences were observed in 90-day postoperative complications, including surgical site infection (0.5% vs. 0.4%, P = .64), pneumonia (0.6% vs. 0.5%, P = .79), sepsis (0.4% vs. 0.3%, P = .51), acute kidney injury (0.7% vs. 0.5%, P = .18), or readmissions (1.2% vs. 1.0%, P = .29). Similarly, no significant differences were found in pulmonary embolism or deep vein thrombosis. Over the 2-year follow-up, TRT use was associated with a higher incidence of total shoulder arthroplasty (0.7% vs. 0.4%, P = .037) but a lower incidence of lysis of adhesions (0.5% vs. 0.9%, P = .032). No significant differences were observed in revision RCR rates (2.6% vs. 2.3%, P = .41). Preoperative TRT use was not linked to increased short-term complications after arthroscopic RCR. Over two years, TRT was associated with higher rates of total shoulder arthroplasty but lower rates of lysis of adhesions, with no difference in revision RCR.