Stroke: estradiol increased endometrial abnormalities without clearly reducing stroke or death in postmenopausal women

Clinical bottom line (level 1b)

  1. Postmenopausal women with a previous stroke or TIA who took estradial compared with placebo were not clearly less likely to have another stroke or die, but were more likely to develop vaginal bleeding (NNH = 4 at 12 months) or endometrial hyperplasia (NNH = 11 at 12 months) .
Viscoli et al: N Engl J Med 2001; 345 : 1243-1249
Expires June 2004

The study

Double-blinded ?concealed randomised trial with intention-to-treat
Setting: 21 acute hospitals, USA

664 patients (aged 46 to 91; mean 71, 100% male) postmenopausal women (amenorrhoea for 12 months or post-hysterectomy) within 90 days of an ischaemic stroke or transient ischaemia attack

Excluded if
  • aged < 44
  • index event disabling (> 5 on NASCET scale)
  • history of breast or endometrial cancer
  • venous thromboembolic event on estrogen-replacement therapy
  • neurological or psychiatric disease complicating evaluation of end-points
  • co-existing condition likely to limit life expectancy
  • already on estrogen

Control Group: (n = 327, 327 analysed): placebo
Experimental Group: (n = 337, 337 analysed): estradiol 17-beta 1 mg daily

100% followed for 12 months

The evidence

Outcome Time to outcome CER EER RRR
(95% CI)		 ARR
(95% CI)		 NNT
(95% CI)
death or non-fatal stroke 12 months 93
(28.4%)		 99
(29.4%)		 -3%
(-31% to 19%)		 -0.94%
(-7.83% to 5.96%)		 -110
(NNT = 13 to infinity;
NNH = 17 to infinity)
vaginal bleeding 12 months 33
(10.9%)		 115
(34.1%)		 -240%
(-380% to -140%)		 -24.0%
(-30.1% to -18.0%)		 -4
(-6 to -3)
endometrial hyperplasia 12 months 3
(0.92%)		 34
(10.1%)		 -1000%
(-3400% to -240%)		 -9.17%
(-12.6% to -5.79%)		 -11
(-17 to -8)

Comments

  1. Patients were stratified by centre and base-line risk of further stroke before randomisation in blocks of four.
  2. Patients were followed for a mean of 33 months.

Citation

  1. Viscoli CM, Brass LM, Kernan WN, et al: a clinical trial of estrogen-replacement therapy after ischemic stroke. N Engl J Med 2001; 345 : 1243-1249
Search Terms: from ACP Journal Club
Contributor: Chris Ball, June 2002
Reviewer:

Clinical Question.
Patient postmenopausal women with a previous stroke
Intervention or Exposure estradiol
Comparison placebo
Outcome death, stroke, vaginal bleeding