Session: OR34-Testosterone Replacement Therapy: Risks and Benefits
Room 6A (San Diego Convention Center)
Purpose: Investigate whether an association exists between TRT use and VTE in hypogonadal men via retrospective cohort and nested‑case‑control analyses.
Methods: Truven Databases were reviewed to obtain data from men with a hypogonadal condition (either treated with an approved TRT product or untreated but with a hypogonadal diagnosis meeting prespecified International Classification of Diseases [ICD‑9] criteria) who were ≥18 years old, continuously enrolled (≥12 months) in a healthcare plan, and did not have a VTE diagnosis during baseline. Propensity‑score (PS) 1:1 matching was used in the cohort analysis to ensure comparability between TRT‑treated and untreated patients (pts) at baseline. In nested‑case‑control analyses, TRT‑treated pts were matched to untreated pts in a 1:4 ratio based on age and calendar year. Among TRT-treated pts, index date was defined as the first TRT prescription date; among untreated pts, index date was assigned at random according to the distribution observed for treated pts. The outcome was defined as incident VTE, primarily Idiopathic VTE; exposure was assessed as primarily any TRT use and by different TRT administration routes. Cox regression and conditional logistic regression models were used in cohort and nested‑case‑control analyses, respectively, to assess hazard ratios (HRs) and odds ratios (ORs) for TRT/VTE associations. Sensitivity analyses were also performed using different TRT exposure and VTE parameters.
Results: 533,223 hypogonadal men met study inclusion criteria. After PS 1:1 matching, 102,650 TRT-treated pts and 102,650 untreated pts with Idiopathic VTE were included in the analysis.
Retrospective cohort analysis revealed an HR for Idiopathic VTE that was 1.08 for all TRT-treated pts (95% CI: 0.91, 1.27; p=0.378), 1.07 for topical/gel TRT‑treated pts (95% CI: 0.88, 1.29; p=0.496), and 1.32 for injectable TRT‑treated pts (95% CI: 0.89, 1.96; p=0.164). Stratification of pts by age (≤65 vs >65 years old) revealed similar, statistically non‑significant results.
Nested‑case‑control analyses of 2,785 TRT-treated cases and 11,119 untreated controls revealed similar, non‑significant findings. The OR for current TRT was 1.02 (95% CI: 0.92, 1.13; p=0.702), while the OR for past TRT was 0.92 (95% CI: 0.82, 1.03; p=0.145). Similar, non-significant findings were observed with age and TRT administration-route stratification.
Results from most sensitivity analyses supported those observed from the cohort and nested‑case‑control analyses.
Conclusion: The results from this study do not support an association between exogenous testosterone replacement therapy in hypogonadal men and an increased risk of venous thrombotic events.
Disclosure: HL: Employee, Eli Lilly & Company, Employee, Eli Lilly & Company. NLO: Employee, Eli Lilly & Company, Employee, Eli Lilly & Company. KB: Employee, Eli Lilly & Company, Employee, Eli Lilly & Company. WW: Employee, Eli Lilly & Company, Employee, Eli Lilly & Company. SPM: Employee, Eli Lilly & Company, Employee, Eli Lilly & Company.
*Please take note of The Endocrine Society's News Embargo Policy at https://www.endocrine.org/news-room/endo-annual-meeting
See more of: Abstracts - Orals, Poster Previews, and Posters