Abstract
Aims
Although the prevalence of cardiometabolic disease is greater in men vs. women, the relative risk of cardiovascular disease (CVD) conferred by cardiometabolic conditions is higher in women than in men. We examined the sex-specific association of cardiometabolic risk factor burden with subclinical echocardiographic cardiac remodelling.
Methods and results
In this cross-sectional study, we examined whether sex modifies the association between cardiometabolic disease burden (measured as metabolic syndrome severity [MetSS] score) with echocardiographic markers of subclinical cardiac remodelling (including global longitudinal strain [GLS] and E/e’ ratio) using multiplicative interaction terms in multivariable-adjusted linear regression models. Among 6182 Framingham Heart Study participants (mean age 51 ± 15 years; 54% women), we observed that sex modifies the association between the MetSS score and subclinical markers of left ventricular systolic and diastolic function. Specifically, a higher MetSS score was associated with worse GLS and E/e’ ratio in women vs. men. For example, every 1-point increase in the MetSS score was associated with 0.35% higher (worse) GLS in women compared with 0.23% higher GLS in men (ß 0.35, standard error [SE] 0.04 in women vs. ß 0.23, SE 0.04 in men, Pint = 0.01). Similarly, the MetSS score was more strongly associated with a higher (worse) E/e’ ratio in women vs. men (ß 0.29, SE, 0.02 in women vs. ß, 0.20, SE, 0.02 in men, Pint < 0.001).
Conclusion
We observed that sex modifies the association of cardiometabolic disease burden with subclinical markers of cardiac dysfunction. Specifically, higher cardiometabolic disease burden was associated with worse GLS and E/e’ ratio in women vs. men.
Lay summary
This study explores whether sex modifies the relationship between cardiometabolic risk and markers of cardiac dysfunction. The relationship between cardiometabolic risk factors and global longitudinal strain and E/e’ ratio varies by sex. Women have a steeper decline in markers of cardiac function than men for any given burden of cardiometabolic risk factors.
Keywords: Cardiometabolic, Risk, Sex, Remodelling, Women, Strain, Echocardiography
Graphical Abstract
Introduction
Over the past several decades, the growing burden of obesity and related metabolic disease has shifted the underlying substrate and profile of cardiovascular disease (CVD).1,2 CVD, once driven predominantly by tobacco use and hyperlipidaemia, is now increasingly a metabolic disease. This shift has been accompanied by an increasing proportion of women impacted by CVD, highlighting female susceptibility to metabolic-related CVD.3 Epidemiologic studies have consistently shown that the absolute prevalence of cardiometabolic risk traits, including obesity, type 2 diabetes (T2D), and kidney disease, is greater in men vs. women, and yet the relative risk of related CVD events is higher in women than in men.4-7 For example, the relative risk for coronary heart disease in the setting of diabetes is higher in women compared to men.6 The mechanisms underlying this phenomenon are incompletely elucidated.
Sex differences in age-related subclinical cardiac remodelling have been well described and may explain why women are more vulnerable to the development of metabolic-related CVD despite lower overall prevalence of cardiometabolic risk factors.8-10 For example, women experience more pronounced concentric left ventricular (LV) hypertrophy and accentuated diastolic dysfunction with advancing age compared with men.8,9 Women also experience a steeper increase in LV mass with advancing age compared with men.10 These sex-specific patterns in cardiac remodelling are particularly notable in the presence of cardiometabolic disease.8,10,11 For example, in longitudinal community-based studies, women displayed greater increments in LV mass with increasing body mass index (BMI), highlighting the unique interaction between cardiac remodelling and cardiometabolic traits in women.10
Subclinical cardiac remodelling may provide additional insights into cardiac structural changes that occur early in response to cardiometabolic disease. For example, global longitudinal strain (GLS) and E/e’ are independent predictors of long-term cardiovascular risk in a variety of clinical settings, with known sex differences in GLS and E/e’.12,13 Consistent evidence demonstrates that women have lower (better) baseline GLS compared with men and that male sex predicts an accelerated decline in GLS over time.14-16 By contrast, women have higher (worse) E/e’ ratio across all age categories,17 and experience more pronounced increases in E/e’ with increasing age compared with men.17 In this context, we investigated the sex-specific association of cardiometabolic risk traits with subclinical echocardiographic measures of cardiac remodelling in the Framingham Heart Study (FHS) participants free of overt CVD (Graphical Abstract). We hypothesized that sex would modify the association of cardiometabolic risk factors and subclinical markers of cardiac dysfunction, and that women experience more adverse cardiac remodelling in response to cardiometabolic disease.
Methods
Data sharing
Framingham Heart Study data, including clinical and echocardiographic data are available through application to dbGAP with accession number phs000007.v32.p13 or can be obtained via formal application to the FHS Research and Publications committee.
Study population
The FHS is a community-based observational cohort study of individuals free of CVD.18 Offspring cohort participants who attended Exam 8 (2005–2008) (n = 3021), Generation 3 cohort participants who attended Exam 1 (2002–2005) (n = 4095), and Omni-1 participants who attended Exam 3 (2007–2008) (n = 298) with available echocardiographic measures of myocardial strain and diastolic function were included in this study. From this sample, we excluded participants with prevalent heart failure (n = 90), prevalent myocardial infarction (n = 152), and missing covariates (n = 990) including GLS, BMI, smoking status, hypertension, hypertension medication, diabetes, waist circumference, dyslipidaemia, Homeostatic Model Assessment of Insulin Resistance (HOMA-IR), metabolic syndrome severity (MetSS) score, age, sex, heart rate and systolic blood pressure (BP), yielding a final sample size of n = 6182 (see Supplementary material online, Figure S1). In the primary analyses, we used a complete-case approach. For the HOMA-IR analyses, individuals with prevalent diabetes were excluded because HOMA-IR cannot be interpreted in the setting of established diabetes, yielding a sample size of n = 5730 for those analyses. All participants provided informed consent, and study protocols were approved by the Boston University Medical Campus Institutional Review Board.
Clinical assessment
All FHS participants underwent detailed medical histories, physical examinations, including vital signs and anthropometric measurements, and fasting blood collection for laboratory studies at each study visit. Covariates of interest included age, sex, race, ethnicity, heart rate, BMI, waist circumference, hypertension, treatment for hypertension, systolic BP, dyslipidaemia, treatment for dyslipidaemia, diabetes, HOMA-IR, and smoking status. BMI was calculated as weight in kilograms divided by the square of height in metres. Waist circumference was measured at the umbilicus in inches. Heart rate was determined from resting 12-lead ECG at a paper speed of 25 mm/s and measured in beats per minute. BP was measured using a mercury column sphygmomanometer and a cuff of appropriate size and reported as the average of two separate measurements obtained on seated participants. Hypertension was defined as systolic BP ≥ 130 mmHg, diastolic BP ≥ 80 mmHg, or reported use of anti-hypertensive medications. Diabetes was defined as a fasting glucose ≥ 126 mg/dL, a random blood glucose ≥ 200 mg/dL, or the reported use of glucose-lowering therapy. Dyslipidaemia was defined as an LDL-C ≥ 130 mg/dL, triglycerides ≥ 150 mg/dL, HDL-C < 40 mg/dL in men or <50 mg/dL in women, non-HDL-C ≥ 160 mg/dL, or current lipid-lowering therapy. Insulin resistance was assessed using HOMA-IR,19 with HOMA-IR > 2 suggestive of insulin resistance.
Cardiometabolic disease burden
The burden of cardiometabolic disease was assessed using the MetSS score, a quantitative assessment of the overall severity of MetS. Components of the MetSS score included: (1) waist circumference ≥ 102 cm in men or ≥ 88 cm in women; (2) triglycerides ≥ 150 mg/dL or dyslipidaemia treatment use; (3) HDL < 40 mg/dL in men or < 50 mg/dL in women; (4) systolic BP ≥ 130 mmHg, diastolic BP ≥ 85 mmHg, or anti-hypertensive therapy use; and (5) fasting blood glucose ≥ 100 mg/dL or glucose-lowering therapy use. Each metabolic risk factor was assigned 1 point, and all points were added up to calculate the individual’s MetSS score. MetSS score ranges from 0 to 5, with a score of ≥3 meeting the criteria for MetS.20,21
Echocardiographic measurements
Participants underwent standardized two-dimensional transthoracic echocardiography using an HP Sonos 550 cardiac ultrasound machine (Phillips Medical Systems; Andover, MA). Echocardiographic data were saved in Digital Imaging and Communications in Medicine format.22 Echocardiograms were interpreted by readers blinded to clinical data as previously described.23
The two pre-specified primary echocardiographic measures of interest were GLS (marker of subclinical systolic dysfunction) and E/e’ (a marker of subclinical diastolic dysfunction). Abnormal GLS was defined as >-18% and abnormal E/e’ ratio was defined as >14.24,25 Secondary systolic echocardiographic measures of interest include LV ejection fraction, global circumferential strain, and longitudinal segmental synchrony. Secondary echocardiographic measures of interest include LV mass, left atrial internal dimension, mitral E wave, mitral A wave, mitral E/A ratio, and mitral e’.
Speckle-tracking-based strain analyses of the LV myocardium were performed on three two-dimensional videos per participant (two-dimensional Cardiac Performance Analysis, version 1.1; Tomtec Imaging Systems; Unterschleissheim, Germany).26 Apical four-chamber and two-chamber views were used to assess GLS, while mid-ventricular parasternal short-axis views were used to assess global circumferential strain. Strain measurements demonstrated excellent inter-observer and intra-observer correlation (inter-observer correlation coefficients ranging from 3.0% to 9.7%, intra-observer coefficients of variation ranging from 3.5% to 10.0%).14
The early systolic mitral annulus velocity, mitral e’, was measured using tissue Doppler imaging at the centre of the lateral mitral annulus. Doppler trans-mitral flow velocities, the mitral E and A waves, were obtained using pulsed wave Doppler in the four-chamber view at the level of the mitral leaflet tips and were measured 3–5 times for each participant.27
Left ventricular ejection fraction was measured using the Teichholz method. Left ventricular synchrony was measured using the standard deviation of time-to-peak strain. Left ventricular mass was ascertained using the Devereux formula. Left ventricular dimensions were measured using the leading-edge to leading-edge technique at end-systole.
Statistical analysis
Baseline clinical and echocardiographic characteristics in men and women were compared using t-tests or chi-squared tests. BMI and systolic BP were all standardized to a mean of 0 and a standard deviation (SD) of 1. Waist circumference was standardized by sex to a mean of 0 and an SD of 1 for each sex. Given its skewed distribution, HOMA-IR was first natural log-transformed and then standardized. In our primary analyses, we evaluated whether sex modified the association of MetSS Score with GLS and mitral E/e’ using a sex × MetSS score interaction term in a sex-pooled sample. We performed linear regression analyses first in a minimally adjusted model accounting for age, sex, and heart rate, and subsequently in a multivariable model further adjusted for current smoking. If sex was observed to be an effect modifier, we performed multivariable-adjusted sex-stratified analyses. We performed a sensitivity analysis excluding prevalent atrial fibrillation. Finally, in exploratory analyses, to examine whether menopause timing on the association between cardiometabolic disease burden and subclinical cardiac dysfunction, we assessed the association of MetSS score with GLS and E/e’ in women with early menopause (<45 years) vs. menopause without early menopause (≥45 years old) using early menopause × MetSS score interaction terms.
In secondary analyses, we evaluated whether sex modified the association of MetSS score with other echocardiographic measures, including subclinical markers of LV systolic dysfunction (LV ejection fraction, longitudinal segmental synchrony, and global circumferential strain) and subclinical markers of LV diastolic dysfunction (LV mass, left atrial systolic diameter, mitral E, mitral A, Mitral E/A, mitral e’) using sex × MetSS score interaction terms.
In exploratory analyses, we further evaluated whether sex modified the association of individual cardiometabolic risk factors, including BMI, waist circumference, hypertension, systolic BP, dyslipidaemia, and HOMA-IR, with subclinical markers of LV systolic and diastolic function (GLS and E/e’ ratio, respectively) using sex × cardiometabolic risk factor interaction terms in a sex-pooled model. For each cardiometabolic risk factor, we adjusted for the appropriate set of confounding variables. If sex was found to be an effect modifier, sex-stratified analyses were performed.
All statistical analyses were performed using R version 4.4.3.28 For primary analyses, a Bonferroni corrected P-value of 0.05/2 primary outcomes = 0.025 was considered significant. For secondary and exploratory analyses, a P-value of <0.05 was suggestive.
Results
Participants
A total of 6182 participants were included in this study, including 3361 (54%) women. Baseline characteristics are displayed in Table 1. Mean age was similar in women and men (women 52 ± 16 years vs. men 51 ± 15 years). Compared with men, women had a lower burden of cardiometabolic disease, including obesity (25% in women vs. 29% in men), hypertension (45% in women vs. 62% in men), dyslipidaemia (55% in women vs. 70% in men), and diabetes (6% in women vs. 9% in men). Fewer women met criteria for MetS with a MetSS score of ≥3 (34% in women vs. 47% in men) compared with men. Baseline echocardiographic measures also differed in women vs. men, with higher LV ejection fraction (66 ± 5% in women vs. 64 ± 5% in men), more negative (better) GLS (−21.1 ± 3.1% in women vs. −19.0 ± 2.9% in men), and higher (worse) E/e’ (6.7 ± 2.1 in women vs. 6.1 ± 1.7 in men) in women vs. men.
Table 1.
Baseline characteristics in men and women
| Baseline clinical measures | Women (n = 3361) | Men (n = 2821) | P value |
|---|---|---|---|
| Age, years | 52 ± 16 | 51 ± 15 | 0.02 |
| Race and ethnicity | 0.02 | ||
| White | 3163 (94) | 2684 (95) | |
| Black | 65 (2) | 33 (1) | |
| Hispanic | 59 (2) | 32 (1) | |
| Other | 73 (2) | 66 (2) | |
| Heart rate, beats per minute | 64 ± 10 | 61 ± 10 | <0.001 |
| BMI, kg/m2 | 27.0 ± 6.0 | 28.4 ± 4.6 | <0.001 |
| Waist circumference, inch | 36.9 ± 6.4 | 39.9 ± 4.9 | <0.001 |
| Obesity, BMI ≥ 30 kg/m2, n (%) | 853 (25) | 829 (29) | <0.001 |
| Hypertension, n (%) | 1522 (45) | 1746 (62) | <0.001 |
| Hypertension medication, n (%) | 813 (24) | 751 (27) | 0.03 |
| Systolic blood pressure, mmHg | 120 ± 17 | 124 ± 15 | <0.001 |
| Dyslipidaemia, n (%) | 1831 (54) | 1973 (70) | 0.001 |
| Dyslipidaemia medication, n (%) | 617 (18) | 707 (25) | <0.001 |
| Low-density lipoprotein (LDL), mg/dL | 108 ± 31 | 113 ± 32 | <0.001 |
| Diabetes mellitus, n (%) | 194 (6) | 258 (9) | <0.001 |
| Glucose-lowering medication, n (%) | 140 (4.2) | 154 (5.6) | 0.02 |
| HOMA-IR, mg*uIU/dL*mL | 2.0 ± 2.1 | 2.4 ± 2.6 | <0.001 |
| Atrial fibrillation, n (%) | 57 (1.7) | 83 (2.9) | 0.001 |
| Current smoking, n (%) | 407 (12) | 347 (12) | 0.85 |
| Metabolic syndrome, n (%) | 1134 (34) | 1337 (47) | <0.001 |
| MetS severity score, median [IQR]; n (%) | 2 [0,3] | 2 [1,4] | <0.001 |
| 0 | 852 (25) | 442 (16) | |
| 1 | 760 (23) | 512 (18) | |
| 2 | 615 (18) | 530 (19) | |
| 3 | 438 (13) | 516 (18) | |
| 4 | 355 (11) | 475 (17) | |
| 5 | 341 (10) | 346 (12) | |
| Echocardiographic Measures | 0.02 | ||
| Left ventricular ejection fraction, % | 66 ± 5 | 64 ± 5 | <0.001 |
| Left ventricular mass, g | 127 ± 25 | 175 ± 33 | <0.001 |
| Global longitudinal strain, % | −21.1 ± 3.1 | −19.0 ± 2.9 | <0.001 |
| Global circumferential strain, % | −30.6 ± 5.4 | −28.8 ± 5.0 | <0.001 |
| Longitudinal segmental synchrony, ms | 88.9 ± 36.8 | 93.3 ± 38.6 | <0.001 |
| Left atrial internal dimension, cm | 3.61 ± 0.47 | 4.04 ± 0.50 | <0.001 |
| Mitral pulse wave E, cm/s | 69 ± 13 | 65 ± 13 | <0.001 |
| Mitral pulse wave A, cm/s | 62 ± 17 | 57 ± 15 | <0.001 |
| Mitral pulse wave e’, cm/s | 11 ± 3 | 11 ± 3 | 0.61 |
| Mitral pulse wave E/e’ | 6.7 ± 2.1 | 6.1 ± 1.7 | <0.001 |
| Mitral E/A | 1.2 ± 0.4 | 1.2 ± 0.4 | 0.86 |
Values are mean ± SD or n (%). Abbreviations: HOMA-IR = Homeostatic Model Assessment for Insulin Resistance, IQR = interquartile range, MetS = metabolic syndrome.
Global circumferential strain was available for n = 5750 participants. Longitudinal segmental synchrony was available for n = 6182 participants. Left ventricular ejection fraction was available for n = 5782 participants. Left atrial internal dimension (systole) was available for n = 6137 participants. Mitral pulsed wave A and E were available for n = 6095 participants. Doppler tissue velocity was available for n = 6090 participants. Left ventricular mass was available for n = 5518 participants. E/e’ was available for n = 6046 participants.
Sex modifies the association of cardiometabolic disease burden with subclinical markers of systolic and diastolic function
In primary analyses, we observed that sex modifies the association of MetSS score with subclinical echocardiographic markers of LV systolic and diastolic function (Figure 1, Tables 2 and 3). Specifically, the association of cardiometabolic burden (as measured by MetSS score) with greater (worse) GLS was more pronounced in women than in men. For example, every 1-point increase in MetSS score was associated with 0.35% higher (worse) GLS in women compared with 0.23% greater GLS in men (ß 0.35, SE 0.04, P-value <0.001 in women vs. ß 0.23, SE 0.04, P-value <0.001 in men, Pint = 0.01, Tables 2 and 3 and Supplementary material online, Table S1). Similarly, women with the greatest burden of cardiometabolic disease (MetSS score 5) had 1.66% higher (worse) GLS compared with women with a MetSS score of 0, while men with a MetSS score of 5 had 1.18% greater GLS vs. men with the lowest MetSS score of 0 (Figure 1). By contrast, sex did not modify the association of MetSS score with secondary markers of subclinical systolic function, including global circumferential strain, longitudinal segmental synchrony, and LV ejection fraction (see Supplementary material online, Tables S2-S5).
Figure 1.
The association of metabolic syndrome severity score with global longitudinal strain and mitral E/e. This figure displays the association of metabolic syndrome severity (MetSS) score with global longitudinal strain (Panel A) and E/e’ ratio (Panel B) in men (blue) vs. women (red). Beta-coefficients displayed represent an X-unit higher global longitudinal strain for each MetSS score compared with the MetSS score of 0 (referent group). Cardiometabolic risk factor burden was assessed using the Metabolic Syndrome Severity (MetSS) Score, which quantifies the presence of cardiometabolic risk factors including waist circumference, triglycerides, HDL-C, blood pressure, and fasting glucose (range 0–5). The arrow indicates worse GLS and E/e’.
Table 2.
Association of metabolic syndrome severity score categories with global longitudinal strain (GLS) and E/e’ in men vs. women
| MetSS score | Women (n = 3361) |
Men (n = 2821) |
||||||
|---|---|---|---|---|---|---|---|---|
| n (%) | β | SE | P | n (%) | β | SE | P | |
|
| ||||||||
| GLS | ||||||||
| 0 | 852 (25) | Reference | 442 (16) | Reference | ||||
| 1 | 760 (23) | 0.54 | 0.15 | <0.001 | 512 (18) | 0.29 | 0.18 | 0.1 |
| 2 | 615 (18) | 1.01 | 0.16 | <0.001 | 530 (19) | 0.68 | 0.18 | <0.001 |
| 3 | 438 (13) | 1.18 | 0.18 | <0.001 | 516 (18) | 0.97 | 0.19 | <0.001 |
| 4 | 355 (11) | 1.7 | 0.2 | <0.001 | 475 (17) | 0.95 | 0.2 | <0.001 |
| 5 | 341 (10) | 1.66 | 0.21 | <0.001 | 346 (12) | 1.18 | 0.22 | <0.001 |
|
| ||||||||
| Women (n = 3285) | Men (n = 2761) | |||||||
|
| ||||||||
| E/e’ | ||||||||
| 0 | 836 (25) | Reference | 435 (16) | Reference | ||||
| 1 | 743 (23) | 0.39 | 0.09 | <0.001 | 503 (18) | 0.21 | 0.1 | 0.04 |
| 2 | 603 (18) | 0.47 | 0.1 | <0.001 | 519 (19) | 0.35 | 0.1 | <0.001 |
| 3 | 432 (13) | 0.84 | 0.11 | <0.001 | 510 (19) | 0.62 | 0.11 | <0.001 |
| 4 | 344 (11) | 1.09 | 0.13 | <0.001 | 466 (17) | 0.87 | 0.11 | <0.001 |
| 5 | 327 (10) | 1.55 | 0.13 | <0.001 | 328 (12) | 0.96 | 0.13 | <0.001 |
The β coefficient represents X-units higher GLS and E/e’ ratio for the MetSS score compared with the MetSS score of 0 (referent group). Higher (more positive) GLS and E/e’ are worse. P examines the association of MetSS with GLS and E/e’ at each MetSS level compared with the MetSS score of 0 (referent group). The multivariable model adjusts for age, sex, heart rate, and current smoking.
Abbreviations: MetSS = metabolic syndrome severity; B = beta-coefficient; SE = standard error.
Table 3.
Association of per 1-point increase in metabolic syndrome severity score with global longitudinal strain (GLS) and E/e’ in men vs. women
| Outcome | Women (n = 3361) |
Men (n = 2821) |
|||||
|---|---|---|---|---|---|---|---|
| β | SE | P | β | SE | P | pint | |
|
| |||||||
| GLS | 0.35 | 0.04 | <0.001 | 0.23 | 0.04 | <0.001 | 0.01 |
|
| |||||||
| Women (n = 3285) | Men (n = 2761) | ||||||
|
| |||||||
| E/e’ | 0.29 | 0.02 | <0.001 | 0.2 | 0.020 | <0.001 | <0.001 |
The β coefficients represent X-units higher GLS and E/e’ ratio per 1-unit increase in MetSS score. Higher (more positive) GLS and E/e’ are worse. P examines the association of MetSS with GLS and E/e’, with MetSS score represented as an ordinal variable. The multivariable model adjusts for age, sex, heart rate, current smoking, and the MetSS score × sex interaction term. Pint examines the MetSS score × sex interaction term with MetSS score represented as an ordinal variable.
Abbreviations: MetSS = metabolic syndrome severity; B = beta-coefficient; SE = standard error.
A higher MetSS score was associated with worse subclinical diastolic function as measured by the E/e’ ratio in women compared with men. Specifically, the association of MetSS score with higher (worse) E’/e was again more pronounced in women vs. men (ß 0.29, standard error [SE] 0.02, P-value <0.001 in women vs. ß 0.20, SE 0.02, P-value <0.001 in men, Pint <0.001, Tables 2 and 3 and Supplementary material online, Table S11). Women with the highest MetSS score had 1.55-unit greater E/e’ vs. the lowest MetSS score vs. 0.96-unit increase in men (Figure 1). In exploratory analyses, sex was also found to modify the association of MetSS score with secondary markers of diastolic function, including left atrial internal dimension, mitral A wave, and mitral e’, with more pronounced associations of MetSS with left atrial dimension, mitral A wave, and mitral e’ in women vs. men (Pint < 0.001 for all) (see Supplementary material online, Tables S5, S7, and S8). Sensitivity analyses excluding participants with atrial fibrillation yielded results consistent with the primary analysis (see Supplementary material online, Table S12). Exploratory analyses examining the impact of early menopause on the association between cardiometabolic disease burden and subclinical cardiac dysfunction showed no effect modification by menopause timing (see Supplementary material online, Table S13).
Sex modification of the association of individual cardiometabolic risk factors with subclinical markers of systolic and diastolic function
We further probed whether sex modifies the association of GLS and E/e’ with individual cardiometabolic risk factors, including BMI, waist circumference, hypertension, systolic BP, dyslipidaemia, diabetes, and HOMA-IR. We found that the associations of BMI and systolic BP with GLS were modified by sex. Specifically, systolic BP was more strongly associated with higher (worse) GLS in women as compared to men (ß 0.23, SE 0.06, P-value <0.001 in women vs. ß 0.09, SE 0.06, P-value 0.15 in men, Pint = 0.03). By contrast, BMI was more strongly associated with higher (worse) GLS in men vs. women (ß 0.29, SE 0.05, P-value <0.001 in women vs. ß: 0.63, SE 0.06, P-value <0.001 in men, Pint = 0.002). The association of waist circumference, dyslipidaemia, and diabetes with GLS was not modified by sex (Figure 2, Table 4, Supplementary material online, Table S1).
Figure 2. Association of individual cardiometabolic risk factors with global longitudinal strain in men vs. women.
Panel A displays the association of systolic blood pressure, HOMA-IR, BMI, and waist circumference with GLS in men (blue) vs. women (red). Beta coefficient represents an X-unit higher GLS per 1-SD increase in cardiometabolic risk factor. HOMA-IR was first log-transformed and then standardized. Panel B displays the association of dyslipidaemia, obesity, diabetes and hypertension with GLS in men (blue) vs. women (red). The arrow indicates worse GLS. Beta coefficient represents a 1-unit higher GLS per presence vs. absence of cardiometabolic risk factor. *HOMA-IR analyses excluded participants with diabetes.
Table 4.
Association of cardiometabolic risk factors with global longitudinal strain and E/e’ in men vs. women
| Cardiometabolic risk factor | Women (n = 3361) |
Men (n = 2821) |
Overall | ||||
|---|---|---|---|---|---|---|---|
| B | SE | P-value | β | SE | P-value | pint | |
|
| |||||||
| Global Longitudinal Strain | |||||||
| Body Mass Index (kg/m2) | 0.29 | 0.05 | <0.001 | 0.63 | 0.06 | <0.001 | 0.002 |
| Waist circumference (inch) | 0.57 | 0.13 | <0.001 | 0.33 | 0.13 | 0.01 | 0.83 |
| Hypertension | 0.68 | 0.12 | <0.001 | 0.38 | 0.11 | 0.001 | 0.054 |
| Systolic blood pressure (mmHg) | 0.23 | 0.06 | <0.001 | 0.09 | 0.06 | 0.15 | 0.03 |
| Dyslipidaemia | 0.40 | 0.11 | <0.001 | 0.17 | 0.11 | 0.13 | 0.18 |
| Diabetes | 0.68 | 0.23 | 0.003 | 0.09 | 0.19 | 0.65 | 0.050 |
| HOMA-IRa (mg*uIU/dL*mL) | 0.10 | 0.07 | 0.16 | 0.18 | 0.07 | 0.01 | 0.28 |
|
| |||||||
| Women (n = 3285) | Men (n = 2761) | Overall | |||||
|
| |||||||
| E/e’ Ratio | |||||||
| Body mass index (kg/m2) | 0.28 | 0.03 | <0.001 | 0.38 | 0.04 | <0.001 | 0.21 |
| Waist circumference (inch) | −0.25 | 0.08 | 0.001 | −0.14 | 0.07 | 0.06 | 0.27 |
| Hypertension | 0.46 | 0.07 | <0.001 | 0.32 | 0.06 | <0.001 | <0.001 |
| Systolic blood pressure (mmHg) | 0.38 | 0.04 | <0.001 | 0.29 | 0.03 | <0.001 | <0.001 |
| Dyslipidaemia | −0.03 | 0.07 | 0.67 | 0.08 | 0.06 | 0.19 | 0.84 |
| Diabetes | 0.66 | 0.14 | <0.001 | 0.50 | 0.11 | <0.001 | 0.047 |
| HOMA-IRa (mg*uIU/dL*mL) | −0.10 | 0.04 | 0.02 | −0.14 | 0.04 | <0.001 | 0.006 |
β coefficients represent a X-units higher global longitudinal strain and E/e’ per 1-SD increase in exposure (for continuous variables) or presence of exposure (for dichotomous variables). Multivariable model adjusts for age, sex, heart rate, cardiometabolic risk factor × sex ± BMI, systolic blood pressure, hypertension treatment, current smoking, and diabetes, exclusive of exposure variable.
Abbreviations: BMI = body mass index; HOMA-IR = Homeostatic Model Assessment for Insulin Resistance.
HOMA-IR analyses excluded participants with diabetes.
Sex also modifies the association of cardiometabolic risk factors, including hypertension, systolic BP, diabetes, and HOMA-IR, with E/e’ ratio (Figure 3, Table 4, Supplementary material online, Table S1). The association of hypertension, systolic BP, and diabetes with higher (worse) E/e’ was more prominent in women than in men (hypertension: ß 0.46, SE 0.07, P-value <0.001 in women vs. ß 0.32, SE 0.06, P-value <0.001 in men, Pint < 0.001; systolic BP: ß 0.38, SE 0.04, P-value <0.001 in women vs. ß 0.29, SE 0.03, P-value <0.001 in men, Pint<0.001; diabetes: ß 0.66, SE 0.14, P-value <0.001 in women, ß 0.50, SE 0.11, P-value <0.001 in men, Pint = 0.047). By contrast, HOMA-IR was associated with lower E/e’ in men than in women (ß -0.10, SE 0.04, P-value 0.02 in women, ß −0.14, SE 0.04, P-value <0.001 in men, Pint = 0.006). The association of other cardiometabolic traits, including BMI, waist circumference, and dyslipidaemia, with E/e’ was not modified by sex (Pint > 0.05 for all).
Figure 3. Association of individual cardiometabolic risk factors with E/e’ in men vs. women.
Panel A displays the association of systolic blood pressure, HOMA-IR, BMI, and waist circumference with E/e’ in men (blue) vs. women (red). Beta coefficient represents an X-unit higher E/e’ per 1-SD increase in cardiometabolic risk factor. HOMA-IR was first log-transformed and then standardized. Panel B displays the association of dyslipidaemia, obesity, diabetes and hypertension with E/e’ in men (blue) vs. women (red). The arrow indicates worse E/e’. Beta coefficient represents a X-unit higher E/e’ per presence or absence of cardiometabolic risk factor.
Discussion
We observed that in a community-based sample of 6182 individuals free of overt CVD, sex modifies the association of cardiometabolic disease burden with subclinical markers of LV systolic and diastolic dysfunction. Specifically, the association of burden of cardiometabolic traits, ascertained by the MetSS score, with worse GLS and E/e’ ratio, was more pronounced in women compared to men. For example, women with the highest MetSS score demonstrated 1.66% higher GLS and 1.55 units higher E/e’ ratio compared with the lowest MetSS score, while men with the highest MetSS score had 1.18% higher GLS and 0.96 units higher E/e’ ratio compared with the lowest MetSS score. Sex also modified the association of individual cardiometabolic risk traits with GLS and E/e’, with hypertension, systolic BP, and diabetes more strongly associated with higher (worse) GLS and E/e’ in women vs. men. Together, these findings demonstrate that adverse subclinical cardiac remodelling in response to cardiometabolic risk exposures is more pronounced among women compared with men, which may contribute to the observation that cardiometabolic traits confer a greater relative risk of CVD in women.
Epidemiologic evidence has consistently demonstrated that, despite lower absolute prevalence of cardiometabolic risk traits in women vs. men, the same metabolic traits confer relatively greater risk of CVD in women vs. men. Data from the Centers for Disease Control and Prevention between 2021 and 2023 demonstrates that the prevalence of cardiometabolic risk traits, including T2D and hypertension, is lower in women vs. men (T2D: 14% in women vs. 15% in men; hypertension: 45% in women vs. 51% in men).29,30 We also observed a lower overall prevalence of cardiometabolic disease in women vs. men, with 34% women meeting criteria for MetS compared with 47% of men. Despite consistent observations that the burden of cardiometabolic disease is lower in women vs. men, the associated relative risk of adverse CVD events is higher in women vs. men. For instance, women with T2D experience 3–4 times greater risk of cardiovascular death compared with men with T2D.31 Similarly, obesity confers a 64% greater risk of CVD in women vs. 46% greater risk in men.32,33 More recently, the American Heart Association introduced the cardiovascular-kidney-metabolic (CKM) syndrome framework to capture the pathophysiological interplay between cardiometabolic risk factors, outlining a 4-stage model of CKM progression from obesity to CVD. In a national sample of 33 868 US adults, women exhibited a lower prevalence of advanced CKM stages but higher cardiovascular mortality risk across all stages of CKM syndrome in women compared with men.7 Despite these consistent observations, why women experience excess CV risk despite lower burden of cardiometabolic disease remains a key unanswered question.
Sex-specific differences in adverse cardiac remodelling have been well described and may explain why women are at greater risk of metabolic dysfunction-related CVD despite an overall lower burden of cardiometabolic risk traits. For example, women experience a more accelerated increase in LV mass and wall thickness than men with advancing age.8-10,34 In a prior FHS longitudinal study of 4217 healthy men and women with serial echocardiographic observations over a 16-year period, women displayed greater increments in LV mass with advancing age and with increasing BMI compared with men.10 Women are also more likely to develop concentric LV hypertrophy in response to pressure overload.35 These sex differences in age-related cardiac remodelling are particularly pronounced in response to cardiometabolic risk exposures. For example, in a study of 2623 Framingham Heart Study subjects, T2D was associated with greater adverse cardiac remodelling, specifically greater increases in LV mass and wall thickness, in women vs. men.36 In another study of 622 individuals without CVD, diabetes and sugar intake were found to be more strongly associated with LV mass-to-volume ratio ascertained by cardiac magnetic resonance imaging in women as compared to men.37 We now extend these previous investigations and show that women also experience more pronounced adverse changes in both subclinical LV systolic and diastolic function in response to cardiometabolic risk burden. Furthermore, we observed that individual risk factors, including hypertension, systolic BP, and diabetes, were more strongly associated with higher (worse) GLS and E/e’ in women as compared to men. Our findings show that early cardiac remodelling in response to cardiometabolic disease is also more pronounced in women than in men.
Sex hormones and aldosterone have been proposed to underlie sex differences in cardiac remodelling and may contribute to our study findings. Oestrogen receptors are expressed in the myocardium, and the presence of oestrogen is thought to attenuate the development of LV hypertrophy via inhibition of myocyte apoptosis.36 Withdrawal of oestrogen following menopause may therefore explain the acceleration in LV hypertrophy with advancing age observed in women9,38 Similarly, aldosterone receptors are also present in cardiac myocytes and known to play a role in myocyte growth. In a study of 2820 healthy participants from the FHS, serum aldosterone levels were associated with LV remodelling, specifically relative wall thickness and LV wall thickness, in women but not men, which may explain why women are more likely to develop LV concentric hypertrophy in response to hypertension and other pressure overload states.39 Indeed, when examining individual cardiometabolic risk factors, we demonstrated that both hypertension and systolic blood pressure were more strongly associated with worse GLS and E/e’ in women compared with men. Moreover, therapies modulating aldosterone pathways including mineralocorticoid receptor antagonists and angiotensin receptor neprilysin inhibitors have shown preferential benefit in women vs. men with heart failure with preserved ejection fraction, one of the end-stage manifestations of adverse LV cardiac remodelling.40 Taken together, these studies highlight the important contribution of sex hormones and aldosterone in sex-specific cardiac remodelling and emphasize the need for further characterization of the precise mechanistic pathways that drive adverse cardiac remodelling in women with cardiometabolic disease.
Our study has several limitations worth noting. First, the overall magnitude of effect sizes was modest and likely reflects subclinical, population-level alterations in cardiac mechanics, rather than changes that should be interpreted as individual clinical decision thresholds; future studies will be needed to determine the clinical significance of our findings. Second, the E/e’ ratio is an estimation of LV filling pressure and is the only one measure used in the assessment of diastolic dysfunction. Third, reproductive history data were not routinely ascertained in the earlier cohorts of the FHS, precluding further investigation of the impact of menopause and sex hormones on subclinical cardiac remodelling. Fourth, serial echocardiograms were not available, precluding assessment of longitudinal changes in cardiac structure and function. Fifth, because the examination periods predated widespread use of newer cardiometabolic therapies, including SGLT2 inhibitors and GLP-1 receptor agonists, we could not evaluate the association of these agents with subclinical cardiac remodelling. Sixth, this was a retrospective, cross-sectional study, and causal inferences cannot be established. Finally, the FHS participants were largely of European ancestry and from the New England region. The generalizability of our findings to individuals from other ancestries or regions is uncertain.
In our community-based sample of individuals free of CVD, we demonstrate that sex modifies the association of cardiometabolic risk burden with subclinical cardiac dysfunction. The associations of cardiometabolic risk factor burden and individual cardiometabolic risk traits with greater GLS and E/e’ ratio, markers of subclinical systolic and diastolic dysfunction, were more pronounced in women. Our findings highlight that exposure to cardiometabolic risk traits may promote greater adverse subclinical cardiac remodelling in women vs. men and may explain why women are more susceptible to metabolic-related CVD.
Supplementary Material
Supplementary material is available at European Journal of Preventive Cardiology.
Funding
E.S.L. is supported by grants from the National Institutes of Health (K23-HL159243, R01HL180648), the Massachusetts Life Sciences Center, and the Tianqiao and Chrissy Chen Institute. The Framingham Heart Study Cois supported by NIH contracts (N01-HC-25195, HHSN268201500001I, 75N92019D00031, 75N92025D00012). E.J.B. is supported in part by the Jay and Louise Coffman Professorship in Vascular Medicine and National Institutes of Health, The National Heart, Lung, and Blood Institute R01HL092577 and National Institute of Arthritis and Musculoskeletal and Skin Diseases (K12AR085635, R01HL080124, R01HL70100, R01HL107385, R01HL142983).
Conflict of interest:
E.S.L. reports consulting fees for SystoleHealth, Inc and advisory board service with Amissa Health and Roon, unrelated to this work. G.F.M. is the owner of Cardiovascular Engineering, Inc., a company that designs and manufactures devices that measure vascular stiffness. The company uses these devices in clinical trials that evaluate the effects of diseases and interventions on vascular stiffness. He also serves as a consultant to and receives grants and honoraria from Novartis, Merck, Bayer, Servier, Philips, and deCODE genetics, and is an inventor on pending patent applications that disclose methods for predicting various measures of biological age using pressure waveforms. The other authors report no relevant conflicts of interest.
References
- 1.Powell-Wiley TM, Poirier P, Burke LE, Després J-P, Gordon-Larsen P, Lavie CJ, et al. Obesity and cardiovascular disease: a scientific statement from the American Heart Association. Circulation 2021;143:e984–e1010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Ndumele CE, Neeland IJ, Tuttle KR, Chow SL, Mathew RO, Khan SS, et al. A synopsis of the evidence for the science and clinical management of cardiovascular-kidney-metabolic (CKM) syndrome: a scientific statement from the American Heart Association. Circulation 2023;148:1636–1664. [DOI] [PubMed] [Google Scholar]
- 3.Roeters Van Lennep JE, Tokgözoğlu LS, Badimon L, Dumanski SM, Gulati M, Hess CN, et al. Women, lipids, and atherosclerotic cardiovascular disease: a call to action from the European Atherosclerosis Society. Eur Heart J 2023;44:4157–4173. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Rajendran A, Minhas AS, Kazzi B, Varma B, Choi E, Thakkar A, et al. Sex-specific differences in cardiovascular risk factors and implications for cardiovascular disease prevention in women. Atherosclerosis 2023;384:117269. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Walli-Attaei M, Joseph P, Rosengren A, Chow CK, Rangarajan S, Lear SA, et al. Variations between women and men in risk factors, treatments, cardiovascular disease incidence, and death in 27 high-income, middle-income, and low-income countries (PURE): a prospective cohort study. The Lancet 2020;396:97–109. [DOI] [PubMed] [Google Scholar]
- 6.Huxley R, Barzi F, Woodward M. Excess risk of fatal coronary heart disease associated with diabetes in men and women: meta-analysis of 37 prospective cohort studies. BMJ 2006;332:73–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Ji H, Sabanayagam C, Matsushita K, Cheng C-Y, Rim TH, Sheng B, et al. Sex differences in cardiovascular-kidney-metabolic syndrome: 30-year US trends and mortality risks—brief report. ATVB 2025;45:157–161. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Cheng S, Xanthakis V, Sullivan LM, Lieb W, Massaro J, Aragam J, et al. Correlates of echocardiographic indices of cardiac remodeling over the adult life course: longitudinal observations from the Framingham Heart Study. Circulation 2010;122:570–578. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Merz AA, Cheng S. Sex differences in cardiovascular ageing. Heart 2016;102:825–831. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Lieb W, Xanthakis V, Sullivan LM, Aragam J, Pencina MJ, Larson MG, et al. Longitudinal tracking of left ventricular mass over the adult life course: clinical correlates of short- and long-term change in the Framingham Offspring Study. Circulation 2009;119:3085–3092. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Ho JE, McCabe EL, Wang TJ, Larson MG, Levy D, Tsao C, et al. Cardiometabolic traits and systolic mechanics in the community. Circ: Heart Failure 2017;10:e003536. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Seko Y, Kato T, Shiba M, Morita Y, Yamaji Y, Haruna Y, et al. Association of the low e′ and high E/e′ with long-term outcomes in patients with normal ejection fraction: a hospital population-based observational cohort study. BMJ Open 2019;9:e032663. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Sharp ASP, Tapp RJ, Thom SAM, Thom SAM, Francis DP, Hughes AD, et al. Tissue Doppler E/e’ ratio is a powerful predictor of primary cardiac events in a hypertensive population: an ASCOT substudy. Eur Heart J 2010;31:747–752. [DOI] [PubMed] [Google Scholar]
- 14.Cheng S, Larson MG, McCabe EL, Osypiuk E, Lehman BT, Stanchev P, et al. Age- and sex-based reference limits and clinical correlates of myocardial strain and synchrony: the Framingham Heart Study. Circ: Cardiovascular Imaging 2013;6:692–699. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Lau ES, Zhao Y, Benjamin EJ, Ramachandran VS, Xanthakis V, Cheng S, et al. Sex differences in left ventricular function and cardiac mechanics. JAHA 2024;13:e035781. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Skaarup KG, Lassen MCH, Marott JL, Biering-Sørensen SR, Jørgensen PG, Appleyard M, et al. The impact of cardiovascular risk factors on global longitudinal strain over a decade in the general population: the Copenhagen City Heart Study. Int J Cardiovasc Imaging 2020;36:1907–1916. [DOI] [PubMed] [Google Scholar]
- 17.Okura H, Takada Y, Yamabe A, Kubo T, Asawa K, Ozaki T, et al. Age- and gender-specific changes in the left ventricular relaxation: a Doppler echocardiographic study in healthy individuals. Circ: Cardiovascular Imaging 2009;2:41–46. [DOI] [PubMed] [Google Scholar]
- 18.Kannel WB, Feinleib M, McNAMARA PM, Garrison RJ, Castelli WP. An investigation of coronary heart disease in families. Am J Epidemiol 1979;110:281–290. [DOI] [PubMed] [Google Scholar]
- 19.Dasgupta R, Shetty SP. Assessment of insulin resistance: from the bench to bedside. In: Metabolic Syndrome: From Mechanisms to Intervensions. Philadelphia: Elsevier; 2024. p351–365. [Google Scholar]
- 20.National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Third report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) final report. Circulation 2002;106:3143–3421. [PubMed] [Google Scholar]
- 21.Grundy SM, Cleeman JI, Daniels SR, Donato KA, Eckel RH, Franklin BA, et al. Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute Scientific statement. Circulation 2005;112:2735–2752. [DOI] [PubMed] [Google Scholar]
- 22.Phillips Medical Systems, Andover, MA, USA. [Google Scholar]
- 23.Nayor M, Enserro DM, Vasan RS, Xanthakis V. Cardiovascular health Status and incidence of heart failure in the Framingham Offspring Study. Circ: Heart Failure 2016;9:e002416. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Thomas JD, Edvardsen T, Abraham T, Appadurai V, Badano L, Banchs J, et al. Clinical applications of strain echocardiography: a clinical consensus statement from the American society of echocardiography developed in collaboration with the European association of cardiovascular imaging of the European Society of Cardiology. J Am Soc Echocardiogr 2025;38:985–1020. [DOI] [PubMed] [Google Scholar]
- 25.Nagueh SF, Sanborn DY, Oh JK, Anderson B, Billick K, Derumeaux G, et al. Recommendations for the evaluation of left ventricular diastolic function by echocardiography and for heart failure with preserved ejection fraction diagnosis: an update from the American Society of Echocardiography. J Am Soc Echocardiogr 2025;38:537–569. [DOI] [PubMed] [Google Scholar]
- 26.Tomtec Imaging Systems GmbH, Unterschleißheim, Germany. [Google Scholar]
- 27.Vasan RS, Larson MG, Levy D, Galderisi M, Wolf PA, Benjamin EJ. Doppler transmitral flow indexes and risk of atrial fibrillation (the framingham heart study). Am J Cardiol 2003;91:1079–1083. [DOI] [PubMed] [Google Scholar]
- 28.A Language and Environment for Statistical Computing. R Foundation for Statistical Computing. https://www.R-project.org/
- 29.National Diabetes Statistics Report Website. Centers for Disease Control and Prevention; 2025. https://www.cdc.gov/diabetes/php/data-research/index.html
- 30.Fryar CD, Kit B, Carroll MD, Afful J. Hypertension prevalence, awareness, treatment, and control among adults age 18 and older: United States, August 2021-August 2023. NCHS Data Brief 2024:CS354233. [PubMed] [Google Scholar]
- 31.Regensteiner JG, Golden S, Huebschmann AG, Barrett-Connor E, Chang AY, Chyun D, et al. Sex differences in the cardiovascular consequences of diabetes Mellitus: a scientific statement from the American Heart Association. Circulation 2015;132:2424–2447. [DOI] [PubMed] [Google Scholar]
- 32.Bushnell C, McCullough LD, Awad IA, Chireau MV, Fedder WN, Furie KL, et al. Guidelines for the prevention of stroke in women: a statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2014;45:1545–1588. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Brown HL, Warner JJ, Gianos E, Gulati M, Hill AJ, Hollier LM, et al. Promoting risk identification and reduction of cardiovascular disease in women through collaboration with obstetricians and gynecologists: a presidential advisory from the American Heart Association and the American College of Obstetricians and Gynecologists. Circulation 2018;137:e843–e852. [DOI] [PubMed] [Google Scholar]
- 34.Piro M, Della Bona R, Abbate A, Biasucci LM, Crea F. Sex-related differences in myocardial remodeling. J Am Coll Cardiol 2010;55:1057–1065. [DOI] [PubMed] [Google Scholar]
- 35.Krumholz HM, Larson M, Levy D. Sex differences in cardiac adaptation to isolated systolic hypertension. Am J Cardiol 1993;72:310–313. [DOI] [PubMed] [Google Scholar]
- 36.Rutter MK, Parise H, Benjamin EJ, Levy D, Larson MG, Meigs JB, et al. Impact of glucose intolerance and insulin resistance on cardiac structure and function: sex-related differences in the Framingham Heart Study. Circulation 2003;107:448–454. [DOI] [PubMed] [Google Scholar]
- 37.Verma MS, Lindsay K, Ni J, Friedrich MG, Luu JM. Sex-specific associations of cardiovascular risk factors with subclinical cardiac remodelling: a magnetic resonance imaging study. Can J Cardiol 2025;41:1836–1844. [DOI] [PubMed] [Google Scholar]
- 38.Mahmoodzadeh S, Eder S, Nordmeyer J, Ehler E, Huber O, Martus P, et al. Estrogen receptor alpha up-regulation and redistribution in human heart failure. FASEB j 2006;20:926–934. [DOI] [PubMed] [Google Scholar]
- 39.Vasan RS, Evans JC, Benjamin EJ, Levy D, Larson MG, Sundstrom J, et al. Relations of Serum aldosterone to cardiac structure: gender-related differences in the Framingham Heart Study. Hypertension 2004;43:957–962. [DOI] [PubMed] [Google Scholar]
- 40.Merrill M, Sweitzer NK, Lindenfeld J, Kao DP. Sex differences in outcomes and responses to spironolactone in heart failure with preserved ejection fraction. JACC Heart Fail 2019;7:228–238. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.