Sex and the CT: An Evolving Story of the Heart


  • Basmah Safdar MD,

  • Judith H. Lichtman PhD,

  • Gail D’Onofrio MD, MS

  • A related article appears on page 147.

  • Supervising Editor: James R. Miner, MD.

Coronary artery disease (CAD) is the number one cause of mortality in the United States.1 Over the past 20 years, significant advances in technology and pharmaceuticals have reduced the overall mortality for acute coronary syndrome (ACS). However, the drop in mortality in men is greater than in women.1,2 This is especially true for women younger than 55 years, who are three times more likely to die compared to men.3,4 Several reasons have been postulated to explain this gap: differences in ACS presentation, lower suspicion for CAD in women by both patients and providers, and new sex-specific biologic mechanisms that often remain undetected. Women with acute myocardial infarction (MI) often delay seeking treatment due to both lack of symptom recognition and inability to recognize personal risk even in the presence of cardiac risk factors.5,6 About half (46%) of women are still unaware that CAD is the leading cause of death in women.7 Physicians may also underestimate risk in women. In a national survey, Mosca et al.8 found that less than 20% of the physicians knew women are more likely than men to die from CAD. Moreover, more women with intermediate Framingham scores are labeled as low risk, compared to men with identical scores. The lack of awareness has contributed to disparities at many levels of care including recognition, diagnosis, management, and prevention of CAD.9–19

In this issue of Academic Emergency Medicine, Ginty et al.20 focus on a cohort of low-risk chest pain patients to determine whether there are sex differences in the receipt of follow-up tests in patients within 1 month of undergoing coronary computerized tomographic angiography (CTA) for ACS symptoms. The study has several strengths: the original data were collected prospectively; there is good representation of women (55%); and the sample is relatively homogenous and low risk, with TIMI scores 0 to 2. Although men were more likely to have follow-up testing, this difference disappeared in analyses stratified for the degree of stenosis of the coronary arteries.

The results of this study need to be interpreted with some caution. First, most patients (86%) in this study did not have coronary stenosis, and stratification for coronary stenosis yielded few patients in each subgroup, making it difficult to draw meaningful conclusions, as the study was powered for the overall rate of testing. Consistent with preexisting literature, the overall rate of testing was higher in men than women (relative risk = 1.51; 95% confidence interval = 1.14 to 1.99). Larger populations are needed to compare the substrata by sex. Second, the data were drawn from a convenience sample, and therefore the true denominator of the patient population at risk for ACS is not reported. Third, the inclusion criteria were poorly described and based on a subjective assessment of “symptoms consistent with ACS.” This could have introduced possible selection bias due to variability in physician assessment of symptoms. No explicit screening process was identified to ensure the adequacy of patient selection. Finally, the authors reported outcomes such as infarction and death. Additional patient-centered outcomes such as persistent or recurrent chest pain and rehospitalization may be more common in women compared to men.21 It is therefore possible that persistent symptoms in women was an effect modifier for additional follow-up testing compared to men, independent of the CTA finding in this cohort.

Two key issues should be considered in the interpretation of these results. First, men and women may present differently with ACS. Although chest discomfort remains the most common presenting complaint for ACS, women are more likely to report atypical symptoms such as dyspnea, indigestion, nausea, back pain, jaw pain, bilateral shoulder pain, and fatigue compared to men.22–29 These atypical prodromal symptoms are linked with hormonal status and occur more commonly in premenopausal and perimenopausal women compared to men and older women.30 Contrary findings have been criticized for their underrepresentation of women.31 Commonly cited reasons that women delay seeking care include failure to recognize their symptoms as cardiac, financial concerns, and sociodemographics factors (older, single, or without insurance).5,32 This study did not address whether symptoms may have varied for women and men, and therefore it is plausible that there may have been differential receipt of CTA for women and men based on symptom presentation. It is plausible that patients with atypical symptoms were less likely to get CTA, and therefore a sizable segment of the relevant population at risk was not included, leading to a selection bias.

A second key issue is the diagnostic limitation of CTA itself in detecting the nonconventional biologic mechanisms of ischemia in women compared to men. Most of our current clinical practice for the detection and management of CAD rely on early studies conducted primarily on men. Recent advances have revealed biologic factors that explain the variable presentations and diagnostic challenges specific to women. Data over the past decade suggest that atherosclerosis may manifest and progress differently in men than women. Ischemic chest pain has traditionally been classified based on the presence of >50% obstruction of the coronary artery.33 In contrast, at least four alternative sex-specific mechanisms have been suggested as causes of ischemic chest pain in patients without classic anatomic obstructive coronary disease. These mechanisms have been described more commonly in women and are often lumped together as “Syndrome X.” They include: 1) microvessel disease causing an inability to augment blood flow in response to stress; 2) coronary slow flow phenomenon causing delayed blood flow to the distal myocardium due to functional obstruction; 3) plaque rupture and ulceration, causing temporary anatomic occlusion; and 4) coronary vasospasm, again causing transient functional obstruction. A substantial proportion (up to 30% of women and 12% of men) with nonobstructive disease have an MI, suggesting the need for expanding the conventional definition of ACS in women.34 Understanding these mechanisms is therefore necessary for the accurate interpretation of new diagnostic modalities used for chest pain evaluations.

Anatomical tests such as CTA, or even conventional angiography, do not detect these sex-specific processes, and that leads to a diagnostic challenge. Microvessel disease refers to diffuse luminal narrowing of the smaller arterioles and prearterioles, as opposed to focal stenosis of the larger coronary arteries that are traditionally thought to cause ischemic chest pain. The Women’s Ischemia Syndrome Evaluation (WISE) study group found that almost half the women who undergo angiography have clean coronaries, compared to only 17% of men. Chest pain in about half of these women with a negative catheterization was in fact found to be due to microvessel disease.35,36 Unfortunately, making this diagnosis can currently only be achieved through sophisticated nontraditional tests such as invasive coronary artery vasoreactivity or with positron emission tomography/magnetic resonance imaging scans.

Coronary slow flow phenomenon, on the other hand, has been described as a cause of rest ischemia in young males, typically smokers. It is defined as delayed progression of contrast injected in coronary vasculature during angiography.37 It relates to abnormally high baseline resistance in the coronary vessels, is associated with angina at rest, and is linked with serious arrhythmias and sudden cardiac death.

Plaque rupture and ulceration, as diagnosed by intravascular ultrasound, has recently been identified to be a frequent culprit for MI in women with nonobstructive disease. In a group of 50 women undergoing coronary angiography, these lesions were often found in segments that were angiographically normal, but correlated well with myocardial edema (indicative of ischemia) on MRI. No clinical or electrocardiographic abnormalities were found to be predictive of these plaques.38

Coronary vasospasm, originally described by Prinzmetal et al. in 195939 as a variant angina, is a syndrome of rest chest pain that was associated with ST-segment elevation. However, unlike typical angina, exercise tolerance in these patients is found to be normal. The pain is caused by spasm around mild CAD and is attributed to endothelial dysfunction. It has classically been associated with women and carries a good prognosis.

Diagnosing all of these mechanisms has important prognostic implications. Chest pain in the presence of so called “clean coronaries” has long been considered benign. However, this concept is rapidly changing. Beltrame et al.37 demonstrated higher rates of mortality and adverse events in patients with coronary slow flow phenomenon compared to those with normal flow. Similarly, the outcome data from the WISE study shows that patients with nonobstructive disease have persistent/recurrent chest pain, poor functional outcomes, high rates of return visits, and high costs of repeat evaluations compared to asymptomatic patients, as well as patients undergoing elective angiography who have clean coronaries (0% obstruction).40–43 Bugiardini et al.36 showed a 2% annual mortality or reinfarction rate in patients with MI in the presence of nonobstructive disease. Patel et al.44 found a > 15% readmission rate within 6 months of the index MI in a similar cohort. The present study by Ginty et al.20 solely addresses the issue of follow-up testing based on anatomical obstruction and not on the basis of patient symptoms. These functional and microvascular obstructions could lead to persistent symptoms and a relative increase in follow-up testing, neutralizing any sex inequalities due to decreased perception. Given that these sex-specific nontraditional mechanisms of ischemia have a latent period before adverse outcomes are observed, a 30-day follow-up period as reported in this study may not be a sufficient time frame to observe sex differences in outcomes. This is particularly important since the majority of the patients have no or nonobstructive CAD in this study. These alternate mechanisms could also possibly explain the unexpected cardiovascular death in a low-risk woman in this study, as well as the higher rates of adverse cardiac events observed in women with 50% to 69% stenosis.20 Further longitudinal studies are needed to understand the implications of persistent chest pain in the presence of a normal CTA. In the future, coronary CTA could potentially help delineate mortality risk in women by assessing the prevalence and prognosis of nonobstructive CAD.45

This study confirms that overall, males get a higher proportion of follow-up tests after undergoing CTA for CAD evaluation in the emergency department, although it suggests that this may be based on the underlying prevalence and significance of obstructive CAD.20 Caution is advised in interpreting these findings, as 1) there may have been bias in the referral for CTA based on symptom presentation (not adequately described or standardized); 2) CTA may not adequately recognize underlying mechanisms that are unique to women; and 3) the lack of findings here should not suggest that sex differences do not exist—just that this study design (with its limitations) did not detect a difference. More work is needed using larger cohorts and more objective measures of symptom presentation to explore and confirm this. Acknowledging and understanding the alternate sex-specific mechanisms of acute ischemia could play an important role in explaining sex differences in the presentation, diagnosis, and management of ACS.