Estrogen therapy and thrombotic risk

Abstract

Post-menopausal hormone therapy increases the risk for venous thrombosis, and possibly myocardial infarction (MI) and ischemic stroke. However, most women using hormone therapy do not suffer thrombosis, and to date our ability to identify women at risk is limited. Thrombosis, arterial or venous, has 2 requisites: a vascular anomaly and a response of the hemostasis system to the anomaly. Consequently, experimental approaches to understand the pathophysiology of thrombosis require definition of vascular anatomy and function as well as characteristics of the blood within the context of genetic background, lifestyle choices and environmental exposures, which influence gene expression. Defining interactions among factors that affect individual propensity to thrombosis will allow physicians to better identify at-risk individuals, for example a woman contemplating estrogen therapy for symptoms of menopause, and prevent adverse thrombotic events.

Introduction

Postmenopausal hormone treatment is an established risk factor for venous thromboembolism (VTE) and may increase myocardial infarction (MI) and stroke in older women (Perez Gutthann et al., 1997, Varas-Lorenzo et al., 1998, Varas-Lorenzo et al., 2000, Writing Group for the Women's Health Initiative Investigators, 2002). However, the vast majority of women so treated do not develop thrombosis. Despite decades of research into the characterization of how hormone treatments affect proteins of the hemostasis system, there still is no clinically viable method to establish thrombosis risk for a woman contemplating hormone therapy (for review, see Cosman et al., 2005). Propensity for thrombosis varies widely among individuals but is independent of vascular disease. The term “thrombophilia” has come to be used in this context. In principle, preventing thrombosis by avoiding or modifying risk exposures among individuals with thrombophilia could have the same outcome as preventing the progression of vascular disease. Strategies for prevention in clinical practice will depend on the morbidity associated with the intervention versus the effect on outcome. That is, the effectiveness of an intervention will be context-sensitive. This concept is exemplified by a recent study that identified age and sex as modifiers of effectiveness of aspirin as primary prophylaxis against MI and stroke: aspirin was effective in preventing MI independent of age in men, but prevented stroke only in older women (Ridker et al., 2005a). The challenge for the clinician-scientist is to define the context. For example, do women with arterial disease who experience thrombosis have features in common with those who experience venous thrombosis? Do drugs or interventions that increase risks of thrombosis, like hormone therapy, move characteristics of the hemostasis system in healthy individuals toward high-risk characteristics? Or is there a genotype that defines high risk and are there stimuli which affect expression of a high risk phenotype? This review will approach the context of thrombotic risk (propensity) in post-menopausal women by evaluating effects of estrogen on characteristics of the vascular wall and coagulability of the blood, by evaluating genetic characteristics of estrogen receptors and coagulation proteins and finally by considering how infection-induced inflammation, as a prototype for environmental exposure, might modulate both characteristics of the vessel wall and the blood toward a thrombogenic phenotype.