Post-menopausal hormone replacement therapy (cont'd): risk-benefit balance in the hot seat.

(1) The results of a randomised placebo-controlled double-blind trial in 16 000 women (the WHI trial), published in 2002, showed that post-menopausal hormone replacement therapy based on sulphoconjugated equine oestrogen and medroxyprogesterone led to an excess risk of serious adverse events (pulmonary embolism, coronary events, stroke, and invasive breast cancer). (2) Further analysis showed that invasive breast cancer diagnosed in women treated with the hormone combination had a similar histology and grade to cancers diagnosed in the placebo group, but that they were larger and more advanced. (3) The excess risk of stroke was mainly due to ischaemic events. In the group treated with the hormone combination, the only identifiable risk factor was lengthy use of hormone replacement therapy before enrollment in the trial. (4) No subgroup of women at risk of coronary events was found apart from women with elevated LDL-cholesterol at enrollment. (5) In the WHIMS subtrial in women aged 65 years or more, the risk of dementia was twice as high in women receiving hormone therapy as in those receiving placebo. (6) In the British ESPRIT placebo-controlled trial, oral estradiol valerate, at a dose of 2 mg/day, was ineffective as secondary prevention of myocardial infarction. (7) Follow-up of a very large British cohort (the Million Women Study) showed a significantly increased risk of breast cancer with all types of hormone replacement therapy, including oestrogen-progestin combinations (relative risk 2.00; 95% confidence interval 1.88-2.12), oestrogen monotherapy (RR 1.30; 95% CI 1.21-1.40) and tibolone (RR 1.45; 95% CI 1.25-1.68). There was no significant difference between oestradiol and equine oestrogen, between medroxyprogesterone acetate and progestins derived from nortestosterone, between oral, transdermal and implanted oestrogen preparations, or between sequential and continuous regimens. (8) A Swedish cohort study and an American case-control study also showed a higher risk of breast cancer linked to progestin use. (9) In a small French epidemiological case-control study, the risk of thromboembolism was higher among women taking oral rather than transdermal estradiol as part of their hormone replacement therapy. (10) In practice, the risk-benefit ratios of the hormone replacement therapies most commonly used in France are poorly characterised. It is therefore logical to use the drugs with which we have most experience (in clinical trials or during lengthy follow-up). The benefits and drawbacks should be regularly discussed with women using hormone replacement therapy, and they should be closely monitored for signs of breast cancer or cardiovascular disease.