Abstract 3613: Lipid Biomarkers, Hormone Therapy, and the Risk of Unprovoked Venous Thromboembolism in Women
Background: Published reports of a relationship between lipid biomarkers and venous thromboembolism (VTE) are inconsistent and controversial, and many do not account for potential effect modifiers such as hormone therapy (HT).
Methods and Results: Among 27,082 initially healthy women followed prospectively (median, 11.4 years) for incident unprovoked VTE, we measured a full panel of lipid biomarkers, including total cholesterol, high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol, triglycerides, and apolipoproteins A-I (apo A-I) and B100. In analyses adjusted for age and body mass index (kg/m2), only HDL-C and apo A-I showed any association with VTE (N=158; hazard ratio (95% CI) per 1 unit SD, 1.17 (1.00 –1.37) and 1.29 (1.11–1.51), respectively). However, after stratifying by HT use at the time of blood collection, none of the lipid biomarkers was associated with future VTE (N=80) in non-users, while higher levels of HDL-C and apo A-I were associated with increased risk of VTE (N=78) among users. Specifically, among HT users the adjusted hazard ratios for future VTE were 1.29 (1.05–1.58) and 1.41 (1.13–1.75) per 1 unit SD increase in HDL-C and apo A-I, respectively (each P≤0.01; Table⇓). In models further adjusted for factor V, prothrombin, and methylenetetrahydrofolate reductase genotypes, risk estimates did not change substantially, suggesting that the differences seen with HT were not due to any confounding by these genotypes.
Conclusions: We observed little evidence that any lipid marker predicted risk of future VTE among non-users of HT. The observation that among HT users high levels of HDL-C and apo A-I relate to risk of VTE likely reflects concomitant prothrombotic effects of hormone therapy, rather than direct effects of HDL-C or apo A-I.