Women and outcomes of coronary artery bypass surgery: do we have an answer?

Coronary artery bypass graft (CABG) surgery is a common revascularization procedure in US women with coronary heart disease. Approximately 180,000 CABG procedures are performed annually on women in the United States, or about 28% of the total performed.1 Despite this widespread use, the risks and benefits of CABG in women continue to be debated, and for decades, researchers have argued whether or not women benefit from this procedure to the same extent as men.2

In part, this uncertainty arises from the fact that major randomized trials of the efficacy of CABG surgery compared with medical therapy have included samples that were 97% men.3 Given that CABG surgery has become, nonetheless, a standard of care for both women and men with coronary heart disease, it is unlikely that a randomized trial will ever be performed in women. Therefore, to assess the benefits of CABG in women, researchers commonly use men as the control group, with the obvious limitations of this approach. These studies have provided conflicting results, both for adverse events and death,4, 5, 6, 7, 8, 9, 10 and for symptom relief and quality of life.7, 8, 11, 12, 13, 14, 15, 16, 17, 18, 19 Overall, clinicians and investigators are left with the suspicion that coronary revascularization poses special risks to women,2 and this may discourage referral of women for CABG and contribute to sex-related disparities for the receipt of this procedure.9, 10

Because CABG yields only a small absolute survival benefit relative to medical therapy among men, for whom clinical trial data are available,3 decisions about undergoing this procedure, in both men and women, are commonly based on the prospect of obtaining symptom relief and improvement in function. Despite the importance of such outcomes, relatively few studies have focused on them, and many have been limited by small samples of women or inclusion of patients who received CABG several decades ago.7, 8, 11, 12, 13, 14, 15, 16, 17 Overall, the results provided by these investigations have been conflicting with respect to sex-related benefits. In some studies, women had more symptoms and poorer functioning after CABG compared with men,7, 8, 11, 12 whereas in other studies there were no significant sex differences.13, 14, 15, 16 Yet, additional investigations pointed out more adverse outcomes among women in some dimensions of quality of life but not in others.17, 18

These conflicting results may be explained in part by differences in methodology. Several studies, particularly those which reported negative findings, were based on samples including fewer than 70 women, thus raising the concern of whether there was sufficient power to detect sex-related outcome differences. In addition, many reports were based on convenience samples, which may have been too selective to provide meaningful data. Furthermore, many studies, both positive and negative, did not take into account preoperative differences in health status between the sexes. Because women and men differ markedly in their preoperative health status, it is preferable to compare the degree of improvement between men and women rather than the absolute functional capacity at follow-up, but failure to adjust for baseline levels may produce biased results because of ceiling or floor effects and the effects of regression to the mean.20

In addition to limitations in study designs and patient samples, a number of methodological difficulties can be identified in this area of study that might contribute to the conflicting results. The first is the fact that women and men referred for CABG surgery represent two very different patient groups. Several investigations,6, 11, 15, 19 including an earlier report from the Post-CABG Biobehavioral Study,21 have consistently described remarkable differences in the demographic, medical, and psychosocial profile of men and women referred for CABG. Women undergoing CABG are older, less educated, have more severe angina and congestive heart failure, lower functional status, and higher level of depressive symptoms. This large number of differences between women and men make the comparison difficult, and studies have been challenged as to their ability to correct for so many potential imbalances that may influence sex differences in outcome. Even if all the relevant factors are measured, the potential for residual confounding remains.22

The presence of such marked differences in age and clinical status raises yet another potential problem. Some authors have suspected that these differences are due to sex bias in the diagnosis or referral of patients with suspected coronary disease.9, 10 If indeed there is a different threshold for referral to CABG surgery in women and in men, with women being referred at a more advanced disease stage than men, the sex comparisons may be biased. Paradoxically, despite being more symptomatic, women referred for CABG have less extensive coronary artery disease than men as determined by coronary angiography results.4, 7, 8, 23, 24 A similar pattern is observed among patients with confirmed acute myocardial infarction, with women presenting in a more severe clinical status but having fewer coronary lesions.25, 26 Although these data raise important questions about the reasons for the more severe presentation in women despite less severe angiographic disease,2 they undoubtedly challenge the notion of a delayed referral of women.

Another challenge has been the definition of an appropriate baseline for the assessment of presurgery health status. Because many patients undergo CABG in an emergent or urgent fashion, and because of the busy presurgery preparatory time, it is often unfeasible to evaluate the patients before surgery. In addition, presumably high anxiety levels before surgery may make patients' responses less valid. For these reasons, many studies have assessed presurgery status in a retrospective fashion, raising, however, concerns of recall bias. The Post-CABG Biobehavioral Study was one of few investigations that has attempted to interview the patients before surgery, but again this was feasible only in a subgroup of patients. Indeed, the data presented in this issue of the American Heart Journal indicate that it may not make much difference whether the health status data are collected before or after CABG surgery.

In this issue, Lindquist and coauthors present a new analysis of the Post-CABG Biobehavioral Study examining sex differences in physical, social, and emotional functioning after CABG surgery. Both women and men showed significant improvements in most of these domains at 1 year after CABG surgery. Additionally, these improvements were overall of similar magnitude in women and men, although women did show less benefit with regard to the symptoms of shortness of breath and tiredness. These results are important in many respects. First, they should be more robust than several previous reports because they are based on a larger sample of patients. In fact, with the inclusion of 269 women, this study remains one of the largest, in terms of number of women enrolled, among studies evaluating functional outcomes after CABG in women. Second, functional status measures encompassed a number of different domains and were measured prospectively at three different time points, whereas several previous studies only focused on chest pain symptoms or lacked a baseline assessment of health status. Third, the 1-year follow-up period was sufficient to allow for full recovery and the achievement of CABG-related functional gains.

Although this study represents a significant contribution to this area of research, it still may not provide a definitive answer to the question whether women fare worse or the same than men after CABG surgery. First, it is based on a relatively old cohort, assembled in the late 1980s and early 1990s; thus, there is the possibility that such results may not be extrapolated to more recent times due to changes in patient characteristics, CABG techniques and concomitant treatments. Second, a selection bias cannot be ruled out given that a large number of patients selected for recruitment were not enrolled, again highlighting the difficulties in applying rigorous methodology in this area of research. Finally, there was a large time frame, spanning several months, for each follow-up assessment, which could increase imprecision and make it more difficult to find differences if there were any.

Without doubt, additional large prospective studies enrolling substantial numbers of women are needed to come to a conclusion regarding whether women have less functional benefit than men after CABG surgery. In the meantime, recognizing the methodological challenges in this area of research, the study by Lindquist et al is reassuring as to whether women attain equal benefit than men after this common procedure. Until more data are gathered, there is no basis for discouraging the use of CABG surgery for the improvement of quality of life of women with coronary heart disease.