Abstract 10: Lipid Biomarkers Predict the Effect of Hormone Therapy on Coronary Heart Disease Risk: A Secondary Analysis of the Heart and Estrogen/Progestin Replacement Study
Introduction: Two large randomized trials (HERS and WHI) showed that among postmenopausal women, combined estrogen and progestin hormone therapy (HT) use does not on average reduce the incidence of coronary heart disease (CHD). Previous analyses have suggested baseline Lp(a) and LDL cholesterol modify the effect of HT on CHD risk.
Hypothesis: Stratifying women based on a combination of their baseline Lp(a) and LDL levels may identify women whose CHD risk is likely to increase or decrease from the use of HT.
Methods: We performed subgroup analyses by baseline LDL and Lp(a) levels of the effect of HT on CHD in HERS. Prior to the development of risk categories, we randomly split the HERS study population (n=2763) into equal-sized derivation and validation cohorts. In the derivation cohort, we assessed how 10-point differences in baseline LDL and Lp(a) concentrations influenced the effect of HT on CHD risk. Based on these observations, we created criteria for categorizing patients into four risk categories then tested the categories in the validation cohort.
Results: In the validation cohort, for patients meeting benefit criteria, average risk criteria, and increased risk criteria, the hazard ratios (HR) for HT on CHD events were 0.55, 0.76, and 1.83 respectively. A subset of the increased risk group meeting the contraindication criteria showed a HR for HT on CHD of 5.3 and a number needed to treat to cause one CHD event of four.
Conclusions: There appears to be a population with substantially increased risk of CHD from HT. There also appears to be a population who benefits from HT as secondary prevention for CHD. There may be clinical utility in testing lipid levels before initiating HT for symptomatic treatment of menopausal symptoms or testing Lp(a) in women with a previous CHD event and offering HT to those who meet the benefit criteria. Investigating whether the decreased risk seen in women who meet the benefit criteria proves to be similar in women without previous CHD could yield valuable information with important public health impact.
Author Disclosures: P. Curl: None.
- © 2015 by American Heart Association, Inc.