BACKGROUND:ECG-based artificial intelligence may enable efficient prediction of incident heart failure (HF) risk to facilitate preventive efforts. Prior models are proprietary, with modest or inconsistent accuracy. We sought to develop and validate a generalizable and publicly available convolutional neural network to predict incident HF using the 12-lead ECG waveform (ECG-to-HF [ECG2HF]). METHODS:We developed ECG2HF in 94 636 patients receiving longitudinal ambulatory care at Massachusetts General Hospital (MGH), and validated it in 3 test sets: MGH, Brigham and Women's Hospital (BWH), and Beth Israel Deaconess Medical Center (BIDMC), among 93 868 individuals aged 30 to 79 years without HF. HF events at 10 years were identified using a validated electronic health record-based natural language processing model. Discrimination was quantified using the area under the receiver operating characteristic curve. We then compared discrimination and net reclassification (at <10%, 10% to 20%, ≥20% 10-year risk categories) using ECG2HF versus the 15-component Pooled Cohorts Equations to Prevent HF score. RESULTS:The test sets comprised MGH (13 954 individuals, 441 events, age 57±13 years, 48% women), BWH (54 396 individuals, 1809 events, age 57±13 years, 55% women), and BIDMC (25 457 individuals, 901 events, age 57±13 years, 53% women). Over 10 years, the cumulative risk of HF was 4.6% (95% CI, 4.1-5.0) in MGH, 5.0% (4.8-5.2) in BWH, and 4.4% (4.1-4.7) in BIDMC. ECG2HF discriminated 10-year incident HF in each test set (area under the receiver operating characteristic curve: MGH 0.86 [0.84-0.87]; BWH 0.85 [0.84-0.86]; BIDMC 0.84 [0.83-0.86]). Compared with the Pooled Cohorts Equations to Prevent HF, ECG2HF provided favorable discrimination (improvement in area under the receiver operating characteristic curve MGH/BWH 0.061 [0.025-0.097]; BIDMC 0.038 [-0.0096 to 0.086]) and net reclassification (NRI MGH/BWH 0.16 [0.077-0.24]; BIDMC 0.23 [0.10-0.35]) of 10-year HF risk. CONCLUSIONS:ECG2HF is a publicly available 12-lead ECG-based artificial intelligence model that discriminates the risk of future HF with favorable and consistent performance across 3 large health care samples from the northeastern United States. ECG2HF may enable efficient prioritization of high-risk individuals for HF-related preventive measures.

BACKGROUND:Peak oxygen consumption (peak VO2), the gold standard measure of cardiorespiratory fitness, may identify women at high risk for pregnancy-related cardiovascular (CV) complications, but ascertainment is not widely scalable. We previously developed and validated a deep learning model to estimate peak VO2 from the resting 12-lead electrocardiogram (ECG). OBJECTIVES:The purpose of this study was to examine the association of deep learning ECG-predicted peak VO2 with incident pregnancy-related CV complications. METHODS:We evaluated ECG-estimated peak VO2 among individuals with a clinical 12-lead ECG from 1 year before pregnancy to 13 weeks of gestation in a multi-institutional electronic health record pregnancy cohort. Multivariable-adjusted mixed-effects logistic regression models examined associations between ECG-estimated peak VO2 with pregnancy-related CV complications up to 1 year postpartum (severe hypertensive disorders of pregnancy, major adverse cardiac events, and maternal death). RESULTS:Among 3,650 pregnancies from 3,437 women (mean age at delivery 33 ± 6 years), median ECG-estimated VO2 was 27.0 mL/kg/min (IQR: 21.9-31.2), and 723 (20%; 95% CI: 19%-21%) experienced a pregnancy-related CV complication over a median follow-up time of 1.7 years (IQR: 1.6-1.8 years). Lower ECG-estimated peak VO2 was associated with a higher complication risk (adjusted OR: 1.09 per 1-metabolic equivalent lower fitness; 95% CI: 1.03-1.17; P < 0.01). Women in the lowest quartile of ECG-estimated peak VO2 had 61% greater odds of CV complications than the highest quartile (OR: 1.61; 95% CI: 1.13-2.30; P = 0.008). CONCLUSIONS:Lower ECG-estimated cardiorespiratory fitness was associated with a higher risk of pregnancy-related CV complications, supporting artificial intelligence-enabled ECG analysis as a scalable tool for antepartum risk stratification of pregnancy-related CV complications.

Cardiovascular disease (CVD) remains the leading cause of death among women, yet CVD risk is frequently underrecognized during midlife, when the menopause transition is accompanied by significant biological changes. Emerging evidence demonstrates that menopause represents a critical window for adverse cardiometabolic and vascular changes that extend beyond the effects of chronologic aging alone. Because obstetrician-gynecologists often serve as the primary clinicians for midlife care in women, they are uniquely positioned to recognize emerging cardiovascular risk, implement preventive strategies, and coordinate care with primary care or cardiology when appropriate. This Clinical Expert Series synthesizes current evidence on menopause-related characteristics associated with CVD risk, including timing and type of menopause, as well as menopause-related symptoms, including vasomotor symptoms, sleep disturbances, and depression. We also review longitudinal data characterizing changes in lipids, blood pressure, glucose metabolism, body composition, and vascular structure and function across the menopause transition. Collectively, these data support framing menopause as a key opportunity for CVD risk identification and early intervention. Incorporating menopause-specific factors into routine CVD risk assessment may enable earlier prevention strategies and ultimately improve long-term cardiovascular outcomes for women.

BACKGROUND:Adverse pregnancy outcomes are a major driver of high maternal mortality in the United States. There are limited data on cardiovascular health (CVH) in reproductive-aged women nationally. OBJECTIVES:The objective of the study was to assess CVH in pregnant, postpartum, and nonpregnant women of reproductive age using the American Heart Association Life's Essential 8 (LE8) score. METHODS:We performed serial cross-sectional analysis of reproductive-aged women 20 to 44 years of age in the United States participating in the National Health and Nutrition Examination Survey from 2015 to 2023. Overall CVH scores (0-100) were calculated using the LE8 core components in nonpregnant women and women with any pregnancy-related condition (defined as currently pregnant, pregnant within 1 year of the survey, and/or breastfeeding). Population weighted analyses were used to compare CVH scores between groups. RESULTS:Among 3,992 participants (representing 46.9 million U.S. women), 10.7% were pregnant, breastfeeding, or postpartum. Overall, the mean LE8 score was 66 corresponding to moderate CVH, with similar scores among women with any pregnancy-related condition compared to nonpregnant women (65 vs 66, P = 0.20). Pregnant women (62) had the lowest LE8 scores compared to postpartum (67), breastfeeding (70), and nonpregnant women (66), respectively, P < 0.001. In addition, high-sensitive C-reactive protein levels were significantly elevated in pregnant women (4.68 mg/dL) compared to postpartum (2.39 mg/dL), breastfeeding (2.40 mg/dL), and nonpregnant women (2.06 mg/dL) (P = 0.006). CONCLUSIONS:Reproductive-aged women in the United States demonstrated moderate CVH, with pregnant women exhibiting lower LE8 scores compared to postpartum and nonpregnant women. These findings highlight modifiable targets for interventions to reduce adverse pregnancy outcomes and future cardiovascular risk.

Background Sex differences in clinical presentations of heart failure (HF) are pervasive, yet differences in invasive hemodynamic measures across HF subtypes remain incompletely characterized. Understanding these distinctions is critical, as sex‐specific physiology influences disease progression and prognostic assessment in HF. Methods We conducted a retrospective cohort study of 1818 patients with HF who underwent clinically indicated right heart catheterization, stratified by preserved versus reduced ejection fraction. We examined the association of sex with invasive hemodynamic measures of right ventricular, pulmonary vascular, and left ventricular function using multivariable linear regression. Results We studied n=709 individuals with heart failure with preserved ejection fraction (HFpEF) (44% women, mean age 68±10 years) and n=1109 with heart failure with reduced ejection fraction (HFrEF) (26% women, mean age 64±12 years). Women and men had similar measures of right ventricular function across HF subtypes (multivariable‐adjusted P >0.05 for all). However, women with both HFpEF and HFrEF exhibited worse pulmonary vascular function, with significantly higher pulmonary vascular resistance and lower pulmonary artery compliance (multivariable‐adjusted P <0.001). Women with both HFpEF and HFrEF had similar cardiac power index, but higher aortic pulsatility index ( P =0.01 for HFpEF; P <0.001 for HFrEF). Conclusions Among a large hospital‐based sample of patients with HF, women and men exhibited similar indices of right ventricular function, despite women having worse pulmonary vascular function. Cardiac power index was similar between sexes in both HFpEF and HFrEF, whereas aortic pulsatility index was higher in women with HFrEF. Taken together, sex differences in hemodynamic measures were largely similar across individuals with HFpEF and HFrEF. Future studies are needed to delineate clinical implications of these sex differences.

BACKGROUND:Accurate prediction of incident heart failure (HF) may help prioritize HF preventive therapies. Deep learning interpretation of echocardiograms may improve HF risk prediction beyond clinical risk models. We trained and validated a deep learning model to predict incident HF from transthoracic echocardiographic images (Echocardiogram-to-Heart Failure, or "Echo2HF") METHODS: Echo2HF was developed using 4,057,664 echocardiogram videos from 70,763 patients receiving longitudinal ambulatory care at Massachusetts General Hospital (MGH). Performance for 10-year incident HF was evaluated in an internal MGH test set and an external test set of 34,802 individuals without prevalent HF from Brigham and Women's Hospital (BWH). Model performance was evaluated using the area under the receiver operating characteristic curve and compared with the Pooled Cohorts Equations to Prevent Heart Failure and the Predicting Risk of cardiovascular disease EVENTs clinical risk scores. RESULTS:Echo2HF was trained in 64,167 individuals and evaluated in a hold-out sample of 6394 individuals from MGH (279 HF events, age 62 ± 17 years, 48% women) and 34,802 individuals from BWH (1280 events, age 62 ± 15 years, 56% women). Echo2HF discriminated incident HF, with 10-year area under the receiver operating characteristic curve of 0.84 (95% confidence interval 0.81-0.87) and 0.84 (95% confidence interval 0.82-0.85) at BWH, with numerically greater discrimination vs both Pooled Cohorts Equations to Prevent Heart Failure and Predicting Risk of cardiovascular disease EVENTs. CONCLUSION:Deep learning analysis of echocardiograms accurately discriminated future HF risk, with favorable performance over current clinical HF scores. Future work should assess whether broader use of artificial intelligence-enabled echocardiographic risk stratification may improve HF prevention and clinical outcomes, including among individuals who do not have a clinical indication for echocardiography.

Importance:Stress and its psychiatric consequences-including depression, anxiety, and posttraumatic stress disorder (PTSD)-are pertinent to women's cardiovascular health, but research on intersections with relevant sex-specific factors (eg, hormonal contraceptives) is lacking. Objective:To examine whether stress-related psychiatric diagnoses moderate associations between hormonal contraceptive use and cardiovascular and thrombotic risk. Design, Setting, and Participants:This retrospective cohort study included electronic health record data collected from a US hospital-based biobank and analyzed from May 2, 2024, to November 3, 2025. Participants were women aged 18 to 55 years who consented into the biobank before or on September 12, 2020. Exposures:Lifetime history of stress-related psychiatric disorders, including depression (major depressive disorder), anxiety (generalized anxiety disorder, social anxiety disorder, or panic disorder), and PTSD, defined by International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) codes and analyzed as separate diagnoses, and lifetime history of combined hormonal contraceptive use, defined by RxNorm codes. Main Outcomes and Measures:The primary outcomes were major adverse cardiovascular events (MACE; defined as ICD-10 codes for infarction, unstable angina, heart failure, coronary revascularization, peripheral vascular disease, peripheral revascularization, stroke, and/or transient ischemic attack) and deep-vein thrombosis (DVT). Three 2-step hierarchical logistic regressions per outcome were conducted. Results:In this sample of 31 824 women (mean [SD] age, 38.5 [10.6] years), over one-third (11 950 women [37.6%]) had hormonal contraceptive use history, and stress-related disorders were common (depression, 9116 women [28.5%]; anxiety, 3533 women [11.1%]; PTSD, 1992 women [6.3%]). Associations were mixed across the stress-related disorders, in that depression and anxiety did not moderate associations between contraceptive use and MACE or DVT. In contrast, PTSD modified the association between contraceptive use and MACE but not that between contraceptive use and DVT. Analyses stratified by PTSD status found that only women without PTSD using contraceptives had lower odds for MACE (odds ratio, 0.69; 95% CI, 0.87-3.24). The odds ratio for MACE among women with PTSD was greater than 1, but the finding was not statistically significant (odds ratio, 1.68; 95% CI, 0.87-3.24). Conclusions and Relevance:In this retrospective cohort study, combined hormonal contraceptive use was associated with lower cardiovascular risk in women regardless of depression or anxiety. These protective associations did not extend to women with PTSD, suggesting that there are unique cardiovascular processes in the context of this stress-related disorder and hormonal contraceptive use that warrant further research.

AIMS:The molecular pathways by which obesity contributes to systemic inflammation are unclear. Eicosanoids are bioactive lipids that govern the upstream initiation of pro- and anti-inflammatory activity. We sought to investigate the association of eicosanoids with obesity, adiposity, and cardiometabolic traits. METHODS:We conducted a cross-sectional analysis of the Multi-Ethnic Study of Atherosclerosis (MESA) study with external validation in two community-based cohorts: Framingham Heart Study and Atherosclerosis Risk in Communities Study. We measured 811 eicosanoids using a directed, non-targeted mass spectrometry-based platform and examined their associations with BMI and eight related adiposity and cardiometabolic traits using multivariable-adjusted linear regression. RESULTS:Among 5101 MESA participants (mean age 63 years, 53% women), we found 255 eicosanoids and related metabolites associated with BMI (FDR q < 0.01 for all). Among these, 82 showed consistent associations across all three cohorts, including 18 metabolites with known identities: 9 were associated with lower BMI including docosahexaenoic acid derivative 19,20-DiHDPA, arachidonic acid derivatives (20-carboxy-arachidonic acid, 5,6-diHETrE, 14,15-diHETrE, 15-HpETE), and eicosapentaenoic acid derivative 5-HpEPE. By contrast, 9 metabolites associated with higher BMI including hydroxyoctadecenoic acid (HOME), 11-hydroxy-9-octadecenoic acid (11-HOME), and adrenic acid. Many BMI-associated metabolites overlapped with other adipose traits including 240 (94%) with waist circumference and 93 (36%) with visceral adipose tissue. CONCLUSION:We identified 82 BMI-associated eicosanoids, including docosanoids and arachidonic acid derivatives associated with lower BMI, while octadecanoids and adrenic acid derivatives were associated with higher BMI. Our findings highlight the role of bioactive lipids and specific pro- and anti-inflammatory pathways that may underlie adverse cardiometabolic consequences of obesity.

BACKGROUND:Cardiac and vascular complications are the leading causes of maternal mortality and morbidity, but the contemporary burden of and secular trends in pregnancy-related cardiovascular complications are not well-characterized. We developed a multi-institutional electronic health record-based pregnancy cohort with rigorously defined cardiovascular outcomes to examine trends in prevalence of maternal cardiovascular comorbidities and cardiovascular disease (CVD) and incidence of pregnancy-related cardiovascular complications. METHODS:We identified pregnancy encounters that occurred between 2001 and 2019 from a primary care electronic health record cohort using International Classification of Diseases and Current Procedural Terminology codes. We used regular expressions to recover estimated gestational age from unstructured notes and used gestational age to define the pregnancy episode for each individual pregnancy. Leveraging this cohort, we quantified and examined trends in the prevalence of maternal cardiovascular comorbidities and CVD and incidence of cardiovascular complications in pregnancies over the course of 19 years of follow-up. We also compared clinical factors for pregnancies with and those without cardiovascular complications. RESULTS:Our pregnancy cohort comprised 56 833 pregnancies among 38 996 individuals (mean age at start of pregnancy, 32±5 years). Regular expressions recovered gestational age for 75% of pregnancies, with good correlation between gestational age ascertained by regular expressions versus manual review (Pearson r=0.9). Among 56 833 pregnancies, overall prevalence of maternal CVD was 4% (age-adjusted 8%) and increased over 19 years (age-adjusted prevalence 1% in 2001 and 7% in 2019; P<0.001). Incidence of pregnancy-related cardiovascular complications was 15% (age-adjusted 17%) and increased over the study follow-up period (age-adjusted incidence 11% in 2001 and 13% in 2019; P<0.001). Cardiovascular complications within 1 year postpartum were more frequent in individuals with greater burden of maternal cardiovascular comorbidities and CVD (diabetes, 6% versus 3%; hypertension, 23% versus 5%; CVD, 10% versus 3%; P<0.001 for all). CONCLUSIONS:In a large-scale electronic health record-based pregnancy cohort, both the prevalence of maternal cardiovascular comorbidities and CVD as well as the incidence of pregnancy-related cardiovascular complications within 1 year postpartum increased over the course of 2 decades. Our findings demonstrate an alarming rise in pregnancy-related cardiovascular complications in a contemporary real-world setting and highlight pregnancy as a crucial life opportunity for cardiovascular health optimization.

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 remodeling. METHODS: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 remodeling (including global longitudinal strain [GLS] and E/e' ratio) using multiplicative interaction terms in multivariable-adjusted linear regression models. RESULTS: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 versus 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, SE 0.04 in women versus 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 versus men (ß 0.29, SE, 0.02 in women versus 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 versus men.