Menopause Live - IMS Updates

Date of release: 12 April, 2010

Early risk of coronary heart disease for younger, postmenopausal HRT users?

The objective of a recently published analysis of the Women’s Health Initiative (WHI) data on coronary heart disease (CHD) was to estimate the effect of therapy with continuous conjugated equine estrogens (CEE) + medroxyprogesterone acetate (MPA) on risk for coronary heart disease (CHD) over time and stratified by years since menopause [1]. The investigators intended to test the ‘timing hypothesis’ by questioning whether there is an early increased risk of CHD for treated, younger, postmenopausal women and whether such a risk would ever disappear. To demonstrate the effect of treatment, they used adherence-adjusted hazard ratios (HRs) and ‘CHD-free survival curves’. The HRs for CEE + MPA therapy in women within 10 years after menopause, as compared to untreated, younger, postmenopausal women were 1.29 (95% confidence interval (CI) 0.52–3.18) for the first 2 years and 0.64 (95% CI 0.21–1.99) for the first 8 years. The authors conclude that, in women who initiated therapy within 10 years of menopause, there was no suggestion of a decreased risk for CHD within the first 2 years of treatment, and that a possible cardioprotective effect became apparent only after 6 years of use, as shown in the ‘CHD-free survival curves’ of users and non-users of hormone therapy (95% CI 2–10 years).


Comment by Ewald Boschitsch In the WHI study, 8506 women were randomly assigned to CEE + MPA therapy and 8102 women were assigned to placebo. Women within 10 years of menopause at randomization numbered 2782 and 2712 in each group, respectively [1]. In the treated group, 31 cases of CHD were observed (14 in the first 2 years), compared to 34 cases in the placebo group (12 in the first 2 years). On the basis of these figures, Toh and colleagues estimated average HRs of 0.64 (95% CI 0.21–1.99) for the first 8 years, and 1.29 (95% CI 0.52–3.18) for the first 2 years. 
In their analysis of the well-known WHI data on CHD [2, 3], Toh and colleagues used sophisticated statistical methods for adjusting for adherence to assigned therapy through ‘inverse probability weighting [4], with the goal of accommodating the variations in adherence over time and the effect of previous treatment use on subsequent adherence. This approach, as argued by the authors, also allowed a comparison of the recent re-analysis of the observational Nurses’ Health Study (NHS) data [5] with the randomized WHI trial data, which were considered to be similar. They state that the pooled data from the two studies suggest a 29% increase in CHD risk during the first 2 years of CEE + MPA use in women within 10 years of menopause, and they declare that these pooled data give ‘the most precise estimates on this topic’. But, at the same time, the authors say that, ‘This result does not attain traditional statistical significance’ and there is a ‘need to be cautious when drawing conclusions’.
The authors’ proposition of the similarity of the WHI and the NHS data, and even more the WHI findings themselves, would actually deserve a rather positive interpretation: in fact, the results show that CEE + MPA slightly reduced the risk of CHD when administered long term and initiated within 10 years after menopause. The adherence-adjusted HR of 0.64 for the first 8 years translates into 11 cases per 1000 women on CEE + MPA vs. 12.5 cases on placebo; the HR of 1.29 for the first 2 years translates into 5 cases vs. 4.5 cases, respectively. Given the small numbers of cases, the results, whether displayed as HRs or as CHD-free survival curves (95% CI 2–10 years), did of course not reach statistical significance.
In a summary for patients, published in the same issue of the Annals of Internal Medicine, the authors proclaim a possible increased risk to be present in the first 2 years in women who start hormone therapy within 10 years after menopause, and declare this increased risk to persist until about 6 years after use. They admit that the findings could have occurred by chance, but assert, as in the original paper, that ‘this information is, and will probably be for a long time, the best available evidence on this topic’. They further warn women in advance to ‘not expect hormone therapy to protect them from heart attacks’ and alert them ‘to worry about a possible slightly increased risk for heart attacks’ [6]. 
The conclusions of this recent WHI analysis are highly speculative and, to some extent, contradictory. The summary for patients might be received rather as a threatening invocation than as an objective clarification of facts. To claim to provide the best available evidence on this topic for a long time appears to be presumptuous and irresponsible.
A further comment by Robert Langer addresses some important methodological aspects
In considering the associations tested in the paper by Toh and colleagues, it is helpful to note that the WHI was designed with the expectation that a coronary benefit might be evident after approximately 9 years [7], and monitoring criteria anticipated potential cross-over at 6 years. The trends for women < 10 years since menopause (presented in Figure 2 in the paper) are consistent with that expectation, showing a cross-over toward benefit between 6 and 7 years after the start of the study [1]. 
Results presented in the paper also support the hypothesis that initiation of hormone therapy in the presence of mature atheroma can trigger coronary events early in treatment, since events within the first 2 years were more likely in older women (HR 2.36, 95% CI excluding 1.0, for the entire population, with an average age of 63 years, but HR 1.29, 95% CI including 1.0 for women within 10 years of menopause) [1]. Thus, the Toh paper affirms the concept that the timing of hormone therapy is important for coronary outcomes. 
Table 1 in Toh’s study presents data from the WHI and the NHS, which are contrasted to allow consideration of differences that may result from experimental versus observational designs. In Table 2, data from the two studies are pooled using a weighting adjustment to increase sample size and statistical power. Results using this technique should be interpreted with caution since there is no adjustment for differences in risk factors known to exist between these two populations [8]; one is an observational cohort of health professionals and the other is a clinical trial of older postmenopausal women recruited to be representative of the US population with regard to ethnicity and general health characteristics.


Ewald Boschitsch
Head of the KLIMAX Menopause Clinic, Vienna, Austria

Robert D. Langer
Principal Scientist and Medical Director, Jackson Hole Center for Preventive Medicine; Adjunct Scholar in Epidemiology, University of Pennsylvania Center for Clinical Epidemiology & Biostatistics; Professor of Family and Preventive Medicine, University of California San Diego (retired)


  1. Toh S, Hernández-Díaz S, Logan R, Rossouw JE, Hernán MA. Coronary heart disease in postmenopausal recipients of estrogen plus progestin therapy: does the increased risk ever disappear? A randomized trial. Ann Intern Med 2010;152:211-17. Published February 16, 2010.

  2. Manson JE, Hsia J, Johnson KC, et al.; Womens Health Initiative Investigators. Estrogen plus progestin and the risk of coronary heart disease. N Engl J Med 2003;349:523-34.

  3. Rossouw JE, Anderson GL, Prentice RL, et al.; Writing Group for the Womens Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Womens Health Initiative randomized controlled trial. JAMA 2002;288:321-33.

  4. Robins JM, Hernn MA, Brumback B. Marginal structural models and causal inference in epidemiology. Epidemiology 2000;11:550-60.

  5. Hernn MA, Alonso A, Logan R, et al. Observational studies analyzed like randomized experiments: an application to postmenopausal hormone therapy and coronary heart disease. Epidemiology 2008;19:766-79.

  6. Summaries for patients: The effect of estrogen plus progestin on coronary heart disease. Ann Intern Med 2010;152:I-40. Original report in Ann Intern Med 2010;152:211-17.

  7. WHI Protocol for clinical trial and observational study components, page 28.

  8. Langer RD. Re: Estrogen plus progestin therapy and breast cancer in recently postmenopausal women. Am J Epidemiol 2009;169:784-5; author reply 785-6.