Postmenopausal Hormone Therapy and Stroke
Abstract & Commentary
Leon Speroff, MD, Professor of Obstetrics and Gynecology, Oregon Health Sciences University, Portland, is Editor for OB/Gyn Clinical Alert
Synopsis: A meta-analysis concluded that postmenopausal hormone therapy increases the risk of ischemc stroke.
Source: Bath PMW, Gray LJ. Br Med J. doi:10.1136/bmj.38331.655347.8F (published online January 7, 2005).
Bath and Gray from the Division of Stroke Medicine in the Queen’s Medical Centre, University of Nottingham, performed a meta-analysis of the published literature regarding the association between postmenopausal hormone therapy and risk of stroke. A total of 28 trials were analyzed, with 39,769 subjects. The following odds ratios were calculated:
| No. of Events | OR and CI | |
| Total stroke | 940 | 1.29 (1.13-1.47) |
| Stroke death/disability | 145 | 1.56 (1.11-2.20) |
| Ischemic stroke | 443 | 1.29 (1.06-1.56) |
| Hemorrhagic stroke | 63 | 1.07 (0.65-1.75) |
| Transient ischemic attacks | 233 | 1.02 (0.78-1.34) |
Thus this study concluded that postmenopausal hormone therapy increases the risk of ischemic stroke, and women who experience stroke and are using hormone therapy have a worse outcome.
Comment by Leon Speroff, MD
The results of this meta-analysis agree with the Women’s Health Initiative (WHI) that reported a relative risk for ischemic stroke in the estrogen-progestin arm of 1.44 (1.09-1.90) and in the estrogen-only arm, 1.39 (1.05-1.84), except for the WHI finding of no increase in fatal strokes.1-3 Examining closely the figure in the meta-analysis that illustrates each study with its odds ratio and confidence interval, only the results from the WHI were statistically significant. In every other study, the confidence interval includes 1.00 and, thus, was not significant. Bath and Gray assigned a weight to each study; the weights assigned to the WHI reports overwhelm the others by a wide margin.
Of course by now, clinicians are aware that both WHI clinical trial arms consisted of older women with a significant proportion having established atherosclerosis. Thus at best, this meta-analysis supports the conclusion that hormone therapy for secondary prevention of cardiovascular disease is ineffective, and may carry an increased risk (a risk that is perhaps limited to the very oldest of women). This meta-analysis also supports the conclusion that hormone therapy does not reduce the risk of stroke; however, with the exception of the WHI (and its older population), the reviewed clinical trials were relatively short in duration, perhaps too short to yield a benefit.
This is another example of the harm that can follow the exercise of meta-analysis. The studies do not support the conclusion that the use of hormone therapy increases the risk of ischemic stroke in all postmenopausal women. The adverse effect is limited to data derived from elderly women in the WHI study, and the WHI study used a dose of estrogen that may have been inappropriately high for these older women. Bath and Gray should have saved themselves their time and effort; this meta-analysis adds nothing to what is already known from the WHI.
The observational studies on hormone therapy and risk of stroke have not been consistent. Some have reported a small increase, others either no effect or a reduction in risk. The cohort studies, however, that have assessed long duration of hormone use have reported a reduction in stroke incidence and stroke mortality (eg, the National Health and Nutrition Examination Survey).4
We should not prescribe hormone therapy to women with cardiovascular disease in the expectation that recurrent vascular events will be prevented. But whether hormone therapy prescribed to relatively young and healthy postmenopausal women provides primary prevention of stroke, remains to be properly documented by the appropriate epidemiologic studies.
References
- Writing Group for the Women’s Health Initiative Investigators. JAMA. 2002;288:321.
- Wassertheil-Smoller S, et al. JAMA. 2003;289:2673.
- The Women’s Health Initiative Steering Committee, JAMA. 2004;291:1707.
- Finucane FF, et al. Arch Intern Med. 1993;153:73.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content