Sonia S. Anand MD, PhD, FRCPC, Department of Medicine, McMaster University, Hamilton General Hospital Campus, DBCVSRI - 237 Barton St. E Rm. C3-102, Hamilton, ON, Canada L8L 2X2. (fax: 905-528-2841; e-mail: email@example.com).
Abstract. Kreatsoulas C, Natarajan MK, Khatun R, Velianou JL, Anand SS (McMaster University; CARING Network, McMaster University; Population Health Research Institute, McMaster University and Hamilton Health Sciences; Interventional Cardiology, Hamilton Health Sciences; Eli Lilly Canada–May Cohen Chair in Women's Health, McMaster University; Michael G. DeGroote-Heart and Stroke Foundation of Ontario Chair in Population Health Research, McMaster University; Population Genomics Program, McMaster University; McMaster University, Hamilton, ON, Canada). Identifying women with severe angiographic coronary disease. J Intern Med 2010; 268:66–74.
Objectives. To determine sex/gender differences in the distribution of risk factors according to age and identify factors associated with the presence of severe coronary artery disease (CAD).
Design. We analysed 23 771 consecutive patients referred for coronary angiography from 2000 to 2006.
Subjects. Patients did not have previously diagnosed CAD and were referred for first diagnostic angiography.
Outcome measures. Patients were classified according to angiographic disease severity. Severe CAD was defined as left main stenosis ≥50%, three-vessel disease with ≥70% stenosis or two-vessel disease including proximal left anterior descending stenosis of ≥70%. Univariate and multivariate logistic regression was used to assess the association between risk factors and angina symptoms with severe CAD.
Results. Women were less likely to have severe CAD (22.3% vs. 36.5%) compared with men. Women were also significantly older (69.8 ± 10.6 vs. 66.3 ± 10.7 years), had higher rates of diabetes (35.0% vs. 26.6%), hypertension (74.8% vs. 63.3%) and Canadian Cardiovascular Society (CCS) class IV angina symptoms (56.7% vs. 47.8%). Men were more likely to be smokers (56.9% vs. 37.9%). Factors independently associated with severe CAD included age (OR = 1.05; 95% CI 1.05–1.05, P < 0.01), male sex (OR = 2.43; CI 2.26–2.62, P < 0.01), diabetes (OR = 2.00; CI 1.86–2.18, P < 0.01), hyperlipidaemia (OR = 1.50; CI 1.39–1.61, P < 0.01), smoking (OR = 1.10; CI 1.03–1.18, P = 0.06) and CCS class IV symptoms (OR = 1.43; CI 1.34–1.53, P < 0.01). CCS Class IV angina was a stronger predictor of severe CAD amongst women compared with men (women OR = 1.82; CI 1.61–2.04 vs. men OR = 1.28; CI 1.18–1.39, P < 0.01).
Conclusions. Women referred for first diagnostic angiography have lower rates of severe CAD compared with men across all ages. Whilst conventional risk factors, age, sex, diabetes, smoking and hyperlipidaemia are primary determinants of CAD amongst women and men, CCS Class IV angina is more likely to be associated with severe CAD in women than men.
Coronary artery disease (CAD) is the leading cause of mortality and morbidity of both men and women in westernized countries accounting for over one-third of total deaths [1, 2]. In women, the annual mortality rate from CAD is greater than that of breast cancer, even amongst the younger groups (i.e. 35–55 years) [1–6]. Despite this importance of CAD for women, there is a persistent perception that CAD is a ‘man’s disease’. Contributing to this notion is the observation of differences in incidence rates according to age; the incidence of CAD in women is lower than men, but rises steadily after the fifth decade and nearly equalizes between the sexes by the seventh decade of life [5, 6]. Correspondingly, the distribution of CAD risk factors varies between men and women across age ranges and failure to consider these differences may have contributed to the belief that women are at lower risk of CAD compared with men [7–11]. In addition, gender differences in the symptoms of CAD exist between women and men, as women are more likely to have symptoms considered atypical compared with men [3, 5, 12–18]. All of these factors likely contribute to the lower referral rates for coronary angiography amongst women compared with men, even in patients who have severe CAD and acute coronary syndromes (ACS) [3, 7–11, 13]. Younger women with ACS have up to 50% higher risk for mortality than their young male counterparts [19, 20]. The higher risk of mortality in these young women may be due in part to the perception that younger women are at very low risk of CAD and therefore diagnostic and therapeutic management is minimal. On the other hand, a lower rate of angiography may be appropriate because when women do undergo coronary angiography, they are more often reported to have ‘normal’ coronary anatomy and are less likely to have severe CAD (i.e. three vessel and left main disease) compared with men [11–13, 21–25]. Although sex-related differences in risk factors for CAD and the presentation of CAD symptoms are well-known, there is little information on whether the variation in the degree of symptom severity between men and women is associated with differences in CAD severity or if these relationships vary by age. There is an urgent need to better understand the presentation of cardiac symptoms in women in order to facilitate diagnosis and treatment, to initiate aggressive risk factor intervention and to improve the quality of life.
The objectives of our investigation were to examine the distribution of risk factors and coronary angiographic patterns of CAD in women and men who were referred for first diagnostic angiogram and to identify factors associated with severe CAD. Specifically we aimed (i) to investigate sex differences in the distribution of conventional risk factors and the angiographic pattern of CAD in young patients ≤60 years of age compared with older patients >60 years of age, (ii) to examine the factors associated with the presence of severe CAD defined as left main stenosis ≥50%, three-vessel disease with ≥70% stenosis or two-vessel disease including proximal left anterior descending stenosis of ≥70%, and (iii) to evaluate the utility of Canadian Cardiovascular Society (CCS) class angina scoring system in predicting severe CAD amongst a cohort of women and men referred for first diagnostic coronary angiography at a tertiary care institution in Canada.
The study sample included 23 771 consecutive men and women who underwent diagnostic coronary angiography between April 1, 2000 and November 15, 2006. Data used were part of the Hamilton Health Sciences Angiography Registry. Details of the database are described elsewhere . Briefly, the purpose of this prospective registry was to document the characteristics of patients waiting for coronary angiography and to document their subsequent angiographic outcomes. The Hamilton Health Sciences is the sole provider of tertiary cardiac care services including coronary angiography, percutaneous coronary intervention and cardiac surgery for most patients, covering the geographical region of Central-South Ontario, Canada, a population of over 2.2 million people. During the study period, year 2000 to 2006, the mean wait time for cardiac catheterization was 56.0 days for outpatients and 8.4 days for inpatients . No difference in wait times was detected between the sexes . Eligible patients were suspected of having CAD and only those without a prior diagnosis of CAD were included in this analysis. This inclusion criterion was intended to capture patients only with suspected CAD that have not been previously diagnosed with CAD as confirmed by the gold standard, cardiac catheterization. Patients were excluded if they were undergoing coronary angiography for reasons other than diagnosing coronary artery disease such as valvular disease, or if they had a prior or recent evidence of an MI (definition of MI included two out of three criteria, including symptoms, EKG changes and/or biomarkers including elevated troponin), previous coronary artery bypass graft surgery (CABGS) or percutaneous coronary intervention (PCI). The Hamilton Health Sciences Angiography Registry has been approved by the Research Ethics Board.
Patient information was prospectively collected at the time of coronary angiography referral using standardized Hamilton-Wentworth Regional Cardiovascular Program Coronary Angiography Consult Forms distributed to all referring physicians in the region. Information was recorded by the referring physician, electronically entered and edited into a computerized database. The Coronary Angiography Consult forms include patient demographic characteristics, reason for referral (coronary disease, cardiomyopathy, valvular disease, other), state of urgency for coronary angiography, anginal symptom class graded according to the Canadian Cardiovascular Society (CCS) Class Grading system 0–IV and patient risk factor profile; history of smoking, diabetes (insulin dependent or oral medication), hyperlipidaemia and hypertension requiring medical treatment, including a comprehensive list of current medications . Coronary anatomy was graded by the angiographer immediately following the procedure using a standardized diagram . In this analysis, we categorized patients according to severity of disease; severe CAD was defined as left main stenosis ≥50%, three-vessel disease with ≥70% stenosis in at least one vessel, or two-vessel disease, including a proximal left anterior descending (LAD) stenosis ≥70%; moderate risk CAD included two-vessel disease (excluding proximal LAD) with ≥70% lesion, one-vessel disease with ≥70% lesion; low risk CAD was defined as lesions with ≤50% stenosis or normal coronary anatomy. The criterion for ‘severe CAD’ was chosen to characterize those that have prognostically significant CAD in terms of surgical revascularization compared with medical therapy .
All analyses were performed using sas, version 9.1 software (Cary, NC, USA) and spss, version 16.0 (Chicago, IL, USA). Baseline characteristics including age, risk factors, coronary anatomy and CCS symptom class were compared between women and men. Continuous variables were expressed as means with standard deviations and probability estimates were obtained using analysis of variance. Dichotomous variables were expressed as percentages with 95% confidence intervals (CI) and probability values were estimated using logistic regression. To present proportions in risk factor prevalence rates, the data were stratified by sex and age. Patients were categorized and analysed according to CAD risk group; patients with angiographically documented severe CAD were compared with patients with low risk CAD (consisting of moderate and low risk groups). Univariate and multivariate logistic regression were used to determine which of the proposed risk factors (independent variables) were significantly associated with severe CAD. The dependent variable was binary and defined as the presence or absence of severe CAD (1 vs. 0 respectively). Independent variables included age, sex, medically treated risk factors, past or present smoking and severe (CCS Class IV) angina. Logistic models employed a backwards elimination process. Interactions were tested and adjusted for the other risk factors in model. Odds ratios and their accompanying 95% confidence intervals (CI) were calculated. All tests employed two-tailed significance testing.
Baseline characteristics of study population
During the study period, April 2000 to November 2006, 31 758 patients were enrolled in the Hamilton Health Sciences Angiography Registry. For this study, 23 771 are included in the analysis, excluding elective patients with prior MI (n = 1405), prior CABGS (n = 3221) and prior PCI (n = 3361). Of 23 771 study patients, 9112 (38.4%) were women and 14 645 (61.6%) were men. The baseline characteristics are presented in Table 1. Briefly, compared with men, women were significantly older (65.2 ± 12.0 vs. 62.3 ± 12.3, P < 0.01), less likely to be past/present smokers (37.3% vs. 57.2%, P < 0.01) and more likely to be hypertensive (65.9% vs. 57.9%, P < 0.01) (Table 1). Amongst young patients (≤ 60 years) referred to coronary angiography, women were slightly older (51.4 ± 6.1 years vs. 50.9 ± 7.2 years respectively), were more likely to be diabetic (20.7% vs. 16.4%), and hypertensive (53.4% vs. 49.5%), less likely to be past/present smokers (49.9% vs. 62.9%), and to have considerable hyperlipidaemia (57.3% vs. 60.1%) compared with men (Table 1).
Table 1. Baseline characteristics of patients in the Hamilton Health Science Coronary Angiography Registry
Total men N (%)
Men ≤ 60 years
Total women N (%)
Women ≤ 60 years
Total men versus total women
14 645 (61.6)
62.3 ± 12.3
50.9 ± 7.2
65.2 ± 12.0
51.4 ± 6.8
Angiographically, we found women were more likely to have normal/mild CAD (39.7% vs. 21.3%, P < 0.01) and less likely to have severe CAD (36.5% vs. 22.3%, P < 0.01) compared with men (Table 2).
Table 2. Coronary anatomy differences between men and women
Total men N (%)
Total women N (%)
aSevere CAD includes left main stenosis ≥50%, three-vessel disease with ≥70% stenosis or two-vessel disease including proximal left anterior descending stenosis of ≥70%.
2-vessel disease (with prox LAD)
All other 2-vessel disease
Left main disease
Low risk CAD
Some differences in symptom severity in CCS angina classification between the sexes was observed; men were more likely to have CCS Class 0 to II angina (31.2% vs. 29.3%, P < 0.01) and CCS Class IV angina (44.0% vs. 42.9%, P < 0.01) compared with women (Table 3).
Table 3. Frequency of CCS classa angina symptoms according to sex and age
CCS Angina class
All male N = 13 933 (%)
All female N = 8621 (%)
<60 years N = 1326 (%)
≥60 years N = 3066 (%)
<60 years N = 332 (%)
≥60 years N = 1319 (%)
CCS Class I – ordinary physical activity such as walking or climbing stairs does not cause angina; angina with strenuous, rapid or prolonged exertion at work or recreation.
CCS Class II – slight limitation of ordinary activity. Walking or climbing stairs rapidly, walking uphill, walking or stair climbing after meals, or in cold, or in wind or under emotional stress or during the few hours after awakening. Walking more than two blocks on the level and climbing more than one flight of stairs at a normal pace and in normal conditions.
CCS Class III – marked limitation of ordinary physical activity. Walking one or two blocks on the level or climbing one flight of stairs in normal conditions and at a normal pace.
CCS Class IV – inability to carry out any physical activity without discomfort –anginal syndrome may be present at rest.
aCCS Class 0 – asymptomatic.
CCS Class 0–II angina
CCS Class III angina
CCS Class IV angina
Risk factors among young women versus young men with angiographically severe CAD
When stratified by age (≤60 years vs. >60 years) differences in risk factor distribution by the presence of severe angiographic CAD were present. Consistent with the overall observations, younger women were less likely to have severe CAD than younger men (19.9% vs. 30.0%, P < 0.01). Young women were more likely, however, to have diabetes (45.7% vs. 24.7%, P < 0.01) and have hypertension (65.1% vs. 55.7%, P < 0.01) compared with young men. Young men, on the other hand, were more likely to past/present smokers compared with young women (64.7% vs. 58.8%, P = 0.04). There were no statistically significant differences in the proportion of women and men with hyperlipidaemia (72.2% women vs. 71.3% men, P = 0.74) (Table 4).
Table 4. Proportion of risk factors in younger patients versus older patients with severe CAD
≤60 years N (%)
>60 years N (%)
≤60 years women versus men
≤60 years N (%)
>60 years N (%)
Women ≤60 years versus >60 years
53.1 ± 6.0
72.0 ± 6.5
53.1 ± 6.1
73.9 ± 6.1
Risk factors among young women versus older women with severe CAD
As expected, older women referred for first diagnostic angiogram were more likely to have severe CAD (80.1% vs. 19.9%, P < 0.01) compared with younger women. However, younger women with severe CAD were more likely to be diabetic (45.7% vs. 32.2%, P < 0.01) and more likely to be past/present smokers (58.8% vs. 32.7%, P < 0.01), compared with older women. On the other hand, older women were more likely to be hypertensive (77.1% vs. 65.1%, P < 0.01) compared with younger women. There were no statistically significant differences in the proportion of women with hyperlipidaemia between the two age strata (P = 0.73) (Table 4).
Angina severity in patients with angiographically severe CAD
Overall, women with angiographically severe CAD were more likely to have severe angina than men. Specifically, women were more likely to have CCS Class IV angina (56.7% vs. 47.8%, P < 0.01), whereas men were more likely to have CCS Class 0 to II symptoms compared with women (23.5% vs. 17.8%, P < 0.01) (Table 3).
Angina severity among young women versus young men with severe CAD
Younger men were more likely to have less severe symptoms or CCS 0 to II angina compared with women (23.0% vs. 18.7%). On the other hand, young women ≤60 years of age with severe CAD were more likely to have CCS Class IV angina than their young male counterparts (54.5% vs. 49.1%) (Table 3).
Angina severity among young women versus older women with severe CAD
Older women with severe CAD were more likely to have CCS Class IV angina than their younger female counterparts (57.2% vs. 54.5%). However there were no differences between CCS Class 0 to II angina and CCS Class III angina between younger and older women (Table 3).
Risk factors associated with angiographically severe CAD
Factors independently associated with the presence of severe angiographic CAD include age (OR = 1.04, 95% CI 1.04–1.05, P < 0.01), male sex (OR = 2.01, 95% CI 1.88–2.14, P < 0.01), diabetes (OR = 2.09, 95% CI 1.94–2.24, P < 0.01), hyperlipidaemia (OR = 1.69, 95% CI 1.58–1.80, P < 0.01), hypertension (OR = 1.43, 95% CI 1.35–1.53, P < 0.01), smoking (OR = 1.13, 95% CI 1.06–1.20, P < 0.01) and CCS class IV anginal symptoms (OR = 1.43, 95% CI 1.34–1.52, P < 0.01) (Table 5). In an adjusted multivariate logistic regression model, the factors independently associated with severe CAD included age (OR = 1.05, 95% CI 1.05–1.05, P < 0.01), male sex (OR = 2.43, 95% CI 2.26–2.62, P < 0.01), diabetes (OR = 2.00, 95% CI 1.86–2.18, P < 0.01), hyperlipidaemia (OR = 1.50, 95% CI 1.39–1.61, P < 0.01), smoking (OR = 1.10, 95% CI 1.03–1.18, P < 0.01) and CCS class IV symptoms (OR = 1.43, 95% CI 1.34–1.53, P < 0.01) (Table 5). An interaction between CCS Class IV angina and sex was identified (P < 0.01) and indicates that women with CCS class IV angina were more likely to have severe CAD compared with men with CCS class IV angina, (OR = 1.82, 95% CI 1.61–2.04 vs. OR = 1.28, 95% CI 1.18–1.39, P < 0.01 respectively) (refer to Fig. 1).
Table 5. Univariate and adjusted multivariate logistic regression model of variables associated with severe CAD
Odds ratio (95% CI)
Odds ratio (95% CI)
aInteraction terms are adjusted for diabetes, hyperlipidaemia, hypertension, smoke and age. Odds ratio values did not change from presented model.
Sex (men versus women)
CCS Class IV angina
Interaction of CCS class IV angina and sexa
Male × CCS class IV angina
Female × CCS class IV angina
In this prospective registry of over 23 000 individuals referred for first diagnostic coronary angiography for CAD, women were found to be older, have more diabetes and hypertension and were less likely to smoke compared with men. Furthermore, the coronary angiography profile of women indicates that women were more likely to have normal/mild CAD and less likely to have severe angiographic CAD. Conventional risk factors and CCS Class IV symptoms were strong predictors of severe angiographic CAD. The presence of CCS Class IV angina appears to be more predictive of severe angiographic CAD in women compared with men. This information should be used by clinicians when deciding which patients to refer to coronary angiography.
Sex differences in the distribution of risk factors between women and men have been previously reported by several investigators and in prospective studies [3–5, 7, 13, 15, 17, 19–21, 24, 25, 30–35]. Despite differences between studies, our findings are consistent with previous reports. Our observation that women are older at time of first referral lends supports to the observation that women develop CAD later in life than men. Moreover, the risk factor profile amongst women within this cohort is also consistent with prior literature supporting women at time of referral are more often hypertensive, whereas men are more likely to smoke.
The proportion of patients with severe angiographic CAD is greater in men than women. A 20% excess in the prevalence of severe CAD remains even after adjustment for age and other risk factors. It is possible that other sex differences may make men more prone to develop obstructive CAD. Whilst the proportion of women with severe angiographic CAD is lower than men, amongst young patients ≤60 years of age with suspected CAD, there is no difference attributed to age. In particular, within this subgroup of young women with severe angiographic CAD, the proportion with diabetes is almost two-fold times higher than amongst men, and the prevalence of hypertension, smoking and elevated lipids is particularly high, over 50%. Prior studies note that the protective ‘female advantage’ of lower CAD prevalence is essentially eliminated amongst diabetic women [5, 36, 37], and that the 10 years of delayed onset in CAD between women and men is largely explained by more frequent risk factors amongst men at younger ages . Unfortunately, diabetic women may receive less treatment and CAD risk factor modification than diabetic men, because of the perception that women are at lower risk of CAD . Despite the perception that younger women are less likely to have severe CAD compared with men, our data emphasize that women with diabetes and other proven cardiovascular risk factors should be considered carefully for diagnostic coronary angiography particularly if they have symptoms suggestive of CAD, regardless of their age and sex. This would allow the appropriate patients to undergo further revascularization that may alter symptoms and/or prognosis .
Whilst women with risk factors have a greater probability of having severe CAD compared with women without risk factors, we and others have observed that women referred for diagnostic coronary angiography, in the outpatient as well as in the ACS setting, are more likely to have normal coronary arteries . Existing evidence suggests that women may indeed experience typical symptoms of angina in the presence of normal coronaries because they suffer vasospasm, excessive plaque/nonobstructive disease and/or endothelial dysfunction [9, 10, 39–41]; the question of the utility of typical symptoms of angina in predicting the presence of severe CAD has been raised [9, 13, 30, 42]. In our study, we observed that the presence of severe anginal (CCS Class IV) symptoms using a standardized symptom classification is a useful predictor, despite the prior controversies reported in the literature [3, 12, 13, 21–23]. In fact, despite a slightly higher and statistically significant proportion of men presenting with CCS Class IV symptoms, the association of CCS class IV symptoms and severe CAD was stronger in women than in men in our study. This finding is of particular interest, as women are often reported to have more atypical symptoms with less severe obstructive CAD. This has been identified as a ‘paradox’, where the prevalence of angina in women is similar to that of men, although men are more often found to have angiographically demonstrated CAD . Our data confirm that women with risk factors and severe angina, particularly CCS Class IV angina, are more likely to have angiographically documented severe CAD. Further, this finding is particularly important since the definition of ‘severe CAD’ used reflects standard criterion and has prognostic significance . Researchers in women’s cardiac health have called for the ‘imminent need’ for better clinical classification to predict the presence or absence of severe CAD in women with suspected CAD . Our data should aid clinicians in determining who should be referred for diagnostic angiography amongst at-risk women with suspected CAD.
It is important to note that although there are limitations inherent to all database analyses our prospective registry represents ‘real world’ clinical practise. Whilst referral bias may limit the external validity, our results are internally valid. Despite the fact that women are referred less often for catheterization, within the ones who were referred, we demonstrate that more women are likely to have normal/mild CAD, and less likely to have severe CAD. In this analysis, patient referrals are dependent on ‘real world’ physician decision making rather than protocol driven angiograms. Consequently, we relied on the summarized reporting of the risk factors by the referring physician and did not have actual laboratory values for diabetes and hyperlipidaemia, or blood pressure readings for hypertensive patients. However, we cross-checked the reported risk factors with medical treatment to minimize this potential bias and as such, our prevalence estimates may be underestimates. Also, much like real world practise, the severity of angiographic stenosis was solely determined by the angiographer performing the procedure. However, it remains unknown if the implications of equal burden of CAD are similar in both men and women. Although the majority of our patients likely had pre-catheterization noninvasive testing, we did not have access to these data in order to determine if symptoms were strongly correlated with noninvasive testing results or if sex differences were present. However our data are internally valid and clearly demonstrate the predictive nature of CCS class IV symptoms. Lastly, the primary purpose of the registry was to document adverse events whilst patients were waiting for their angiogram and follow-up data were not collected. Despite this, it is important to note that this study represents one of the largest series of women analysed alongside men.
Women referred for first diagnostic angiography are more likely to have normal/mild CAD and to have lower rates of severe CAD compared with men across all ages. Whilst conventional risk factors including age, sex, diabetes, smoking and hyperlipidaemia are primary determinants of CAD, CCS Class IV angina is more strongly associated with severe CAD amongst women than in men. These findings have implications for physicians to better identify women at risk and to target diagnostic and treatment strategies accordingly.
Conflict of interest statement
There are no conflicts of interest to report from any of the authors.
We would like to thank the Hamilton Health Sciences and Population Health Research Institute for their database support.