Background: There is apparently compelling evidence, from observational studies, that hormone replacement therapy (HRT) may have benefits in reducing cardiovascular events in post-menopausal women. However, these observational data are subject to biases and confounding and require support from formally designed randomised controlled trials of the effects of HRT on cardiovascular disease risk.
Objectives: To assess the effects of HRT for the primary and secondary prevention of cardiovascular diseases in post-menopausal women.
Search strategy: We searched MEDLINE (1998 to December 2002)), EMBASE (1998 to December 2002), the Cochrane Controlled Trials Register (CCTR) (Issue 4 2002), the National Research Register (1998 to present), Clinical Trials.gov (1998 to present), and the database of Spanish Clinical Trials (1998 to present) and reference lists of articles.
Selection criteria: Randomised controlled trials comparing HRT with controls (placebo or no treatment) with a minimum follow up of 6 months for treating or preventing cardiovascular disease in postmenopausal women with or without cardiovascular disease.
Data collection and analysis: Three independent reviewers extracted information from the articles, solving discrepancies by consensus. All outcomes studied were dichotomous. Risk ratios and 95% confidence intervals (CI) were calculated for each study and plotted. Random effects meta-analysis was used in efficacy outcomes (cardiovascular events) and fixed-effects meta-analysis in variables regarding side effects (deep venous thrombosis).
Main results: No protective effect of HRT was seen for any of the cardiovascular outcomes assessed: all cause mortality, cardiovascular death, non-fatal MI, venous thromboemboli or stroke. Higher risks of venous thromboembolic events (Relative risk (RR) 2.15, 95% CI 1.61 to 2.86), pulmonary embolus (RR 2.15, 95% CI 1.41 to 3.28), and stroke (RR 1.44, 95% CI 1.10 to 1.89) was found in those randomised to HRT compared with placebo. No substantial heterogeneity (p <0.1) was detected in any of the outcomes studied.
Authors' conclusions: At present, a recommendation for initiating HRT for the reason of preventing cardiovascular events in post-menopausal women (with or without cardiovascular disease) should not be made. Women with other risk factors for venous thromboembolic events should be discouraged from using HRT if the sole goal is to prevent cardiovascular events.