Correspondence

Sex Bias in the Care of Patients with Cardiovascular Disease?

N Engl J Med 1994; 331:883September 29, 1994

Article

To the Editor:

The article by Mark et al. (April 21 issue)1 provides further evidence of variability in the diagnostic evaluation of women and men with known or suspected coronary artery disease. On the surface, the results of these authors appear to contradict data recently reported from our institution2.

In our study, 391 women and 449 men referred for the noninvasive evaluation of possible coronary artery disease had similar rates of abnormalities on stress testing, nuclear perfusion imaging, or both (19 percent vs. 22 percent, P not significant). The male cohort subsequently underwent significantly more diagnostic testing (62 percent vs. 38 percent, P = 0.002) and coronary revascularization (5 percent vs. 2 percent, P = 0.03). The women in the study had a higher two-year rate of cardiac death or myocardial infarction (7 percent vs. 2.4 percent, P = 0.002) and had more risk factors (1.3 ±1.0 vs. 1.1 ±1.0, P = 0.002).

The Duke data-base study enrolled women who were considered by their physicians to have a lower probability of coronary artery disease before testing than the men; 27 percent of the men and 12 percent of the women had abnormal initial noninvasive-test results. Whereas patients with a history of myocardial infarction were excluded from our study, 26 percent of men and 15 percent of women studied at Duke reported a history of myocardial infarction; 13 percent and 6 percent, respectively, had electrocardiographic Q waves. The population studied at Duke contained a relatively small percentage of patients with diabetes mellitus as compared with our study (men, 8 percent vs. 23 percent; women, 9 percent vs. 27 percent). The probability of coronary artery disease before testing thus differed between these two studies and may have contributed to differences in outcomes of testing.

Rates of cardiac catheterization (as a function of the percentage of patients with abnormal initial test results) were 48 percent for men and 44 percent for women at Duke and 45 percent and 34 percent, respectively, at our hospital. Our follow-up data indicating worse outcomes in women suggest that this medical resource may have been underused in the female cohort.

It is interesting but somewhat disconcerting to observe such variability in the results of separate studies dealing with the same important question. The differences between studies noted above may account for the apparent lack of concordance between two studies reported within one month from experienced tertiary-referral medical centers. Recognizing these differences may have the beneficial effect of clarifying the criteria for patient enrollment in future multicenter trials that will establish (or refute) the contention that female sex is a potential source of bias in the medical care of patients with suspected coronary artery disease.

D. Douglas Miller, M.D.
Saint Louis University Health Sciences Center, St. Louis, MO 63110-0250

Leslee Shaw, Ph.D.
Veterans Affairs Medical Center, Durham, NC 27705-3897

2 References

1. 1

   Mark DB, Shaw LK, DeLong ER, Califf RM, Pryor DB. Absence of sex bias in the referral of patients for cardiac catheterization. N Engl J Med 1994;330:1101-1106
   Full Text | Web of Science | Medline

2. 2

   Shaw LJ, Miller DD, Romeis JC, Kargl D, Younis LT, Chaitman BR. Gender differences in the noninvasive evaluation and management of patients with suspected coronary artery disease. Ann Intern Med 1994;120:559-566
   Web of Science | Medline