The first "free" report was conducted by examining US Centers for Medicare & Medicaid Services (CMS) coding files from 1999 to 2005 (before the big ICD recalls). Patients records included were 65 years or older with Medicare fee-for-service coverage and diagnosed with acute myocardial infarction and either heart failure or cardiomyopathy (based on Medicare diagnosis codes) but no prior cardiac arrest or ventricular tachycardia (ie, the primary prevention cohort [n = 65,917 men and 70,504 women]), or with cardiac arrest or ventricular tachycardia (ie, the secondary prevention cohort [n = 52,252 men and 47,411 women]). This trial was limited by the lack of clinical data (no measures like ejection fraction or confirmation of arrhythmias occured - only diagnosis codes were used) and so was limited at the outset. But despite its limitations, no mater how it wsa cut in multivariate analyses, men were more likely than women to receive ICD therapy (hazard ratio [HR], 3.15; 95% confidence interval [CI], 2.86-3.47).
The second trial, conducted by the same Duke researchers, examined actual clinical data from a voluntary "Get with the Guidelines" heart failure quality improvement initiative called OPTIMIZE-HF and supported by GlaxoSmithKline (makers of a major heart failure drug, cardevilol, marketed as Coreg® and Coreg CR®) and the American Heart Association. It occurred more recently (2005-2007: after the defibrillator recall period). In this study, left ventricular ejection fraction data, paired with diagnosis codes, were studied. It too, demonstrated that women were 40% less likely than men to receive an ICD.
So why the difference?
Well, it's tough to say for sure. One item discussed in the papers was the increased likelihood for women to have a form of "impaired relaxation" heart failure (rather than from systolic dysfunction), but the second trial used only people with ejection fractions (the amount of blood squeezed from the heart with each heartbeat which is normally greater than 50%) that were less than 30%. It seems ICDs were still more common in men than women.
Another reason might be because of confounding factors, like women may have more co-morbidities, especially since they usually get older than men. After all, controlling for everything in these observational studies is difficult. But the first trial still revealed a propensity for men to receive an ICD in both the under-75 crowd or the over-75 crowd.
Perhaps (and I say this reluctantly) it is because doctors are sexist. Especially male doctors. Could it as simple as that? Probably not. But men still outnumber women in cardiovascular subspecialties and we might just not be thinking that women get heart disease. Also, most of the major ICD trials had more men than women, so doctors might have a bias against referring a female rather than a male.
These findings raise all kinds of questions. It would be interesting to see if others out there in the blog-o-sphere have some other ideas.
But one thing is for sure, you can bet Medtronic (also tacitly a research contributor to these studies) and the other ICD manufacturers will have an e-mail in my inbox tomorrow with these findings.
Lesley H. Curtis, PhD; Sana M. Al-Khatib, MD, MHS; Alisa M. Shea, MPH; Bradley G. Hammill, MS; Adrian F. Hernandez, MD, MHS; Kevin A. Schulman, MD. "Sex Differences in the Use of Implantable Cardioverter-Defibrillators for Primary and Secondary Prevention of Sudden Cardiac Death." J Am Med Assoc 2007;298:1517-1524.
Adrian F. Hernandez, MD, MHS; Gregg C. Fonarow, MD; Li Liang, PhD; Sana M. Al-Khatib, MD, MHS; Lesley H. Curtis, PhD; Kenneth A. LaBresh, MD; Clyde W. Yancy, MD; Nancy M. Albert, PhD; Eric D. Peterson, MD, MPH. "Sex and Racial Differences in the Use of Implantable Cardioverter-Defibrillators Among Patients Hospitalized With Heart Failure" J Am Med Assoc 2007;298:1525-1532.
2 Oct 2007 1930PM CST Addendum: An accompanying editorial in JAMA looks at the gender issue as a glass half full rather than half empty:
(HeartWire) Dr Rita Redberg (University of California, San Francisco), who wrote an accompanying editorial, told heartwire: "They've definitely established that white men are more likely to get ICDs, but we all know that defibrillators are not the only thing that white men are more likely to get." In her editorial, she says, "The results are troubling, but not for the expected reasons. In other words, the bad news may not be for women and minorities, but for white men who are undergoing a procedure that, for primary prevention, has not been shown to extend their lives. By reporting the first outcomes data for ICDs in the Medicare population, the study by Curtis et al should stimulate national dialogue on this crucial question."Her issue was with the primary prevention cohort, she had no issue with the secondary prevention cohort.