A systematic review and meta-regression analysis to examine the ‘timing hypothesis’ of hormone replacement therapy on mortality, coronary heart disease, and stroke

Abstract Background The ‘Timing Hypothesis’ states that the benefits and harms of hormone replacement therapy (HRT) are related to the proximity with which it is begun following the onset of menopause. The primary aim of this analysis was to test for heterogeneity of treatment effect for HRT using Chi 2 and I 2 tests for younger versus older initiators of HRT. The secondary aim was to perform a meta-regression with mean age at trial baseline as the covariate for various outcomes. Methods Younger initiation trials were defined as those with mean age of participants 60 years. The primary endpoints included all-cause mortality, cardiac mortality, coronary heart disease (CHD) events (a composite of cardiac mortality and nonfatal myocardial (MI)), and a composite of stroke, transient ischemic attack (TIA) and systemic embolism. Results Thirty-one RCTs were identified comparing HRT users to nonusers ( n  = 40,521). There was significant heterogeneity of treatment effect between younger versus older HRT initiators for all-cause mortality (Chi 2  = 9.74, p  = 0.002, I 2  = 89.7%), cardiac mortality (Chi 2  = 4.04, p  = 0.04, I 2  = 75.2%), and CHD events (Chi 2  = 3.06, p  = 0.08, I 2  = 67.3%). Both groups experienced an increase in stroke, TIA and systemic embolism (1112/18,774 in the HRT group versus 734/18,070 in the control group; OR = 1.52; 95% confidence interval (CI) = 1.38–1.67). When performing the meta-regression, as age increased the treatment effect of HRT was increased for stroke, TIA and systemic embolism (point estimate 0.006 with a standard error of 0.002) ( p  = 0.0003). Conclusion Younger initiation of HRT may be effective in reducing death and cardiac events. However, younger HRT initiators remained at an increased risk of stroke, TIA and systemic embolism and this risk increased as average age increased. Younger menopausal women using HRT to treat vasomotor symptoms do not appear to be at an increased risk of dying or experiencing CHD events.