Introduction

Ischemic heart disease (IHD) is the leading cause of death for women worldwide [1]. This statistic is no different in the United States (US), where IHD accounted for nearly 160,000 female deaths in 2015 [1]. Recent estimates demonstrate that 7.4 million women in the US are living with IHD [1]. Owing in part to national educational campaigns, there has been improved awareness, recognition, and treatment of IHD in women since 1979, resulting in an initial decline in female mortality. However, awareness of the burden of IHD in women, its risk factors, and recognition of symptoms remain subpar in both the lay and medical communities [2]. In 2012, only 56% of women were aware of IHD being their leading cause of death [1].

As a result, perhaps, since 2010, mortality trends in women seem to have stabilized, and more concerningly appear to be on the rise [1]. Certain subsets of women, particularly the young (< 55 years old) and ethnic minorities continue to experience worse outcomes compared to age-matched men [1, 3]. Women are still less likely to receive preventive and guideline-directed care than men with similar ASCVD risk scores [3]. When prescribed appropriate therapies, women are also less likely to be treated aggressively and are less likely to achieve optimal effects from these therapies, underscoring a disparity in diagnosis and management of IHD in women [3].

This review outlines the contemporary data on the sex-specific differences in pathophysiology, clinical presentation, traditional and novel risk factors, diagnostic evaluation, and management strategies for stable IHD in women. Contemporary challenges in diagnosis and management of IHD in women that contribute to ongoing disparities are highlighted.

Impact of ischemic heart disease

As the number one cause of death in both men and women in the US [4], IHD has a huge impact on healthcare costs. Cardiovascular disease accounted for 14% of the total health expenditure between 2013 and 2014, with hospitalization for IHD being a major contributor to this [1]. These costs are projected to increase > 100% by 2035 [1]. In 2013, hospitalizations for IHD, including myocardial infarction, were among the most expensive [1]. Women with IHD have higher resource utilization as evident by more office visits, more avoidable hospitalizations, and higher rates of recurrent angina, heart failure, and myocardial infarction (MI) mortality, compared to men [5]. Compared to men, women with angina have inferior functional status scores, even after adjusting for comorbidities and severity of IHD [6]. This in turn translates to higher healthcare costs and potentially poor compliance, highlighting the substantial impact that IHD has in women, not only at an individual level but also at the population level [7].

Risk factors and risk assessment

Traditional cardiovascular risk factors are overall similar between the sexes. The nine risk factors identified in the INTERHEART study: abnormal lipids, smoking, hypertension, diabetes, abdominal obesity, psychosocial factors, consumption of fruits, vegetables, and alcohol, and regular physical activity account for 94% of the risk of MI in women and 90% in men [8]. Biological differences between men and women may affect the expression of cardiovascular risk factors, and impart a differential risk for women (Table 1). For instance, smoking, hypertension, diabetes, and physical inactivity have a higher odds ratio of predicting risk of MI in women compared to men [8, 9]. The cumulative effect of the conventional modifiable risk factors confers a nearly 2-fold relative increased risk of MI in older women and 8-fold increased risk in younger women compared to men, highlighting the importance of primary prevention in young women [8]. Table 1 highlights novel cardiac risk factors that are unique to women including early menopause or menarche, gestational diabetes, pregnancy-related hypertension, autoimmune disorders, and psychosocial stresses [9, 16, 20,21,22, 23••]. The importance of incorporating these risk factors is increasingly being recognized, and both American and European cardiology societies have made recommendations for consideration of these novel risk factors in assessing a woman’s risk [23••, 24] (Table 1).

Table 1 Association of risk factors with IHD in women
Full size table

Despite 48% of women having three or more traditional risk factors for IHD [10], discussion of risk factors and individual risk assessment does not occur consistently in women [25••]. Multiple calculation tools exist to predict the risk of IHD; most do not account for the sex-based differential risk of the traditional risk factors [26, 27]. The American Heart Association and American College of Cardiology (AHA/ACC)’s atherosclerotic cardiovascular disease (ASCVD) risk assessment tool is perhaps the most widely used and distinguishes risk based on sex [28]. It, however, does not factor in any of the novel risk factors specific to women [28]. Despite the higher number of risk factors, most global risk scores underestimate the true risk and characterize a greater percentage of women as ‘low risk” compared to their male peers [29]. There are also no clear recommendations on how to incorporate novel risk factors into risk scores, except for the European Society of Cardiology’s recommendation to use a 1.5 multiplier in rheumatoid arthritis and autoimmune disease [23••].

Clinical presentation

Biological variations including differences in coronary artery size, hormonal influences, autonomic innervation, and hematologic and electrophysiologic indices contribute to differences in symptom presentation of IHD between men and women [30] (Fig. 1). Only 31% of young women present with chest pain, compared to 42% of men [31, 32]. The evaluation of symptomatic women is hampered by use of the definition of “typical angina” in traditional models such as the Diamond-Forrester tools [33]. These risk models are derived from large cohorts of men and are more reflective of the male pattern of exertional chest pain. Among patients diagnosed with acute coronary syndromes, women were less likely to report typical chest pain and were more likely to report atypical chest pain, abdominal discomfort, loss of appetite, dyspepsia, nausea, vomiting, dyspnea, hand numbness, palpitations, dizziness, fatigue, or weakness [31, 32, 34]. Presence of these non-specific symptoms and vague presentations further hamper the timely recognition and diagnosis of IHD in women [31, 34]. In particular, younger women often present with absence of chest pain, have a delayed presentation, and suffer worse outcomes [32, 34, 35].

Fig. 1
figure 1

Biologic reasons for sex-specific manifestations of ischemic heart disease. Biologic reasons for sex-specific manifestations of ischemic heart disease include difference in anatomic parameters, autonomic and electrical indices, hematologic factors, cardiac function, as well as hormonal influences. LV left ventricle, QTc corrected QT interval, IHD ischemic heart disease.

Full size image

Pathophysiology

The dynamic interaction of hormonal influences, atypical risk factors, and smaller, more vasoreactive coronary arteries in women result in a female-specific phenotype of IHD [36]. Symptomatic women are less likely to experience the traditionally described male pattern of flow-limiting epicardial atherosclerosis [37]. Instead, they are more prone to epicardial and microvascular spasm, vascular inflammation, myocardial bridging, and dysfunction of the endothelium and microvasculature [37,38,39,40]. Women have a lower atheroma burden and exhibit more diffuse, less obstructive disease compared to men [39, 41]. Thus, symptomatic women at risk for IHD frequently (60–70%, compared to ~ 30% men) have angiographically normal coronaries or non-obstructive disease on invasive coronary angiography (ICA) [37, 42].

The appearance of normal coronaries on ICA may be misleading, as women may still have a high burden of microvascular disease that portends an adverse prognosis [43•]. Normally, the coronary flow increases 2.5–5-fold in response to physiologic or pharmacologic stress (coronary flow reserve (CFR)) [44]. However, patients with microvascular dysfunction are unable to increase coronary flow due to impaired reactivity of the coronary microvasculature, resulting in reduced CFR and myocardial perfusion and ultimately myocardial ischemia [45]. Microvascular disease, while present in men, is far more prevalent in women [39]. Although the traditional IHD risk factors are implicated in development of microvascular disease, the exact mechanism for its development and progression is poorly understood.

While both macrovascular and microvascular coronary disease may operate independently, they more frequently operate in concert and have been hypothesized to be a continuum of sex-specific response to vascular injury [41]. Consequently, to accommodate the full spectrum of coronary atherosclerosis in women, the term IHD is more apt.

Sex hormones play a key role in the differential expression of IHD between the sexes. Premenopausal women have a lower incidence of IHD compared to age-matched men, but the incidence of IHD steeply rises following menopause [10]. Estrogen plays a protective role against the development of atherosclerosis by inhibiting smooth muscle proliferation, matrix deposition, and promoting re-endothelization following vascular injury [46]. However, exogenous hormone replacement therapy after menopause is not beneficial in delaying progression of atherosclerosis and has not resulted in improving cardiac mortality [47].

Diagnostic evaluation of stable IHD in women

The AHA’s statement on the role of non-invasive testing for women with suspected IHD provides a sex-specific algorithm that incorporates both functional stress imaging and anatomic imaging [48••]. Functional stress testing includes exercise treadmill testing (ETT), echocardiography, single-photon emission computed tomography (SPECT), positron emission tomography (PET), myocardial perfusion imaging (MPI), and cardiac magnetic resonance imaging (CMR); in contrast, anatomic assessment relies on coronary computed tomography angiography (CCTA) and ICA. Contemporary guidelines on diagnosing IHD predominantly focus on stress imaging techniques that traditionally look for hemodynamically significant luminal stenosis, warranting revascularization [48••, 49]. Advances in our understanding of the pathophysiology of IHD in women have rendered these algorithms inadequate. Although a stress test may exclude epicardial stenosis in symptomatic women, they often continue to experience symptoms and high mortality, underscoring the need to supplement our testing with additional imaging to explore the full spectrum of IHD that encompasses evaluation of obstructive and non-obstructive plaque, and dysfunction of the coronary microvasculature and endothelium [50, 51].

The basis of choosing the appropriate diagnostic test to evaluate for IHD in women depends on a number of factors, including test availability, local expertise, patient age, body habitus, ability to exercise, and an individual’s risk profile and pretest probability of having IHD [48••]. Women at low risk for IHD are generally not considered candidates for further diagnostic testing. Given that ETT is cheap, readily available, and has high negative predictive value, it is recommended as the first-line test in symptomatic women at an intermediate risk for IHD, with a normal resting ECG and able to perform maximal exercise [48••]. However, women have a higher proportion of obesity and physical inactivity, and are consequently unable to produce maximal exercise, and have ECG changes related to hormonal influences, all of which result in lower diagnostic accuracy with ETT compared to imaging studies [14, 48••, 52]. Symptomatic women at intermediate-to-high risk for IHD, with resting ST segment ECG abnormalities or unable to exercise, are generally referred for stress imaging to assess for stress-induced wall motion or myocardial perfusion abnormalities [48••]. Table 2 highlights the advantages and disadvantages of each modality, along with the sensitivity and specificity of detecting obstructive atherosclerosis in women [53].

Table 2 Advantages, disadvantages, and diagnostic accuracies of testing modalities for IHD in women
Full size table

Stress echocardiography

Stress echocardiography confers a higher diagnostic accuracy than ETT [55] and is predictive of IHD events in women, including MI and death, even in the absence of obstructive coronary artery disease (CAD) on angiography [56, 57]. A 2-year cardiac event-free rate of 97% was observed in those with a stress echocardiogram, while an abnormal study was associated with a 4-fold higher risk of cardiac events [57, 58].

Stress myocardial perfusion imaging

Stress MPI, with either SPECT or PET, offers comparable diagnostic accuracy to stress echocardiography [58]. However, its use may be limited in younger women due to the exposure to ionizing radiation. Improvements in scanning protocols, scanner types, and isotopes have improved versatility with resultant radiation doses as low as 1 mSv [59]. In addition to high diagnostic accuracy, stress MPI offers incremental prognostic value over clinical variables, ECG and LVEF in symptomatic women at risk for IHD [60, 61]. A normal stress MPI scan is associated with < 1% annual cardiac event rate compared to over 3-fold with abnormal MPI study in women [61].

Compared to SPECT, PET imaging, owing to its higher spatial resolution and inbuilt attenuation correction, reduces breast attenuation artifacts commonly seen in obese women, and allows for improved visualization of small perfusion defects, particularly in women who generally have smaller hearts, with lower radiation exposure [54, 60, 62]. Stress MPI with PET also allows for the evaluation of myocardial flow reserve, which is an assessment of the absolute blood flow across coronary arteries [50, 60, 63]. A diminished myocardial flow reserve (defined as < 1.9–2.0) is suggestive of underlying vascular dysfunction and microvascular disease [60, 63].

Cardiac MRI

CMR is gaining popularity as a robust non-invasive modality that can accurately assess global and regional systolic left ventricular function, myocardial perfusion, and scar. Owing to its higher spatial resolution, CMR has demonstrated higher diagnostic accuracy compared to SPECT MPI, with a sensitivity of 88.7% and specificity of 83.5% in women [64]. Prognostic value of CMR is similar between men and women [65]. Women with an abnormal dobutamine CMR had a 4-fold higher risk of MI or cardiac death when compared to women with a normal CMR [66]. On the other hand, a normal stress CMR in women is associated with an annual rate of 0.3% of major adverse cardiac events (MACEs) [65].

Advanced stress CMR techniques, including the assessment of myocardial flow reserve and perfusion reserve index, also allow for the detection of impaired coronary vasoreactivity and endothelial dysfunction [67•]. In a study of 113 symptomatic women without obstructive CAD on coronary angiography, 57% were found to have abnormal subendocardial perfusion seen on adenosine stress MRI but not rest images, consistent with coronary microvascular disease [68]. These perfusion defects, even in the absence of obstructive coronary atherosclerosis, were associated with MACE and cardiac death [65]. Females with evidence of ischemia on CMR have an annual MACE rate of 15%, while those without ischemia have an MACE rate of only 0.3% [65].

Coronary computed tomography angiography

CCTA is an emerging anatomic approach to evaluating IHD and provides valuable insights regarding the extent and severity of coronary atherosclerosis, luminal stenosis, plaque composition, and the presence of arterial remodeling [48••, 67•]. CCTA offers excellent risk stratification, segregating women with normal coronary arteries, non-obstructive CAD, and obstructive CAD to cardiac event rates of 0.2, 1.2, and 2.1%, respectively [69]. Importantly, presence of non-obstructive atherosclerosis on CCTA was not benign and associated with a higher symptom burden and 2-fold higher mortality compared to no disease [69].

Coronary angiography

ICA remains the “gold standard” for the diagnosis of obstructive CAD in both men and women. ICA is currently recommended as the initial strategy for CAD evaluation in sudden cardiac death survivors or patients with potentially life-threatening arrhythmias, or those with signs and symptoms of heart failure [49]. Both men and women who have intolerable ischemic symptoms despite appropriate guideline directed medical therapy and who are candidates for revascularization, should undergo ICA [49]. It is also recommended that patients who have a high pretest probability of severe ischemia who are candidates for revascularization should undergo ICA, a recommendation tempered by uncertainty of the added benefit of revascularization above medical therapy in this group of patients [70]. ICA is not without risks, and its incremental value in directing further management should be carefully considered before proceeding.

Nearly two thirds of women with suspected IHD have “normal coronaries” or non-obstructive disease on ICA [37, 42]. Absence of flow-limiting epicardial disease is not necessarily a benign finding as short- and long-term prognoses in these women is poorer compared to the background population without angina [37, 43•, 71]. Performance of advanced angiographic techniques may demonstrate the presence of occult coronary abnormalities, including non-obstructive plaque by intravascular ultrasound, endothelial dysfunction on acetylcholine testing, and microvascular dysfunction on adenosine testing [39, 72]. Despite “normal coronaries” on ICA, there is a high prevalence of atherosclerotic plaque on intravascular ultrasound as coronary artery remodeling may limit the sensitivity of the angiogram for the presence of atherosclerosis [72]. Optical coherence tomography may also provide additional information on plaque morphology [73].

Patients with intermediate (non-obstructive) lesions, and a small subset of patients with visually normal coronaries, may have abnormal fractional flow reserve (FFR), suggesting hemodynamically obstructive disease for which revascularization may be of benefit [49]. Women have been observed to have higher FFR measurements for similar degrees of angiographic stenosis severity when compared to men [74, 75]. Potential explanations for this include smaller myocardial mass supplied by the stenosed vessel, inaccurate visual estimation of stenosis due to smaller vessel size, or a higher prevalence of left ventricular hypertrophy and diastolic dysfunction in women which may impact microvascular function and maximal hyperemia, among others [76]. An FFR-guided strategy of revascularization is similarly beneficial in both sexes [74]. FFR should therefore be employed liberally in evaluation of intermediate lesions in women as symptoms may be due to microvascular dysfunction rather than epicardial CAD.

Prognosis and outcomes in women with IHD

Paradoxically, even though women have a lower burden of atherosclerosis, they experience a higher incidence of angina, worse quality of life, recurrent hospitalization, and mortality [37, 39, 77]. This is in-part explained by the higher prevalence of non-obstructive coronary atherosclerosis and presence of microvascular disease in women [37, 42]. The 5-year event rate for MI, hospitalization for heart failure, stroke, or cardiac death was 16.0% for symptomatic women with mild non-obstructive disease, 7.9% with no atherosclerosis, and 2.4% in asymptomatic women, matched for risk factors [78]. Similarly, symptomatic women with no obstructive coronary disease had a 10-year all-cause mortality rate of 13%, which was 5-fold higher than in an age-matched asymptomatic reference cohort [43•]. Additionally, a large meta-analysis comprising 26 studies reported, patients with endothelial dysfunction and microvascular disease have a 2.3–4.5-fold increased risk of cardiovascular events [79]. Women with non-obstructive coronary artery disease experience worse outcomes than men [80]. Women are three times more likely than men to experience a MACE within the first year of angiography [80]. Furthermore, given recurrent symptoms in absence of flow-limiting stenosis, they have a high rate of repeat coronary angiography (15.7% at 5 years) [81]. Despite these compelling findings, treatment of women with no-flow obstructing lesions often remains limited to reassurance and results in increased hospitalizations, recurrent coronary angiography, and worse outcomes in response to refractory symptoms.

Contemporary challenges and existing knowledge gaps

A large body of evidence suggests that women experience a greater symptom burden and incur more frequent office visits and hospitalizations, resulting in greater healthcare costs [5, 11, 81]. The reasons for the disparity in outcomes are multifactorial and relate to inadequate understanding and appreciation of biological differences, lack of recognition of atypical symptoms and novel risk factors in women, subconscious referral bias, and suboptimal application of existing evidence-based guidelines [11].

Recognition and awareness of IHD in women

Women experience longer delays in seeking medical care, and time from medical contact to revascularization time, exceeding that recommended by guidelines, compared to men [82]. Potential reasons include inadequate awareness of disease burden and the different clinical presentations in women, both by public and the medical community [25••, 83]. In 2014, only 55% women were aware of IHD being the leading cause of death in women, with even lower rates of awareness in ethnic minorities, and women with lower education and income [25••]. Although most women have ≥ 3 risk factors, only 52% considered themselves at risk at the time of their index MI [84]. Even when women recognize they may be having a MI, only 50% seek urgent medical attention [25••, 83]. Women’s delay in seeking care may also be related to frequently receiving inaccurate and inconsistent responses from the medical community, as their symptoms have often being minimized and attributed to non-cardiac etiologies [2]. Concerningly, the rates of awareness are actually worse among physicians with only 39% of primary care physicians viewing IHD as the top concern during routine clinic visits with their female patients [25••]. Furthermore, only 53% cardiologists and 44% primary care physicians utilized the ASCVD risk calculator for risk assessment of their female patients, and women’s risk of IHD was consistently underappreciated [25••]. Even when clinicians suspect symptoms of IHD, the diagnostic tools may not be adequately sensitive or specific in women [14, 29].

Summary and recommendations

Although in the last decade there has been an expansion of our understanding of sex-based differences in risks, symptoms, assessment, and intervention of IHD, these biological differences are still underappreciated. The same definition of “typical angina” continues to be used for both sexes, despite this representing a male-pattern of symptomatology. The same conventional risk assessment scores are used in both sexes, excluding the novel risk markers prevalent in women, resulting in underestimation of risk in women [14, 29]. The same normal thresholds for troponin assays are used to determine presence of MI in both sexes, despite data suggesting these thresholds may lack sensitivity in women [85]. Women are often subject to the same diagnostic algorithms as men, which are insufficient to detect the varied ischemic etiologies of chest discomfort in women. Reasons for the suboptimal appreciation of biological differences include underrepresentation of women in clinical trials and registries, and male dominant results being extrapolated to women [86].

Evolving knowledge about microvascular disease needs to be better incorporated into IHD guidelines and practice styles. Presently, there are only a handful of contemporary guidelines to address IHD prevention, diagnosis, and management in women, and these, too, are poorly implemented [11, 25••], highlighting a more pervasive subconscious bias in the medical community. This potentially calls for a system-wide paradigm shift to abandon classic definitions and embrace a more sex-conscious practice of medicine. Adequately powered research to answer gender-focused questions is needed to better establish female-specific risk scores, diagnostic thresholds, diagnostic algorithms, and management pathways. Research comparing the different imaging modalities for diagnosing non-obstructive plaque and microvascular disease may be beneficial. Discovery of novel treatment strategies for women that do not begin with identification of flow-limiting stenosis would be paramount [41]. Improved recognition of the sex-differences must also translate into inclusion of sex- and gender-based curricula into medical training, and at scientific meetings, to transform practice paradigms, along with public health education campaigns to ultimately result in improved IHD outcomes in women.