Review of 28 Studies Links HT to Higher Stroke Risk


Hormone therapy is associated with a significantly increased risk of stroke, based on studies involving nearly 40,000 patients.

A review of 28 studies ranging in size from 59 to 16,608 adults and with follow-up times of 0.7–6.8 years showed a significant association between HT use and an increased risk of total stroke, with an odds ratio of 1.29. The review supports previous studies that showed an association between increased risk of stroke and hormone therapy, reported Philip Bath, M.D., and Laura Gray of the University of Nottingham, Nottingham, England (BMJ [Epub ahead of print], January 2005. Article DOI number: 10.1136/bmj.38331.655347.8F. Available from:

Twelve studies included women taking estrogen only; 16 included women taking estrogen plus progesterone. The average ages ranged from 55 to 71 years, and three studies of estrogen combined with progesterone included men. All but five studies were placebo-controlled, and 11 small trials recorded no stroke events.

Overall, 2% of patients randomized to no HT suffered strokes, but the risk of stroke among women randomized to HT increased 29%, primarily because of the increase in ischemic stroke.

In addition, the severity of stroke increased with HT use; the chance of a poor functional outcome, defined as either death or disability and dependency, was 56% higher among women randomized to HT.

In particular, HT use was associated with a significant increase in the risk of ischemic stroke in 16 studies (OR 1.29). HT use also was significantly associated with an increased risk of nonfatal stroke in 21 studies (OR 1.23), and with an increased risk of stroke leading to death or dependency in 14 studies (OR 1.56).

HT was not significantly associated with hemorrhagic stroke or transient ischemic attacks in 17 studies and 22 studies, respectively. However, none of the studies showed any association between HT and a reduction of stroke risk, despite data from previous observational studies suggesting a protective effect of HT against cerebrovascular events.

The investigators suggested that phytoestrogens, which were not included in these studies, may have yielded different results.

They also noted that the estrogen dose in several trials was higher than the standard starting doses of conjugated equine estrogen and estradiol in Great Britain, 0.625 mg and 1 mg, respectively, and studies of lower doses may yield different results.

Next Article: