Exogenous Testosterone Use Increases the Risk of Central Serous Chorioretinopathy.

Abstract

OBJECTIVE To evaluate the association between exogenous testosterone (ExoT) use and central serous chorioretinopathy (CSCR). DESIGN Retrospective observational cohort study using large-scale electronic health record (EHR) data. SUBJECTS/PARTICIPANTS Patients receiving exogenous testosterone therapy, identified from the MarketScan and STARR databases. METHODS Data were accessed from two sources: Merative™ MarketScan® Commercial Database, and Stanford's Clinical Data Warehouse (STARR), which aggregates de-identified patient records from Stanford Health Care. Patients on testosterone therapy were included and categorized by CSCR status. Demographic factors such as sex (administrative field), race, and ethnicity were assessed. Laboratory values (testosterone, hematocrit, RBC count, cortisol) were compared in STARR, with limited availability in MarketScan. Logistic regression analyses were performed in MarketScan adjusting for age and sex. A sensitivity analysis restricted to patients exposed to ExoT prior to CSCR diagnosis was also performed. MAIN OUTCOME MEASURES The primary outcomes were CSCR prevalence and adjusted odds ratios (AOR) for CSCR risk in patients on exogenous testosterone. Secondary outcomes included differences in laboratory values and treatment requirements (photodynamic therapy [PDT] and intravitreal injections). RESULTS In STARR, individuals with CSCR on exogenous testosterone had significantly higher mean testosterone levels (p=0.001), hematocrit (p=0.022), and RBC counts (p=0.005) compared to those without CSCR. In MarketScan, laboratory values trended in the same direction but were not statistically significant, likely reflecting limited sample sizes. Logistic regression in MarketScan showed that exogenous testosterone was significantly associated with increased CSCR risk (AOR: 8.05; 99% CI: 6.04-10.73). In both datasets, there were no significant differences in treatment rates (PDT or intravitreal injections) between ExoT and non-ExoT users. In a sensitivity analysis restricted to patients who received ExoT prior to CSCR diagnosis, no significant laboratory differences were observed. CONCLUSIONS This study demonstrates a significant association between exogenous testosterone use and increased CSCR risk, highlighting the importance of monitoring patients on testosterone therapy for potential ocular symptoms, especially among high-risk demographic groups.