Latest HRT trial results show risk of dementiaBMJ 2003; 326 doi: https://doi.org/10.1136/bmj.326.7401.1232-b (Published 05 June 2003) Cite this as: BMJ 2003;326:1232
Postmenopausal women who take combined oestrogen and progestogen hormone replacement therapy (HRT) face twice the risk of developing dementia than women who do not, a new study says (JAMA 2203;289:2651-62).
The finding deals a fresh blow to the practice of routine hormonal supplementation in postmenopausal women and contradicts previous observational and epidemiological studies that indicated that oestrogen replacement therapy could protect against Alzheimer's disease and vascular dementias (JAMA 2002;288:2123-9, 2170-1).
As reported in last week's BMJ (31 May, p 1161) the drug company Wyeth, which funded the study, has admitted that it secretly briefed a number of medical societies in advance of the study's publication, knowing it was going to be potentially damaging for the company.
The study investigated the effect of HRT in preventing dementia. Researchers from the Women's Health Initiative's memory study followed 4532 postmenopausal women aged 65 years or older who were free of probable dementia at the start of the study in 1996. The study controlled for history of diabetes and for the use of statins, which are thought to be protective against vascular dementia.
The researchers used the mini-mental state examination to screen for dementia, adjusting the scores for educational level. Initially they used a score of 72 or lower as a cut off for women who had eight or fewer years of schooling and a score of 76 for women with more than eight years. After 16 months this protocol was altered to increase the sensitivity of detection of dementia dementia, with new cut off points of 80 and 88.
Women with low scores underwent further neuropsychological tests and screens for possible confounders of dementia, such as depression, alcohol misuse, syphilis, vitamin B12 deficiency, and hypothyroidism. Women suspected of having dementia were further examined by a consensus panel of three experts consisting of two certified neurologists and one psychiatrist. Screening for dementia continued annually throughout the study period.
The women, who were recruited from 39 clinical centres, were randomised into either a placebo group (n=2303) or an HRT group (n=2229). Women in the HRT group were given one tablet a day containing 0.625 mg of conjugated equine oestrogen and 2.5 mg of medroxyprogesterone acetate, a combination sold under the brand name Prempro (Premarin and Provera).
The study continued until July 2002, when it was halted because of concerns that the cardiovascular risks and the risks of cancer of the uterus and breast from taking combined HRT exceeded the benefits.
Overall, 61 women were given a diagnosis of probable dementia: 40 in the HRT group and 21 in the placebo group. The hazard ratio for probable dementia was 2.05 (95% confidence interval 1.21 to 3.48; P=0.01).
Alzheimer's disease was the most common dementia in both groups, affecting 20 women in the HRT group and 12 women in the placebo group. Thirty nine women in the HRT group and 36 women in the placebo group had a stroke during the follow up period. Mild cognitive impairment, believed to be a precursor of dementia, did not differ between the groups (hazard ratio 1.07 (0.74 to 1.55).
It is unclear why HRT should increase the risk of dementia, but researchers speculate that it may be due to an increase in lacunar strokes. Another study in the same issue of JAMA (289:2673-84) analysed the effect of combination HRT on the risk of stroke and found that women who took HRT had a 31% (2% to 68%) higher risk of ischaemic and haemorrhagic stroke. A study evaluating the effect of oestrogen supplementation alone is ongoing.