A new meta-analysis finds that ICDs are not associated with a mortality benefit in women. The study, by Ghanbari and colleagues, appears in the Archives of Internal Medicine.
934 women and 3,810 men were randomized in the 5 trials included in the meta-analysis. For men, the trials, both individually and in the combined analysis, resulted in a statistically significant overall reduction in mortality.
“Most clinical trials have been heavily weighted toward men; therefore, generalization of the results to women remains questionable. The best answer to this problem would be to perform a clinical trial that specifically targets women with heart failure to test the hypothesis of whether implantable cardioverter-defibrillator implantation reduces their overall mortality [death] rate,” write the authors.
In an accompanying editorial Rita Redberg laments the paucity of data in women:
Approximately 30 percent of implantable cardioverter-defibrillator recipients are women. However, data supporting the efficacy of implantable cardioverter-defibrillators for primary prevention in women is sparse. In other words, implantable cardioverter-defibrillators are being implanted in hundreds of thousands of women without substantial evidence of benefit, apparently based on the assumption that, to paraphrase the old saying, ‘What’s good for the gander is good for the goose.’
CardioBrief solicited comments on this analysis from the two best-known cardiologists in the blogosphere, each of whom has plenty of experience with ICDs. Both cardiologists strongly rejected a take-away message hinting that women should not be candidates for ICDs.
Here’s the comment from Dr Westby Fisher, an EP at NorthShore University HealthSystem in Evanston, IL and a Clinical Associate Professor of Medicine at University of Chicago’s Pritzker School of Medicine. He is the author of the popular and frequently-cited blog, Dr. Wes, which he describes as the “musings in the life of an internist, cardiologist and cardiac electrophysiologist.” He also sells t-shirts for charity at MedTees.com.
Ghanbari et al’s meta-analysis on the use of implantable cardiac defibrillators for the prevention of sudden cardiac death with advanced heart failure erroneously concludes “Implantable cardioverter-defibrillator therapy for the primary prevention of sudden cardiac death in women does not reduce all-cause mortality.” The study correctly notes that women were underrepresented in all of the five trials they were able to identify that addressed the issue, but like all meta-analyses performed on the basis of literature reviews, their paper suffers from selection biases, measurement biases, and intervention biases. Further, since each study they included in their analysis was underpowered to draw conclusions regarding the treatment differences between men and women. Instead, the authors should have concluded that the selection bias toward women in cardiovascular trials studying ICDs needs further investigation.
This study suffers from the usual problems of meta-analysis, and perhaps from more than the usual problems. Rather than enumerating those problems, I’ll just note that a meta-analysis such as this can be used to generate hypotheses, but ought not to be used to make definitive conclusions.
So it is not appropriate to conclude (as this study does) that “ICD therapy for the primary prevention of SCD in women does not show any benefit to all-cause mortality.” A more appropriate conclusion would be that, in the pooled data from the women enrolled in the studies selected for this meta-analysis, no overall improvement in all-cause mortality could be detected.
Women with heart failure are different from men with heart failure. They tend to be older, they have more hypertension and diabetes (and the medical problems that go along with diabetes), and they live longer than men from the time of the diagnosis of heart failure. It is not clear that their risk of sudden arrhythmic death is the same as it is for men with heart failure.
But ICDs prevent sudden arrhythmic death equally well in men and women. So it is not appropriate to conclude that ICDs do not “work” in women, but rather, that the relative risk of arrhythmic may be different in women. It is entirely possible, and even likely, that appropriately selecting which women with heart failure ought to receive ICDs (i.e., determining which individuals have a relatively high risk of sudden arrhythmic death) would be a different process than for men.
This meta-analysis ought to stimulate more research to learn how to better identify those women with heart failure whose risk of sudden arrhythmic death significantly threatens their overall survival. I fear that, instead, it will be used merely as an excuse to withhold even more ICDs from women with heart failure.
I will not hold my breath waiting for an editorial pointing out that, since “none” of the women in these trials benefitted from ICDs, their inclusion in these trials “diluted” the benefit ICDs gave to the men – and that, therefore, the “number needed to treat” calculations for the men in these trials must be substantially lower than previously believed. (So, it must be all the more important to throw ICDs in men with heart failure.)
Here is the press release from the AMA:
Implantable Defibrillators Not Associated With Reduced Risk of Death in Women With Heart Failure
CHICAGO – Implantable cardioverter-defibrillators do not appear to be associated with a reduced risk of death in women with advanced heart failure, according to a meta-analysis of previously published research in the September 14 issue of Archives of Internal Medicine, one of the JAMA/Archives journals.
Heart failure affects about 5.3 million Americans, almost half of them women, according to background information in the article. Patients with heart failure are six to nine times more likely than the general population to experience sudden cardiac death. In addition to medication, treatment for heart failure involves preventing sudden cardiac death through the implantation of a cardioverter-defibrillator. This therapy is supported by numerous clinical trials, the authors note. However, questions remain regarding the benefits in certain patient subgroups, including women.
Hamid Ghanbari, M.D., and colleagues at Providence Hospital Heart Institute and Medical Center, Southfield, Mich., searched for randomized clinical trials of implantable defibrillator therapy for heart failure patients published between 1950 and 2008 that included data on the risk of death for female patients. Five eligible trials that included 934 women were identified.
None of the five trials demonstrated a significant benefit of defibrillator implantation over medical therapy for women. When the researchers pooled the data and performed a meta-analysis, the implantable cardioverter-defibrillator was not associated with decreased all-cause mortality in women. Among the 3,810 men in the studies, however, a statistically significant decrease in death rate was found in each of the five trials alone and in the combined meta-analysis.
There are several possible reasons for the sex differences in these results, the authors note. Among patients with heart disease, women have about one-fourth the risk of sudden cardiac death as men. This may be because women have different patterns of arrhythmias and also because they have more co-occurring illnesses that may increase their risk of death from other causes. Therefore, a larger study population may be needed to show any benefit of defibrillator implantation in women.
“Most clinical trials have been heavily weighted toward men; therefore, generalization of the results to women remains questionable. The best answer to this problem would be to perform a clinical trial that specifically targets women with heart failure to test the hypothesis of whether implantable cardioverter-defibrillator implantation reduces their overall mortality [death] rate,” the authors write. Because clinical guidelines already recommend defibrillator treatment to prevent sudden cardiac death, such a trial may be difficult to propose, they note. “However, on the basis of our findings it seems that a trial targeting women is needed, and a meta-analysis such as ours may be an appropriate first step to explore this hypothesis.”
(Arch Intern Med. 2009;169:1500-1506.)
Editorial: What’s Good for the Gander May Not Be Good for the Goose
“Approximately 30 percent of implantable cardioverter-defibrillator recipients are women. However, data supporting the efficacy of implantable cardioverter-defibrillators for primary prevention in women is sparse,” writes Rita F. Redberg, M.D., of University of California, San Francisco, and editor of Archives of Internal Medicine, in an accompanying editorial. “In other words, implantable cardioverter-defibrillators are being implanted in hundreds of thousands of women without substantial evidence of benefit, apparently based on the assumption that, to paraphrase the old saying, ‘What’s good for the gander is good for the goose.’”
“It is important to know the benefits of implantable cardioverter-defibrillator use in women, especially considering the known risks of morbidity and mortality,” Dr. Redberg continues. “Ghanbari et al rightly conclude that further studies are needed. Part of the reason for the lack of sex-specific data for devices may be related to the lack of Food and Drug Administration guidance in this area. There is reason to be optimistic that this deficit will start to be corrected in the near future.”
“Until then, meta-analyses such as the one by Ghanbari et al are the best way to determine if the goose is doing as well as the gander,” she concludes.
(Arch Intern Med. 2009;169:1460-1461)
# # #