Article Text

Download PDFPDF

Persistent sex differences in outcomes after coronary heart disease: time to move from observation to action
  1. Rabea Asleh
  1. Heart Institute, Hadassah University Medical Center, Faculty of Medicine, Hebrew University of Jerusalem, Jerusalem, Israel
  1. Correspondence to Dr Rabea Asleh, Heart institute, Hadassah University Medical Center, Faculty of Medicine, Hebrew University of Jerusalem, Jerusalem, Israel; rasleh{at}

Statistics from

Coronary heart disease (CHD) remains a leading morbidity and mortality threat in both men and women, affecting millions of individuals globally. Multiple studies over the last three decades have repeatedly demonstrated significant sex differences in baseline characteristics, clinical presentation and coronary angiographic features, as well as in outcomes across various types of CHD.1 2 Specifically, women presented with first CHD at older age accompanied by a higher prevalence of risk factors and comorbidities compared with men. Studies have consistently demonstrated that the unadjusted outcome of women after acute coronary syndrome is significantly worse than men. This disparity in outcomes was largely attenuated after adjustment for age, comorbidities and other confounders in some, but not all, studies. Women were found to have more angina symptoms despite decreased prevalence of significant epicardial coronary artery disease and ischaemia, and were less likely than men to undergo coronary revascularisation or be treated with guideline-based medical therapies. Moreover, women were found to be at greater risk of procedural complications as well as early and late mortality following coronary revascularisation, with more noticeable differences observed in young women compared with their male counterparts. Despite the increasing awareness of a considerable gap in care and the growing understanding of differences in the pathophysiology of CHD, these disparities in outcomes between men and women persist.

In this issue of the Journal, Akyea and colleagues3 revisit this question in a large retrospective population-based study derived from the UK Clinical Practice Research Datalink GOLD registry of primary care electronic records and provide further insight into the sex disparities in adverse clinical outcomes after incident CHD. These outcomes included recurrent major adverse cardiac events (MACE), defined as the composite of recurrent CHD, any stroke, peripheral vascular disease (PVD), heart failure (HF) or cardiovascular-related mortality (the primary outcome), and the individual outcomes of this composite outcome and all-cause mortality (secondary outcome). Overall, 143 702 adult individuals who developed incident CHD (stable or unstable angina, myocardial infarction (MI) or coronary revascularisation) from 1998 through 2017 were included in the analysis (~44% women; median age 73 and 66 years for women and men, respectively). Most baseline characteristics, including demographics, important comorbidities and drug prescription, differed significantly between men and women. Specifically, women at the time of incident CHD were older, with lower socioeconomic status, and had higher prevalence of chronic kidney disease, atrial fibrillation, hypertension, hyperlipidaemia, cancer and depression as compared with men. Furthermore, women were more likely than men to be prescribed antiarrhythmic, antihypertensive, antiplatelet, diuretic and beta-blocker drugs at the time of presentation with incident CHD. Over a median follow-up time of 13 years after the incident CHD, 63.8% of all individuals developed MACE, and recurrent CHD, stroke, PVD, HF, cardiovascular death and all-cause mortality occurred in 46.3%, 4.0%, 1.1%, 5.5%, 6.9% and 20.5%, respectively. Using multivariable Cox proportional hazard regression and competing risk models, women were found to have 29%–32% and 38%–40% decreased incidences of MACE and recurrent CHD, respectively, while they experienced a large excess risk of stroke, HF and all-cause mortality (26%–32%, 9%–13%, and 5%–11% increased risk, respectively) as compared with men. The risk of cardiovascular death was similar between the two groups.

The authors are to be commended for performing a comprehensive analysis of a large population with incident CHD and investigating different fatal and non-fatal outcomes after a long-term follow-up. Importantly, most studies examining sex-related differences in outcomes in the setting of CHD have mainly included selected populations from clinical trials or hospital registries, in which women were remarkably under-represented, thereby not reflecting community practice. While most studies have focused on differences in all-cause mortality and short-term outcomes between men and women with recurrent cardiovascular events, little evidence is available from population-based studies for long-term outcomes, particularly non-fatal outcomes, including recurrent CHD events, stroke and HF after incident (first ever) CHD. The current results, therefore, extend upon multiple studies evaluating sex disparities in outcomes across the spectrum of ischaemic heart disease and examining long-term non-fatal outcomes. However, there exist substantial heterogeneity and methodological differences across reports examining these sex differences in outcomes following CHD, which may result, in many cases, in conflicting results. The complexity of the statistical analysis is also reflected in this study as multiple approaches were used, including adjusted incidence ratios, Cox regression models with and without competing risks, and win ratio analysis. For example, the incidence rate ratio (adjusted for age and socioeconomic status) of all-cause mortality was found to be lower in women than in men (HR 0.92, 95% CI 0.90 to 0.94), while adjusted Cox analysis and competing risk model (using Fine-Gray subdistribution hazard model) were suggestive of a significantly increased mortality risk among women compared with men. These findings may indicate that the outcomes may be largely influenced by the range of statistical methods employed and the spectrum of potential confounding variables included in multivariable models in studies investigating disparities in outcomes by sex. Accounting for competing risks is important, and perhaps more appropriate than standard Cox regression models, when investigating fatal and non-fatal cardiovascular events in the community, particularly in populations susceptible to competing risks (eg, death due to non-cardiovascular events when investigating MACE or cardiovascular death). The win ratio analysis is a validated method and was additive in this study for assessment of the effect of being female on the fatal cardiovascular outcome within the composite MACE because this approach requires ranking of outcomes in a hierarchical fashion by severity, thereby giving a higher importance to more clinically significant events.

It is worth noting that the significant reduction in MACE observed in this study was driven by a remarkable reduction in recurrent CHD among women, whereas other components of MACE and all-cause mortality were significantly increased in women compared with men after incident CHD. This discordance in the sex disparities in outcomes is puzzling and merits further discussion. Although it is well established that women have lower age-specific rates of incident CHD than men, less evidence is available regarding CHD events among individuals with established cardiovascular disease. In one large study,4 involving approximately 340 000 adults hospitalised with MI and over 1.3 million matched adults without a history of CHD from a contemporary healthcare database in the USA, the lower risk of MI, CHD and all-cause mortality in women versus men was significantly attenuated after MI. Interestingly, the reduction in recurrent CHD observed in the current study by Akyea et al 3 among women was not translated into reduced cardiovascular death, and all-cause mortality was even higher among women. The definition and severity of these recurrent CHD events in this population are undetermined. Therefore, one may wonder which proportion of these events represented major (MI and those requiring revascularisation) versus minor (angina treated only medically without the need for coronary intervention) events, and whether there were differences in CHD type and severity between men and women across the spectrum of ischaemic heart disease that might influence cardiovascular mortality. In a large geographically defined community cohort of patients with incident MI, we have recently shown that women experienced a significantly increased risk of recurrent MI as compared with men after using propensity scores to equalise the most important baseline characteristics between men and women.5 This increased risk of recurrent MI was in contrast to the risk of HF, cardiovascular death and all-cause mortality, which did not differ by sex after propensity score adjustment. Data from the SWEDEHEART registry comparing sex differences in all-cause mortality after MI with those in the general community have shown that, although overall mortality rates were similar between men and women, women with MI had higher excess mortality attributed to MI than men with MI.6 Therefore, these findings might be explained by a possible increase in cardiovascular mortality in women experiencing recurrent CHD. In support of this explanation is that MACE analysis using a win ratio approach suggested that women had more fatal outcome in the composite MACE than men. Additionally, the increased rates of HF and stroke after incident CHD and possibly other unadjusted comorbidity diseases in women may balance their reduced risk of recurrent CHD and ultimately lead to comparable cardiovascular death rates and even higher all-cause mortality rates.

The increased risk of HF in women after incident CHD is another important observation of this study. Fewer studies have previously examined HF development after acute or chronic CHD, with inconsistent results.5 7 8 In a cohort of patients with known or suspected CHD who underwent vasodilator stress cardiac magnetic resonance (CMR) imaging, a significantly increased risk of new-onset HF was reported in women, especially for patients with preserved ejection fraction, which was independent of the ischaemic burden detected by CMR or other comorbidities.7 A large population-based study has shown that, although the unadjusted rates of HF in hospital or after discharge were higher in women than men after MI, this outcome difference between sexes disappeared for non-ST segment elevation MI, but remained for those with ST segment elevation MI after appropriate adjustment.8 These disparities by sex in HF risk are likely multifactorial and may include sex-specific differences in clinical presentation, making diagnosis more difficult and delaying effective therapies for HF in women, response to medical therapy, and a higher myocardial vulnerability to coronary artery occlusion and/or less ischaemic preconditioning in women with CHD. Furthermore, women after an CHD event are less likely to receive coronary revascularisation and less often discharged on guideline-directed medical therapy than men. Beyond coronary revascularisation, there might be potential sex differences in left ventricular remodelling and chronic hypoperfusion attributed to differences in microvascular flow and non-culprit atherosclerotic coronary artery disease burden leading to myocardial stiffness and impairment in systolic and diastolic function.

The population-based data are a strength of this study, as is the long follow-up for both fatal and non-fatal outcomes after incident CHD. However, some limitations should be noted. As in any observational study based on electronic health records with high frequency of missing data and variability in risk factors and outcomes documentation, potential ascertainment and misclassification biases are possible. Moreover, the study included a heterogeneous cohort with the whole spectrum of acute and chronic CHD with different pathophysiology and response to treatments. Therefore, differences in outcomes by sex may vary according to the subtype of CHD. In the absence of a control group without incident CHD, it is unclear whether long-term outcomes, such as stroke, PVD or HF, are always directly linked to CHD or whether they represent an independent event, particularly in mild CHD cases. Finally, important information that may influence outcomes, such as gender-based differences in prescription and adherence to evidence-based medical therapies after CHD, extent of coronary artery disease, coronary revascularisation, and type of HF, is lacking.

How should we move forward from observation to untangling the sex disparities and improving outcomes after CHD? Study after study, although drawn from different cohorts and examining different outcomes, has concluded persistent disparities in outcomes following CHD. It is evident that these disparities are attributed to a complex interplay of physiological, biological and perhaps psychosocial issues that differ in men and women. Specifically, most studies have shown that baseline characteristics and risk factor profile (age, comorbidities and CHD characteristics), clinical management, and coronary anatomy/physiology all differ by sex and likely contribute to differences in outcomes following CHD. Time to act! Consistent with previous studies, the study by Akyea and colleagues3 provides important data on persistent sex differences in outcomes after CHD, thereby underscoring the need to develop strategies aimed at improving inequalities in cardiovascular outcomes (figure 1). Future efforts should be guided towards increasing the representation of women in both preclinical and clinical studies and further investigating the mechanisms underlying sex differences in biological, behavioural and psychosocial factors associated with CHD. Only then, we will be able to unravel these disparities and develop new strategies and sex-specific algorithms to bridge the outcome gap between men and women after CHD.

Figure 1

Schematic depicting sex differences in clinical characteristics, coronary physiology and clinical management responsible for sex-based differences in outcomes and suggested strategies to bridge the outcome gap between men and women with CHD. CAD, coronary artery disease; CHD, coronary heart disease; MACE, major adverse cardiac events; PVD, peripheral vascular disease; SCAD, spontaneous coronary artery dissection.

Ethics statements

Patient consent for publication



  • Twitter @rasleh

  • Contributors RA was responsible for editorial writing and preparation of the figure and all relevant materials in this article.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Provenance and peer review Commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles