Journal of Clinical Medicine · Nov 15, 2026

Documented Hormone Replacement Therapy Use and Subsequent Rotator Cuff Tear Diagnoses and Repair Procedures: A Propensity Score-Matched Cohort Study

By Yang MY, Hung LW

Background: The association between hormone replacement therapy (HRT) use and rotator cuff outcomes remains uncertain. This study evaluated the association of baseline HRT documentation with subsequently recorded rotator cuff tear diagnoses and repair surgery in women with menopausal diagnoses. Methods: Using the TriNetX global federated health research database, we identified women aged ≥50 years at a qualifying menopausal or climacteric diagnosis recorded between January 2010 and December 2022. HRT exposure was classified according to the presence or absence of ICD-10-CM Z79.890 on or before the index date. Patients with prior rotator cuff tear or repair were excluded before 1:1 propensity-score matching. Matching included age, cardiometabolic comorbidities, osteoporosis, systemic corticosteroid use, tobacco use/nicotine dependence, systemic inflammatory disorders, and postprocedural ovarian failure. Outcomes recorded were rotator cuff tear diagnoses and repair procedures over 3 years. Hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated using Cox proportional hazards models. Sensitivity analyses examined an alternative propensity-score specification and a medication-corroborated exposure definition. E-values quantified sensitivity to unmeasured confounding. Results: The primary matched analysis included 27,918 women per group. Mean age was 59.2 years in both groups, and all displayed baseline characteristics had standardized mean differences <0.1. Mean follow-up was 1035 days in the HRT group and 1028 days in the comparator group. Observed event proportions were 2.28% versus 1.90% for rotator cuff tear diagnoses (636 versus 531 events; HR, 1.192; 95% CI, 1.063–1.338) and 0.35% versus 0.20% for repair procedures (97 versus 57 events; HR, 1.691; 95% CI, 1.219–2.345). Sensitivity analyses showed directionally consistent associations. E-values for the point estimate and lower confidence limit were 1.67 and 1.32 for tear diagnoses and 2.77 and 1.74 for repair procedures. Conclusions: Baseline HRT documentation was associated with subsequently recorded rotator cuff tear diagnoses and repair procedures. Residual confounding by pretreatment hormonal status and treatment indications remains possible. These findings do not establish causality or support changing HRT initiation, continuation, or discontinuation on the basis of rotator cuff risk.

doi.org/10.3390/jcm15228952Populations

AgingResearch.News