HORMONE EXPOSURE AND VENOUS THROMBOEMBOLISM IN COMMERCIALLY-INSURED WOMEN 50 TO 64 YEARS OF AGE

Objective Determine whether hormone-associated venous thromboembolism (VTE) risk varies by exposure route and formulation in 50-64 year-old US women. Design Nested case-control study. Setting Large US commercially-insured population with patient-level claims data. Participants Women aged 50-64 years with at least one year of enrollment. Controls were matched to incident cases (10:1) on VTE date and case’s age (+/− 2yrs). Exclusions included prior VTE, intravascular vena cava (IVC) filter within twelve months, and anticoagulant exposure within 14 days. Exposures All estrogen and progestogen prescriptions (with route and formulation) filled within 12 months prior to index date were coded as current (0-60 days), past (61-365 days), or none. Contraceptives were categorized separately. Outcome Acute VTE cases were identified with ICD codes plus anticoagulant, IVC filter, or death within 30 days. Results Conditional logic regression analyses controlled for differences between cases (n=20,359) and controls (n=203,590) in Elixhauser comorbidities and VTE risk factors. Odds ratios (OR) were as follows: for current oral, unopposed estradiol 1.24 (95% CI: 1.09 to 1.40) or conjugated equine estrogen (CEE) 1.46 (95% CI: 1.28 to 1.68); for progestogens with estradiol 1.14 (95% CI: 0.95 to 1.37), with CEE 1.52 (95% CI: 1.25 to 1.84), or with ethinyl estradiol 2.35 (95% CI: 1.71 to 3.25). Current transdermal estradiol had the lowest ORs, whether unopposed, 0.70 (95% CI: 0.59 to 0.83) or combined with progestogens, 0.73 (95% CI: 0.56 to 0.96), but varied by progestogen. The OR for estrogen-progestogen contraceptives was 5.22 (95% CI: 4.67 to 5.84) compared to no exposure and 4.24 (95% CI: 3.64 to 4.98) compared to combined MHT. Conclusions In 50-64-year-old women, transdermal menopausal hormone therapy (estradiol with or without progestogens) did not elevate VTE risk. In contrast, contraceptives markedly increased VTE risk. Summary Boxes What is already known on this topic? Randomized controlled trials indicate that relative risk for venous thromboembolism (VTE) is approximately twice as high with menopausal hormone therapy (MHT) containing conjugated equine estrogen (CEE) with or without medroxyprogesterone acetate compared to no hormone exposure. Recent large, observational studies in the UK and Europe suggest that estradiol is lower risk than CEE and transdermal estradiol does not raise VTE risk compared to no hormone exposure, but results may not generalize to the United States because of differences in formulary, prescribing patterns, and background VTE incidence. What this study adds Using a large medical record database for US commercially-insured women 50-64 years of age, results confirmed that VTE risk was higher for oral compared to transdermal MHT and transdermal MHT (unopposed estrogen or combined with a progestogen) did not increase risk for VTE compared to no hormone exposure. However, unique US prescribing patterns included MHT with transdermal estradiol plus oral progestogens and MHT with ethinyl estradiol. MHT estrogen formulation affected VTE risk: ethinyl estradiol had higher risk than CEE, and CEE had higher risk than estradiol. Combined hormonal contraceptives (oral, vaginal, transdermal) had a markedly higher increase in VTE compared to MHT.