WHI Analyzes Cardiovascular Risk by Age and Years Since Menopause

Abstract & Commentary

By Leon Speroff, MD, Editor, Professor of Obstetrics and Gynecology, Oregon Health and Science University, Portland, is Editor for OB/GYN Clinical Alert.

Synopsis: The risk of cardiovascular disease associated with postmenopausal hormone therapy is found only in older women.

Source: Rossouw JE, et al. Postmenopausal hormone therapy and risk of cardiovascular disease by age and years since menopause. JAMA. 2007;297:1465-1477.

The Women's Health Initiative (WHI) investigators conducted a secondary analysis of the two canceled clinical trial arms.1 The results in the estrogen-only arm, the combined estrogen-progestin arm, and with the participants combined were separated into age groups at randomization (50-59, 60-69, and 70-79) and according to years since menopause (< 10, 10-19, and 20 or more). An increased risk for coronary heart disease (CHD) was present only in the oldest women in the trials. The only statistically significant reduced risk was for total mortality in women age 50-59.

Statistically Significant Increased Risks
All subjects combined:
HR=1.28 (CI=1.03-1.58) in women 20+ years since menopause
HR=1.50 (CI=1.17-1.92) in women ages 60-69
HR=1.77 (CI=1.05-2.98) in women < 10 years since menopause
  Stroke HR=1.62 (CI=1.15-2.27) in women ages 60-69
  CHD HR=1.48 (CI=1.04-2.11) in women ages 70-79
    HR=1.66 CI=1.14-2.41) in women 20+ years since menopause
Table Adapted by L. Speroff

The authors calculated absolute risks and found no increases for CHD, stroke, or total mortality in women ages 50-59. In fact, only the increase in CHD events in women 20+ years since menopause reached statistical significance. The authors concluded that there was no apparent increase in cardiovascular disease risk in treated women close to their menopause.


The data in this WHI report are not new. A careful reading of the original adjudicated reports reveals that the risk of CHD was present only in the oldest women in the trials.2, 3 What is new is that the WHI finally emphasizes this finding to the public, almost 5 years since the first dramatic report in July, 2002, that we remember so well.4

This report from the WHI is a complicated statistical exercise involving 137 calculations of risk and 182 analyses of subgroups. I cannot authoritatively assess the validity of these tests; however, some things do attract this clinician's eyes. First is pooling of the two canceled trials in order to achieve greater numbers a reliable and valid exercise? The women in the two trials differed considerably. The women in the estrogen-only arm had a higher prevalence of obesity, abnormal EKGs, hypertension, diabetes, hypercholesterolemia, and a previous history of cardiovascular disease. More women in the estrogen-only arm used hormone therapy previously, and for longer durations. Although I am not convinced it is appropriate to combine the subjects of the two arms; nevertheless, the combined results are not markedly different compared with the results in the two arms analyzed separately. However, because of the marked differences in patient characteristics comparing the two arms, I do not believe it is appropriate for the WHI authors to imply that a difference that is clinically important can be derived from comparing estrogen-only treatment with combined estrogen-progestin treatment.

The authors emphasize that, overall, hormone therapy increased the risk of stroke (HR = 1.32; CI = 1.12-1.56 in the combined analysis of the two arms). However, in the subgroup analyses, the only significant increase in stroke occurred in the estrogen-only arm in women ages 60-69. The only significant increase in CHD occurred in those women in their 70s. The important point to be emphasized can be found in the authors' discussion: When women with prior cardiovascular disease or those older than 60 years were excluded, the risk of stroke in women less than 10 years since their menopause was not significantly increased.

There is a statement in the methods section that drew my attention: "Due to the compressed timeline for the initial publications, 13 additional adjudicated cases each of CHD and stroke form the CEE + MPA trial were available for this analysis." Why in the world didn't the WHI investigators wait for their full results, finish their full analysis, and provide the public with accurate information? The initial public impression, reinforced by the early publications, was that the risk of cardiovascular disease was robust, and that it applied to all postmenopausal women using hormone therapy. What is the opposite of robust? Pick your word and apply it to the analyses in this current report. Even the WHI authors recognize this by calculating estimates of absolute risk and stating that there is no increase in absolute risks due to hormone therapy in women ages 50-59.

Do these results rule out the possibility that hormone therapy given to young postmenopausal women may prevent CHD? I think not. In the WHI, the number of women in this young age group was small, the duration of hormone exposure relatively short, and long-term follow-up is yet to be reported. The number of women ages 50-54 or less than 5 years since menopause was too small to allow analysis. A meta-analysis of 23 randomized hormone therapy trials concluded that treatment reduced the risk of coronary heart disease events in younger women (OD = 0.68; CI = 0.48-0.96) compared with older women (10 or more years since menopause or greater than 60 years of age).5 This is a conclusion that is less firm than at first apparent, because most of these trials were not designed to measure an endpoint of cardiovascular disease. However, another meta-analysis by the same authors concluded that hormone therapy reduced overall mortality in women with an average age less than 60.6 There is a growing story that adequate estrogen exposure prior to the onset of clinical events provides protection against cardiovascular disease.


  1. Rossouw JE, et al. JAMA. 2007;297:1465-1477.
  2. Manson JE, et al. New Engl J Med. 2003;349:523-534.
  3. Hsia J, et al. Arch Intern Med. 2006;166:357-365.
  4. Rossou JE, et al. JAMA. 2002;288:321-333.
  5. Salpeter SR, et al. J Gen Intern Med. 2006;21:363-366.
  6. Salpeter SR, et al. J Gen Intern Med. 2004;19:791-804.