Abstract

Background. Postmenopausal estrogen (E) therapy, especially in combination with progestin (P) doubles the relative risk of venous thrombosis (VTE). Risk with hormones is higher with increasing age, obesity and with factor V Leiden. We studied coagulation markers as susceptibility factors for postmenopausal hormone-related VTE.

Methods. The Women’s Health Initiative program included two placebo-controlled double-blind randomized trials of two E regimens, E (conjugated equine estrogens) or E+P (E + medroxyprogesterone acetate), in 16,608 postmenopausal women aged 50–79. We performed a nested case control study that measured baseline levels of coagulation markers in 215 women who developed VTE during follow up and 867 age-matched controls. The joint effects of treatment assignment to either E regimen vs placebo and prespecified abnormal levels of each coagulation factor on relative risk of VTE were estimated by logistic regression adjusting for age, race, body-mass index and type of E regimen.

Results. Low levels of protein C and free protein S (<5th percentile), high D-dimer (top quartile), and high plasmin antiplasmin complex (PAP) and prothrombin fragment 1–2 (top decile) were all associated with risk of VTE with adjusted odds ratios (95% CI) of 2.0 (1.0–4.1), 2.9 (1.5–5.6), 2.8 (2.0–4.0), 2.5 (1.6–4.0) and 1.9 (1.2–3.1), respectively. Elevated factors II, VIII, IX and fibrinogen were not VTE risk factors. Compared to women with normal coagulation marker levels assigned to placebo, the joint odds of VTE with either E regimen plus an abnormal coagulation marker were more than additive compared to the separate effects of hormones and coagulation abnormalities for low protein C, low free protein S, and elevated D-dimer, PAP and F1–2. The odds ratios of VTE with the combination of an abnormal coagulation factor and assignment to hormones were (in order listed in prior sentence), 4.5 (95% CI 2.0–10.2), 6.7 (3.0–14.5), 6.1 (3.7–10), 5.8 (3.2–10.5) and 4.4 (2.4–7.7).

Conclusions. We report new findings of elevated F1-2 and PAP as VTE risk factors in women in this prospective study nested in trials of E or E+P versus placebo. Protein C or S values below the 5th percentile were also clinically relevant even though they do not represent inherited deficiency. Lower protein C and free protein S, and higher D-dimer, F1-2 and PAP all identified women at increased risk of VTE with hormones. If our findings are confirmed in management studies, measurement of these factors might assist women with decision-making on safety of E or E+P.

Author notes

Disclosure: No relevant conflicts of interest to declare.