Barrett-Connor Elizabeth
Division of Epidemiology, Department of Family and Preventive Medicine, School of Medicine, University of California, San Diego, CA, USA.
Am J Epidemiol. 2007 Sep 1;166(5):506-10. doi: 10.1093/aje/kwm214.
Largely on the basis of results from meta-analyses of observational studies, postmenopausal estrogen was widely prescribed to prevent coronary heart disease. However, epidemiologic studies, no matter how consistent and coherent, are not sufficient to recommend mass preventive therapy to healthy women. In fact, all three large clinical trials failed to confirm estrogen's expected cardiac protection. The most persistent explanatory hypothesis for the "trial failure" was the age of the participants, based on the thesis that estrogen in recently menopausal women could prevent the development of coronary artery plaque but, given to older women with vulnerable plaque, would have a null or even harmful effect. The timing hypothesis is plausible, but the prespecified subgroup analyses in both Women's Health Initiative trials showed no significant interaction with age or years since menopause. The best opportunity to test the timing hypothesis was lost when 1,000 Women's Health Initiative women younger than 60 years had coronary artery calcium scans to evaluate the effect of estrogen on plaque burden, but no women 60 years or over were similarly examined. Therefore, this ancillary study can examine the effect of estrogen treatment on coronary calcium in women younger than 60 years but will not be able to determine if the effect is different in older women. In the meantime, publicized statements in multiple venues have promoted the timing hypothesis as fact, confusing patients and physicians who do not realize that the hypothesis is stronger than the evidence.
主要基于观察性研究的荟萃分析结果，绝经后雌激素被广泛用于预防冠心病。然而，流行病学研究无论多么一致和连贯，都不足以向健康女性推荐大规模预防性治疗。事实上，所有三项大型临床试验都未能证实雌激素预期的心脏保护作用。对于“试验失败”最持久的解释性假设是参与者的年龄，其依据是绝经后不久的女性使用雌激素可以预防冠状动脉斑块的形成，但对于有易损斑块的老年女性使用雌激素则没有效果甚至有害。时间假设看似合理，但两项妇女健康倡议试验预先设定的亚组分析均未显示与年龄或绝经后年限有显著相互作用。当1000名60岁以下的妇女健康倡议参与者进行冠状动脉钙化扫描以评估雌激素对斑块负荷的影响时，最佳的检验时间假设的机会丧失了，因为没有对60岁及以上的女性进行类似检查。因此，这项辅助研究可以研究雌激素治疗对60岁以下女性冠状动脉钙化的影响，但无法确定在老年女性中效果是否不同。与此同时，多个场合的公开声明将时间假设当作事实进行宣扬，使那些没有意识到该假设比证据更有力的患者和医生感到困惑。
