The Importance of Nontraditional and Sex-Specific Risk Factors in Young Women With Vasomotor Nonobstructive vs Obstructive Coronary Syndromes.

A total of 114 women from the WHC cohort had premature (M)INOCA from a vasomotor entity (n = 37 MINOCA; n = 77 INOCA), and 140 women from the SAVEBC cohort had premature obstructive CAD (n = 74 MI-CAD; n = 66 non-MI-CAD) and met the inclusion criteria, as shown in Figure 1 . Figure 1 Flow diagram of exclusion steps by which eligible S tudy to A void Cardio v ascular E vents in B ritish C olumbia (SAVEBC) and Leslie Diamond Women’s Heart Health Clinic Registry (WHC) patients were filtered. CA, coronary angiogram; CAD, coronary artery disease; CMD, coronary microvascular dysfunction; CTA, computed tomography coronary angiogram; INOCA, ischemia with no obstructive coronary arteries; MI, myocardial infarction; MINOCA, MI with no obstructive coronary arteries; (M)INOCA, MINOCA or INOCA; (N)STEMI, (non-)ST-elevation myocardial infarction; SCAD, spontaneous coronary artery dissection; WHC, Leslie Diamond Women’s Heart Health Clinic Registry. Flow diagram of exclusion steps by which eligible S tudy to A void Cardio v ascular E vents in B ritish C olumbia (SAVEBC) and Leslie Diamond Women’s Heart Health Clinic Registry (WHC) patients were filtered. CA, coronary angiogram; CAD, coronary artery disease; CMD, coronary microvascular dysfunction; CTA, computed tomography coronary angiogram; INOCA, ischemia with no obstructive coronary arteries; MI, myocardial infarction; MINOCA, MI with no obstructive coronary arteries; (M)INOCA, MINOCA or INOCA; (N)STEMI, (non-)ST-elevation myocardial infarction; SCAD, spontaneous coronary artery dissection; WHC, Leslie Diamond Women’s Heart Health Clinic Registry. In the WHC cohort, most women with MINOCA had probable or definite vasospasm (19; 51%); the rest had probable or definitive CMD (9; 24%) or an undetermined diagnosis of probable CMD or vasospasm still awaiting final testing, but with non-vasomotor entities ruled out (9; 24%). Most women with INOCA had CMD (52; 68%), and the rest had vasospasm (17; 22%) and an undetermined diagnosis of CMD or vasospasm (8; 10%). In the SAVEBC cohort, 50 women with MI-CAD presented with non-ST elevation MI (68%) and the other 24 with ST-elevation MI (32%). Among the non-MI-CAD patients, 35 presented with stable angina (53%), 22 with unstable angina (33%), and 9 with other indications for undergoing coronary angiography (14%). Table 1 summarizes the frequency of demographic and clinical variables across diagnostic subgroups. The median ages (interquartile range) at the first coronary angiography in each subgroup of women were 49 years (42-55 years) for those with MINOCA, 50 years (41-55 years) for those with INOCA, 50 years (44-55 years) for those with MI-CAD, and 50.5 years (44.5-55 years) for those with non-MI-CAD. Significant differences were found between women with (M)INOCA (from the WHC), compared to those with obstructive CAD (from SAVEBC) as follows: women with (M)INOCA had higher proportions of self-reported Caucasian race ( P < 0.01); being partnered ( P < 0.01); history of migraines ( P < 0.001); preeclampsia or gestational hypertension ( P  = 0.04); and more children. In contrast, women with obstructive CAD had significantly higher proportions of dyslipidemia ( P < 0.001), diabetes ( P < 0.001), hypertension ( P < 0.01), obesity ( P < 0.001), and smoking history ( P < 0.001). Table 1 Frequencies of risk factors in the Leslie Diamond Women’s Heart Health Clinic Registry (WHC) and S tudy to A void Cardio v ascular E vents in B ritish C olumbia (SAVEBC) cohorts further stratified by diagnostic subgroups (MINOCA, INOCA, MI-CAD, non-MI-CAD) Variable WHC SAVEBC WHC (n = 114) SAVEBC (n = 140) P (WHC vs SAVEBC) MINOCA at age ≤ 55 y (n = 37) INOCA at age ≤ 55 y (n = 77) MI-CAD at age ≤ 55 y (n = 74) Non-MI CAD at age ≤ 55 y (n = 66) Age at first CA/CTA, y (IQR) 49 (42, 55) 50 (41, 55) 50 (44, 55) 50.5 (44.5, 55) 49 (41, 55) 50 (44, 55) 0.08 Race  Caucasian (“White”) 24 (65) 56 (73) 30 (41) 27 (41) 80 (70) 57 (41) < 0.01  Non-Caucasian 11 (30) 20 (26) 29 (39) 25 (38) 31 (27) 54 (39)  South Asian 2 (5) 7 (9) 7 (9) 5 (8) 9 (8) 12 (9)  East Asian (“Chinese, Japanese, Korean”) 5 (14) 2 (3) 5 (7) 4 (6) 7 (6) 9 (6)  First Nations (“Aboriginal”) 0 2 (3) 6 (8) 2 (3) 2 (2) 8 (6)  African-Canadian (“Black”) 0 1 (1) 0 0 1 (1) 0  Other (including mixed race) 4 (11) 8 (10) 11 (15) 14 (21) 12 (11) 25 (18)  Missing ∗ 2 (5) 1 (1) 15 (20) 14 (21) 3 (3) 29 (21) Partnered (married, common law) 36 (97) 61 (79) 36 (49) 32 (48) 97 (85) 68 (49) < 0.01  Missing 0 0 18 (24) 20 (30) 0 38 (27) Dyslipidemia 14 (38) 39 (51) 46 (62) 48 (73) 53 (46) 94 (67) < 0.001 Diabetes 3 (8) 10 (13) 24 (32) 22 (33) 13 (11) 46 (33) < 0.001  Type 1 0 2 (3) 0 2 (3) 2 (2) 2 (1)  Type 2 3 (8) 8 (10) 24 (32) 20 (30) 11 (10) 44 (31) Gestational diabetes 3 (8) 8 (10) 9 (12) 16 (24) 11 (10) 25 (18) Hypertension 13 (35) 30 (39) 42 (57) 34 (52) 43 (38) 76 (54) < 0.01 Obesity (BMI ≥ 30) 8 (22) 17 (22) 35 (47) 26 (39) 25 (22) 61 (44) < 0.001  Missing 0 0 5 (7) 1 (2) 0 6 (4) Smoking history  Never 33 (89) 66 (86) 34 (46) 38 (58) 99 (87) 72 (51) < 0.001  Current 1 (3) 2 (3) 17 (23) 9 (14) 3 (3) 26 (19)  Previous 3 (8) 9 (12) 23 (31) 19 (29) 12 (11) 42 (30) Family history of premature CVD 16 (43) 31 (40) 31 (42) 32 (48) 47 (41) 63 (45) 0.1  Missing 0 0 15 (20) 5 (8) 0 20 (14) Depression and/or anxiety 13 (35) 26 (34) 22 (30) 19 (29) 39 (34) 41 (29) 0.54 Hypothyroidism 2 (5) 14 (18) 12 (16) 10 (15) 16 (14) 22 (16) 0.6 Migraines 19 (51) 23 (30) 2 (3) 5 (8) 42 (37) 7 (5) < 0.001 Autoimmune disorder(s) (any) 3 (8) 14 (18) 18 (24) 12 (18) 17 (15) 30 (21) 0.21 Preeclampsia or gestational hypertension 7 (19) 6 (8) 3 (4) 2 (3) 13 (11) 5 (4) 0.04  Missing 4 (11) 10 (13) 19 (26) 15 (23) 14 (12) 34 (24) Number of children  0 or 1 8 (22) 12 (16) 21 (28) 21 (32) 20 (18) 42 (30) 0.04  2 or 3 22 (59) 50 (65) 36 (49) 31 (47) 72 (63) 67 (48)  4+ 3 (8) 5 (6) 6 (8) 3 (5) 8 (7) 9 (6)  Missing 4 (11) 10 (13) 11 (15) 11 (17) 14 (12) 22 (16) Number of nonterm pregnancies 0.68  0 18 (49) 25 (32) 32 (43) 33 (50) 43 (38) 65 (46)  1 7 (19) 16 (21) 16 (22) 14 (21) 23 (20) 30 (21)  2+ 2 (5) 18 (23) 14 (19) 8 (12) 20 (18) 22 (16)  Missing 10 (27) 18 (23) 12 (16) 11 (17) 28 (25) 23 (16) Values are n (%), unless otherwise indicated. BMI, body mass index; CA, coronary angiogram; CAD, coronary artery disease; CTA, computed tomography coronary angiogram; CVD, cardiovascular disease; INOCA, ischemia with no obstructive coronary arteries; IQR, interquartile range; MI, myocardial infarction; MINOCA, MI with no obstructive coronary arteries. ∗ Missing proportions provided when > 5%. Frequencies of risk factors in the Leslie Diamond Women’s Heart Health Clinic Registry (WHC) and S tudy to A void Cardio v ascular E vents in B ritish C olumbia (SAVEBC) cohorts further stratified by diagnostic subgroups (MINOCA, INOCA, MI-CAD, non-MI-CAD) Values are n (%), unless otherwise indicated. BMI, body mass index; CA, coronary angiogram; CAD, coronary artery disease; CTA, computed tomography coronary angiogram; CVD, cardiovascular disease; INOCA, ischemia with no obstructive coronary arteries; IQR, interquartile range; MI, myocardial infarction; MINOCA, MI with no obstructive coronary arteries. Missing proportions provided when > 5%. The cumulative burden of traditional and nontraditional risk factors is summarized in Tables 2 and 3 , respectively, with relative proportions shown in Figure 2 (traditional risk factors) and Figure 3 (nontraditional risk factors). Χ 2 tests comparing the cumulative burden of risk factors demonstrate significant differences between cohorts in the number of traditional risk factors ( Fig. 2 A; P < 0.001) and nontraditional risk factors ( Fig. 3 A; P  = 0.03). The most frequent number of traditional risk factors seen per patient was higher for the patients with obstructive CAD (mode = 3), compared to that for women with nonobstructive CAD (mode = 1). Most strikingly, over 40% of the MINOCA group had just one traditional risk factor. The opposite trend was seen with nontraditional risk factors: in women with nonobstructive CAD, the most common number of risk factors was one; in women with obstructive CAD, the most common number was zero. Table 2 Traditional cardiovascular risk factors stratified by cohort ( S tudy to A void Cardio v ascular E vents in B ritish C olumbia [SAVEBC] vs Leslie Diamond Women’s Heart Health Clinic Registry [WHC]) and by diagnostic subgroup (MI-CAD, non-MI-CAD, MINOCA, INOCA) Cohort Diagnosis Traditional cardiovascular risk factors ∗ , n (%) 0 1 2 3 4 5 6 WHC MINOCA 5 (13.5) 15 (40.5) 10 (27.0) 6 (16.2) 0 1 (2.7) 0 INOCA 17 (22.1) 19 (24.7) 15 (19.5) 16 (20.8) 9 (11.7) 1 (1.3) 0 Total 22 (19.3) 34 (29.8) 25 (21.9) 22 (19.3) 9 (7.9) 2 (1.8) 0 SAVEBC MI-CAD 3 (4.1) 8 (10.8) 16 (21.6) 23 (31.1) 13 (17.6) 9 (12.2) 2 (2.7) Non-MI-CAD 2 (3.0) 9 (13.6) 14 (21.2) 20 (30.3) 14 (21.2) 5 (7.6) 2 (3.0) Total 5 (3.6) 17 (12.1) 30 (21.4) 43 (30.7) 27 (19.3) 14 (10.0) 4 (2.9) CAD, coronary artery disease; INOCA, ischemia with no obstructive coronary arteries; MI, myocardial infarction; MINOCA, MI with no obstructive coronary arteries. ∗ Traditional risk factors quantified were dyslipidemia, diabetes (type 1 or 2), hypertension, obesity, past or current smoker, and family history of premature cardiovascular disease. Table 3 Nontraditional cardiovascular risk factors stratified by cohort ( S tudy to A void Cardio v ascular E vents in B ritish C olumbia [SAVEBC] vs Leslie Diamond Women’s Heart Health Clinic Registry [WHC]) and by diagnostic subgroup (MI-CAD, non-MI-CAD, MINOCA, INOCA) Nontraditional cardiovascular risk factors, ∗ n (%) Cohort Diagnosis 0 1 2 3 4 5 WHC MINOCA 11 (29.7) 12 (32.4) 11 (29.7) 2 (5.4) 1 (2.7) 0 INOCA 24 (31.2) 29 (37.7) 18 (25.4) 6 (7.8) 0 0 Total 35 (30.7) 41 (36.0) 29 (25.4) 8 (7.0) 1 (0.9) 0 SAVEBC MI-CAD 35 (47.3) 23 (31.1) 14 (18.9) 2 (2.7) 0 0 Non-MI CAD 30 (45.5) 26 (39.4) 8 (12.1) 2 (3.0) 0 0 Total 65 (46.4) 49 (35.0) 22 (15.7) 4 (2.9) 0 0 CAD, coronary artery disease; INOCA, ischemia with no obstructive coronary arteries; MI, myocardial infarction; MINOCA, MI with no obstructive coronary arteries. ∗ Nontraditional cardiovascular risk factors quantified were depression and/or anxiety, any autoimmune disease, hypothyroidism, migraines, and preeclampsia and/or gestational hypertension. Figure 2 Histograms of total traditional cardiovascular (CV) risk factors by ( A ) cohort and ( B ) diagnostic subgroup (MI-CAD vs non-MI-CAD vs MINOCA vs INOCA). The traditional CV risk factors quantified were as follows: dyslipidemia, diabetes (type 1 or 2), hypertension, obesity, past or current smoker, and family history of premature CV disease. CAD, coronary artery disease; INOCA, ischemia with no obstructive coronary arteries; MI, myocardial infarction; MINOCA, MI with no obstructive coronary arteries; SAVEBC, S tudy to A void Cardio v ascular E vents in B ritish C olumbia; WHC, Leslie Diamond Women’s Heart Health Clinic Registry. Figure 3 Histograms of total nontraditional cardiovascular (CV) risk factors by ( A ) cohort and ( B ) diagnostic subgroup (MI-CAD vs non-MI-CAD vs MINOCA vs INOCA). The nontraditional CV risk factors quantified were as follows: depression and/or anxiety, any autoimmune disease, hypothyroidism, migraines, and preeclampsia/gestational hypertension. CAD, coronary artery disease; INOCA, ischemia with no obstructive coronary arteries; MI, myocardial infarction; MINOCA, MI with no obstructive coronary arteries; SAVEBC, S tudy to A void Cardio v ascular E vents in B ritish C olumbia; WHC, Leslie Diamond Women’s Heart Health Clinic Registry. Traditional cardiovascular risk factors stratified by cohort ( S tudy to A void Cardio v ascular E vents in B ritish C olumbia [SAVEBC] vs Leslie Diamond Women’s Heart Health Clinic Registry [WHC]) and by diagnostic subgroup (MI-CAD, non-MI-CAD, MINOCA, INOCA) CAD, coronary artery disease; INOCA, ischemia with no obstructive coronary arteries; MI, myocardial infarction; MINOCA, MI with no obstructive coronary arteries. Traditional risk factors quantified were dyslipidemia, diabetes (type 1 or 2), hypertension, obesity, past or current smoker, and family history of premature cardiovascular disease. Nontraditional cardiovascular risk factors stratified by cohort ( S tudy to A void Cardio v ascular E vents in B ritish C olumbia [SAVEBC] vs Leslie Diamond Women’s Heart Health Clinic Registry [WHC]) and by diagnostic subgroup (MI-CAD, non-MI-CAD, MINOCA, INOCA) CAD, coronary artery disease; INOCA, ischemia with no obstructive coronary arteries; MI, myocardial infarction; MINOCA, MI with no obstructive coronary arteries. Nontraditional cardiovascular risk factors quantified were depression and/or anxiety, any autoimmune disease, hypothyroidism, migraines, and preeclampsia and/or gestational hypertension. Histograms of total traditional cardiovascular (CV) risk factors by ( A ) cohort and ( B ) diagnostic subgroup (MI-CAD vs non-MI-CAD vs MINOCA vs INOCA). The traditional CV risk factors quantified were as follows: dyslipidemia, diabetes (type 1 or 2), hypertension, obesity, past or current smoker, and family history of premature CV disease. CAD, coronary artery disease; INOCA, ischemia with no obstructive coronary arteries; MI, myocardial infarction; MINOCA, MI with no obstructive coronary arteries; SAVEBC, S tudy to A void Cardio v ascular E vents in B ritish C olumbia; WHC, Leslie Diamond Women’s Heart Health Clinic Registry. Histograms of total nontraditional cardiovascular (CV) risk factors by ( A ) cohort and ( B ) diagnostic subgroup (MI-CAD vs non-MI-CAD vs MINOCA vs INOCA). The nontraditional CV risk factors quantified were as follows: depression and/or anxiety, any autoimmune disease, hypothyroidism, migraines, and preeclampsia/gestational hypertension. CAD, coronary artery disease; INOCA, ischemia with no obstructive coronary arteries; MI, myocardial infarction; MINOCA, MI with no obstructive coronary arteries; SAVEBC, S tudy to A void Cardio v ascular E vents in B ritish C olumbia; WHC, Leslie Diamond Women’s Heart Health Clinic Registry. Among the 30% of women from the WHC study with only one traditional risk factor, family history of premature CVD was the most common (38%), followed by dyslipidemia (24%), and then hypertension and obesity (15% each). Among the 12% of patients in SAVEBC with only one traditional risk factor, the most common were dyslipidemia (29%) and smoking history (29%), followed by family history of premature CVD (18%). Having 5 or 6 traditional CV risk factors was observed in over 6 times more patients in the SAVEBC cohort (12.9%) than in the WHC cohort (1.8%). For nontraditional CV risk factors, the most frequently reported number of risk factors among WHC patients was one (36%). The most common was a history of migraines (37%), followed by depression and/or anxiety (34%) and hypothyroidism (12%). Among the SAVEBC patients, 46.5% had zero nontraditional risk factors, but among the 35% with only one nontraditional risk factor, the most common was depression and/or anxiety (45%), followed by any autoimmune disease (31%) and hypothyroidism (18%). Having 3 or 4 nontraditional CV risk factors was observed over twice as often in the WHC patients (7.9%) compared to the SAVEBC patients (2.9%) Odds ratios (ORs) and 95% confidence intervals (CIs) from the univariable regression models are summarized in Table 4 . Most traditional risk factors, except for family history of CVD, were significantly associated with a reduced odds of (M)INOCA compared to any obstructive CAD, as shown in Figure 4 A. The risk factors most strongly associated with an increased odds of (M)INOCA (OR > 2) were history of migraines (OR 10.50 [95% CI 4.75, 26.66], P < 0.001); preeclampsia or gestational hypertension (OR 3.02 [95% CI 1.09, 9.72], P  = 0.04); and having 2-3 children (OR 2.26 [95% CI 1.22, 4.29], P  = 0.011). Table 4 Association between risk factors and selected diagnoses Outcome (M)INOCA (ref: any obstructive CAD ∗ ) OR (95% CI) INOCA (ref: non-MI CAD) OR (95% CI) MINOCA (ref: MI-CAD) OR (95% CI) Sociodemographic variables  Age at earliest CA/CTA (per y) 0.94 (0.90, 0.99) 0.95 (0.89, 1.00) 0.94 (0.87, 1.00)  Race: non-Caucasian 0.41 (0.23, 0.71) 0.39 (0.18, 0.81) 0.47 (0.19, 1.12)  Not partnered 0.35 (0.18, 0.67) 0.60 (0.26, 1.38) 0.05 (0.00, 0.26) Traditional cardiovascular risk factors  Dyslipidemia 0.43 (0.25, 0.71) 0.38 (0.19, 0.78) 0.37 (0.16, 0.83)  Diabetes 0.26 (0.13, 0.50) 0.30 (0.13, 0.69) 0.18 (0.04, 0.58)  Hypertension 0.51 (0.31, 0.84) 0.60 (0.31, 1.17) 0.41 (0.18, 0.92)  Obese (BMI ≥ 30) 0.34 (0.19, 0.58) 0.43 (0.20, 0.88) 0.27 (0.10, 0.65)  Family history of premature CVD 0.64 (0.38, 1.08) 0.62 (0.32, 1.23) 0.69 (0.30, 1.57)  Smoking history: current 0.08 (0.02, 0.25) 0.13 (0.03, 0.62) 0.06 (0.00, 0.32)  Smoking history: former 0.21 (0.10, 0.41) 0.27 (0.11, 0.66) 0.13 (0.03, 0.43) Nontraditional risk factors and female-specific variables  Depression and/or anxiety 1.18 (0.69, 2.01) 1.21 (0.59, 2.47) 1.18 (0.50, 2.73)  Hypothyroidism 0.83 (0.41, 1.66) 1.20 (0.49, 2.92) 0.28 (0.04, 1.09)  Migraines 10.50 (4.75, 26.66) 4.94 (1.75, 13.92) 36.89 (9.30, 238.93)  Any autoimmune disorder(s) 0.66 (0.33, 1.25) 1.02 (0.43, 2.38) 0.28 (0.06, 0.92)  Preeclampsia and/or gestational hypertension 3.02 (1.09, 9.72) 2.41 (0.47, 12.47) 4.67 (1.19, 23.05)  Number of children: 2 to 3 2.26 (1.22, 4.29) 2.82 (1.22, 6.53) 1.60 (0.62, 4.42)  Number of children: 4+ 1.87 (0.62, 5.61) 2.92 (0.59, 14.41) 1.31 (0.23, 6.37)  Number of nonterm pregnancies: 1 1.16 (0.59, 2.25) 1.51 (0.62, 3.66) 0.78 (0.26, 2.20)  Number of nonterm pregnancies: 2+ 1.37 (0.67, 2.82) 2.97 (1.11, 7.93) 0.25 (0.04, 1.05) OR and 95% CI are from univariable logistic regression models. Boldface indicates statistical significance of P < 0.05. BMI, body mass index; CA, coronary angiogram; CAD, coronary artery disease; CTA, computed tomography coronary angiogram; CI, confidence interval; CVD, cardiovascular disease; INOCA, ischemia with no obstructive coronary arteries; MI, myocardial infarction; MINOCA, MI with no obstructive coronary arteries; (M)INOCA, MINOCA or INOCA; OR, odds ratio; ref, referent. ∗ Any obstructive CAD: includes MI-CAD and non-MI CAD. Figure 4 Forest plots of odds ratios with 95% CI for (M)INOCA compared to any obstructive CAD. ( A) Univariable odds ratios; ( B ) multivariable (adjusted) odds ratios. CA, coronary angiogram; CI, confidence interval; CAD, coronary artery disease; CTA, computed tomography coronary angiogram; CVD, cardiovascular disease; (M)INOCA, myocardial infarction with no obstructive coronary arteries (MINOCA) or ischemia with no obstructive coronary arteries (INOCA); T1, Type 1; T2, Type 2. Association between risk factors and selected diagnoses OR and 95% CI are from univariable logistic regression models. Boldface indicates statistical significance of P < 0.05. BMI, body mass index; CA, coronary angiogram; CAD, coronary artery disease; CTA, computed tomography coronary angiogram; CI, confidence interval; CVD, cardiovascular disease; INOCA, ischemia with no obstructive coronary arteries; MI, myocardial infarction; MINOCA, MI with no obstructive coronary arteries; (M)INOCA, MINOCA or INOCA; OR, odds ratio; ref, referent. Any obstructive CAD: includes MI-CAD and non-MI CAD. Forest plots of odds ratios with 95% CI for (M)INOCA compared to any obstructive CAD. ( A) Univariable odds ratios; ( B ) multivariable (adjusted) odds ratios. CA, coronary angiogram; CI, confidence interval; CAD, coronary artery disease; CTA, computed tomography coronary angiogram; CVD, cardiovascular disease; (M)INOCA, myocardial infarction with no obstructive coronary arteries (MINOCA) or ischemia with no obstructive coronary arteries (INOCA); T1, Type 1; T2, Type 2. The following risk factors, which are known to be strongly associated with obstructive CAD, were strongly and negatively associated with (M)INOCA as compared to their obstructive counterparts (OR < 0.5): current smoker (OR 0.08 [95% CI 0.02, 0.25], P < 0.001); former smoker (OR 0.21 [0.10, 0.41], P < 0.001); type 1 or type 2 diabetes (OR 0.26 [95% CI 0.13, 0.50], P < 0.001); obesity (OR 0.34 [95% CI 0.19, 0.58], P < 0.001); not partnered (OR 0.35 [95% CI 0.18, 0.67], P  = 0.002); non-Caucasian race (OR 0.41 [95% CI 0.23, 0.71], P  = 0.002; and dyslipidemia (OR 0.43 [95% CI 0.25, 0.71], P  = 0.001). Risk factors that were not significantly associated with (M)INOCA compared to any obstructive CAD were family history of CVD, depression and/or anxiety, hypothyroidism, and any autoimmune disease. After iterative multivariable regression models were fitted and assessed, the final model of (M)INOCA relative to any obstructive CAD included the following ( Fig. 4 B): age of earliest angiography (OR 0.96 [95% CI 0.90, 1.03], P  = 0.25; diabetes (OR 0.15 [95% CI 0.05, 0.40], P  < 0.001); depression and/or anxiety (OR 2.75 [95% CI 1.16, 6.92], P  = 0.025; migraines (OR 6.84 [95% CI 2.56, 21.74], P < 0.001; current smoking history (OR 0.06 [95% CI 0.01, 0.24], P < 0.001; previous smoking history (OR 0.16 [95% CI 0.06, 0.42], P < 0.001; and preeclampsia or gestational hypertension (OR 5.85 [95% CI 1.40, 30.58], P  = 0.022).

This study included women with (M)INOCA enrolled in the Leslie Diamond Women’s Heart Health Clinic Registry (WHC) and women with obstructive CAD in the S tudy to A void Cardio v ascular E vents in BC (SAVEBC) biobank. Eligibility criteria for this study were met if women from either study had signs and symptoms of ischemia and/or an acute coronary syndrome and had undergone coronary angiography (CA) and/or cardiac CT angiography (CTA) before or at age 55 years to delineate their coronary anatomy. The earliest date of CA/CTA was February 1995 for the WHC and September 2015 for SAVEBC; the last date for both cohorts was July 10, 2022. For the WHC, women with MINOCA or INOCA and final diagnoses of definite or probable coronary vasospasm or CMD were included. MINOCA was defined per the 2019 scientific statement from the American Heart Association, 21 and INOCA was defined per the 2020 European Society of Percutaneous Cardiovascular Interventions (EAPCI) expert consensus document on ischemia with nonobstructive coronary arteries. 22 Probable and confirmed CMD and coronary vasospasm were defined using the Co ronary Va somotion D isorder I nternational S tudy (COVADIS) group definition. 23 WHC patients were excluded if their diagnosis after the first CA/CTA was obstructive CAD, non-(M)INOCA cardiac (or noncardiac) entities (ie, arrythmia, esophageal spasm), or if the final diagnoses were those other than vasomotor etiologies (ie, SCAD). For the SAVEBC study, female patients with obstructive CAD (defined as any epicardial vessel with ≥ 50% stenosis) were included in this study. SAVEBC patients were excluded if their presentation had any nonatherosclerotic component (ie, SCAD, vasospasm). The WHC is a quaternary outpatient cardiology clinic located at Vancouver General Hospital in BC, Canada, comprising a multidisciplinary team of cardiologists, nurse practitioners, and a psychiatrist specializing in women’s heart health. The clinic has access to specialized testing for (M)INOCA, including coronary reactivity testing, optical coherence tomography (OCT), and adenosine cardiac magnetic resonance imaging (MRI). The WHC registry began recruitment of patients referred to the clinic in 2016 and has 322 patients enrolled, of which 259 patients (80%) have (M)INOCA. Data are collected from physician consult notes and diagnostic test reports uploaded to electronic medical records, including information about comorbidities (including CV risk factors) and cardiac presentation characteristics. The SAVEBC biobank began recruitment in 2015 and is a prospective study of patients who present with premature CAD with stenosis of at least 50% in at least one epicardial artery at age ≤ 50 years for male patients and age ≤ 55 years for female patients. A detailed description of the study design and rationale is provided elsewhere. 24 Briefly, and in contrast to those in the WHC, patients in the SAVEBC study were recruited from cardiac catheterization laboratories and cardiology wards at Vancouver General Hospital, St. Paul’s Hospital (Vancouver, BC) and Kelowna General Hospital (Kelowna, BC). The SAVEBC study also collects information on CV risk factors, comorbidities, presentation characteristics, physical examination, and laboratory test results. This information is collected directly from patients during the study visits and from paper-based and electronic medical records, including medical charts, and pharmacy dispensing records. In addition, reports are collected from the Cardiac Services BC Registry, a province-wide electronic information system collecting information on all patients who have received cardiac procedures (coronary angiography, percutaneous coronary interventions, coronary artery bypass surgery, valve procedures, and implantable devices) in the province. Shared demographic and clinical covariates that are collected in both cohorts were identified and then extracted. Variable names and definitions were standardized prior to merging the 2 datasets. Sociodemographic variables were consolidated between patient self-report on study questionnaires and any mention in physician consult notes. Self-reported race (Caucasian, South Asian, East Asian, First Nations, African-Canadian, and Other) was summarized as Caucasian or non-Caucasian due to the high proportions of Caucasians in both cohorts. Partnered status was recorded as partnered (married or common law) or not partnered (single, divorced, or widowed). Traditional CV risk factors included the following: dyslipidemia (total cholesterol ≥ 240 mg/dL [6.2 mmol/L], low-density lipoprotein cholesterol ≥ 160 mg/dL [4.1 mmol/L], high-density lipoprotein cholesterol ≤ 40 mg/dL [1.0 mmol/L]; triglycerides ≥ 200 mg/dL [2.3 mmol/L], or treatment of dyslipidemia 25 ); hypertension (systolic/diastolic blood pressure ≥ 140/90 mm Hg per the Canadian hypertension guidelines 26 ); diabetes (type 1 or 2: fasting plasma glucose ≥ 126 mg/dL (7 mmol/L) in at least 2 baseline measurements, hemoglobin A1c ≥ 6.5%, physician diagnosis, or treatment of diabetes 27 ); family history of premature CV disease (CVD; age < 55 years for men, < 65 years for women); obesity (body mass index ≥ 30 kg/m 2 ); and smoking (former, current, or never). Nontraditional clinical risk factors and female-specific risk modifiers included the following: history of a clinical diagnosis of depression and/or anxiety, hypothyroidism, diagnosis of migraines (with or without aura), diagnosis of an autoimmune disease, preeclampsia (gestational systolic/diastolic blood pressure ≥ 140/90 mm Hg and proteinuria per the American College of Obstetricians and Gynecologists guidelines 28 ) or gestational hypertension, gestational diabetes, number of children (0-1, 2-3, 4+), and number of nonterm pregnancies, such as miscarriage, abortion, and/or stillbirth (0, 1, or 2+). Baseline characteristics were reported as medians for continuous variables, and counts with proportions for categorical variables. Cumulative burden was calculated by summing the number of traditional CV risk factors (range 0-6) and nontraditional CV risk factors (range 0-5). The main comparison was between the 2 cohorts; both cohorts were further stratified by the presence vs absence of MI. Categorical variables were compared using χ 2 tests; continuous variables were compared using the Kruskal-Wallis test. To assess the strength of the association between each risk factor and the relevant diagnoses, several univariable logistic regression tests were conducted. The primary comparisons of interest were associations with (M)INOCA relative to any obstructive CAD. The associations between each risk factor and diagnoses were then assessed for INOCA relative to non-MI-CAD, and MINOCA relative to MI-CAD, with the associations expressed as odds. Iterative, multivariable logistic regression models were constructed to identify factors independently associated with (M)INOCA compared to any obstructive CAD. In the first model, covariates were selected based on clinical importance, as follows: age at earliest angiography; diabetes (type 1 or type 2); hypertension; depression and/or anxiety; and family history of premature CVD. In the second model, the 3 variables with the strongest effect sizes in the univariable analysis were added to the model—preeclampsia and/or gestational hypertension, smoking status, and migraines. Log likelihood changes, Akaike’s criterion, and the number of degrees of freedom were assessed to optimize the model fit.

This study demonstrated distinct differences in risk factor profiles of women with premature (M)INOCA—notably nontraditional CV risk factors such as migraines, preeclampsia or gestational hypertension, and depression and/or anxiety—compared to those with premature obstructive CAD, characterized mostly by traditional CV risk factors. These differences suggest different mechanisms and etiologies, underscoring a need for clinicians to collect and assess emerging nontraditional and sex-specific variables as risk factors that put patients at higher relative risk of (M)INOCA. The results of this study support the hypothesis of possible shared vasomotor disorder mechanisms among (M)INOCA, migraines, and preeclampsia or gestational hypertension. This concept of a systemic vasomotor disorder is not new; previous studies have identified strong associations with migraines and Raynaud’s syndrome in patients with vasospastic angina. 29 However, very few molecular (ie, genetic) studies of vasomotor (M)INOCA etiologies are available that could connect shared molecular mechanisms of these disorders. Multiple routes toward dysfunctional modulation of vascular tone by the endothelium have been proposed—enhanced coronary vasoconstrictive reactivity at the microvascular level, impaired endothelium-independent and -dependent vasodilatory capacity, and increased microvascular resistance due to structural factors (eg, vascular remodelling, luminal narrowing). 30 Furthermore, a sex-specific effect of high estrogen levels (relative to postmenopausal levels) may occur on subtypes of these pathologies, as estrogen promotes nitric oxide (NO) production via activation of NO synthase and subsequent endothelium-dependent vasodilation. 31 More research is required to understand the differences and shared mechanisms among cardiac vasomotor subtypes with noncardiac vasomotor etiologies (ie, migraines), and the role of sex hormones. A noteworthy finding was the emergence of depression and/or anxiety as an independent variable in the adjusted multivariable model, as it was not significant in the univariable models for (M)INOCA. Depression and anxiety are highly comorbid conditions with each other, and they occur in women with IHD more frequently than in men with IHD. 32 A persistent depressive state has been associated with abnormal vascular reactivity, predominantly through autonomic dysfunction and cumulative effects over time from chronic low-grade inflammation. 33 Depression and anxiety can have acute and chronic effects on the dysregulation of vasovagal tone through persistent activation of the hypothalamic-pituitary-adrenal axis, which exhibits sex differences in vasoconstrictive responsiveness. 34 Thus, prolonged activation may further exacerbate risk of (M)INOCA in women with a higher predisposition toward vasomotor syndromes. Altogether, these findings support a complex and bidirectional mechanism of depression and/or anxiety contributing to (M)INOCA risk. Pregnancy has lifelong impacts on women’s physiology, with numerous studies demonstrating pregnancy’s long-term benefits on immune 35 and CV health. 36 A proposed explanation for this link is improved endothelial function resulting in greater bioavailable NO during pregnancy that persists postpartum. 37 , 38 In the current study, having 2-3 children was associated with increased odds of (M)INOCA, which may also be interpreted as having 2-3 children being associated with a lower risk of CAD. This relationship has been reported previously as a J-shaped curve, with having 2-3 children being associated with fewer CV events than having 0-1 or 4+ children. 39 Because our findings relate to (M)INOCA relative to CAD, this question needs to be examined in future studies comparing women with CAD or (M)INOCA to women with no IHD. The prevalences of traditional CV risk factors, dyslipidemia, diabetes, and smoking status in our (M)INOCA cohort are similar to those in the general Canadian population of 45%, 11%, and 10%, respectively. 40 , 41 , 42 This similarity differs from migraines, which were observed to occur over 4 times as frequently in the (M)INOCA cohort (36%), compared to the Canadian population (8.3%). 43 Focusing on traditional CV risk factors allows for reasonable prediction of possible premature obstructive CAD but not premature (M)INOCA, as most (M)INOCA patients did not have these traditional risk factors. Limitations of this study include its small sample size (particularly of MINOCA patients), the lack of racial and/or ethnic diversity, and the omission of other sex-specific variables emerging as CV risk factors, owing to the absence of systematic collection in either or both cohorts, such as premature menopause (at age < 45 years), endometriosis, and polycystic ovarian syndrome. The extent to which the results from this study would apply to non-Caucasian populations is unclear; for example, several studies have shown significantly higher prevalence of vasospasm by provocative testing in people of Japanese, 44 Taiwanese, 45 and South Korean 46 descent, as compared to Caucasians. 22 However, the risk factor profiles of the young women within these populations are not clear.

Our study on risk factors associated with vasomotor etiologies of premature (M)INOCA compared to those associated with premature obstructive CAD provides further evidence that these entities demonstrate distinct risk factor profiles in young women. Specifically, nontraditional risk factors, such as migraines, preeclampsia or gestational hypertension, and depression and/or anxiety increase the risk of (M)INOCA and should be documented reliably in all patients presenting with possible IHD. Future studies with larger sample sizes, both sexes, and systematic collection of nontraditional risk factors are needed to improve our understanding of the mechanisms of vasomotor pathology in these premature-onset populations.