
OR WAIT null SECS
Lau discusses the importance of developing and applying technology to predict potential adverse cardiovascular outcomes in pregnant patients.
Lower estimated cardiorespiratory fitness has been associated with higher risks of pregnancy-related cardiovascular complications, according to a recent study.1
Presented at the American Heart Association’s Scientific Sessions 2025 in New Orleans, Louisiana, by Emily Lau, MD, Massachusetts General Hospital and professor of cardiology at Harvard University, and colleagues, these data also indicate the predictive power of artificial intelligence-enabled electrocardiograms (ECG), which were utilized during the trial.1
Cardiovascular disease is the leading cause of indirect maternal mortality; it accounts for >33% of all pregnancy-related maternal deaths. Additionally, ≥50% of maternal deaths happen post-partum. However, other research has indicated up to 68% of pregnancy-related deaths through cardiovascular conditions are preventable.2
The editorial team at HCPLive sat down with Lau to discuss the study and its implications. Lau discussed the methods used to collect and analyze the data used in the study, noting the artificial intelligence model they utilized to filter through screening pregnancies at the Massachusetts General Brigham system.
“One of the things we wanted to do is to develop tools that can help us identify women who are at high risk for developing a cardiovascular complication, so that we might be able to help target preventive therapies and identify which women need greater screening during their pregnancy” Lau told HCPLive. “We are using a really novel model; we had previously developed an EKG deep learning model to estimate cardiorespiratory fitness from the 12-lead ECG.”
Lau and colleagues had previously developed a deep learning model to accurately estimate peak oxygen uptake (VO2) using a resting 12-lead ECG. The team examined ECG-estimated VO2 among patients who underwent clinical 12-lead ECG testing between 1 year before pregnancy and 13 weeks of gestation. Investigators used age-adjusted logistic regression models to examine the association between estimated peak VO2 and subsequent pregnancy-related cardiovascular complications ≤1 year postpartum, which included maternal death, severe hypertensive disorders of pregnancy, and major adverse cardiac events.1
A total of 3437 women were included in the study, resulting in a total of 3650 pregnancies. Mean age at delivery was 33 +/- 6 years, and the median ECG-estimated VO2 was 26.5 mL/kg/min. Investigators noted that 26% of patients experienced a pregnancy-related cardiovascular complication.1
Lower ECG-estimated peak VO2 was associated with greater risk of a pregnancy-related complication (odds ratio [OR], 1.18 per 1-unit lower metabolic equivalent [MET = 3.5 kg/m2]; 95% CI, 1.15-1.23; P <.001). Women in the lowest quartile of ECG-estimated peak VO2 had almost 2 times the odds of developing a complication compared to the highest quartile (OR, 2.36; 95% CI, 1.91-2.93; P <.001).1
Ultimately, investigators noted the strong and independent association between lower estimated cardiorespiratory fitness and pregnancy-related cardiovascular complications. They also highlighted the effectiveness of a routinely-performed AI-enabled analysis of ECGs in antepartum care, which they believe may enable scalable risk assessment for identifying high-risk pregnancies.1
“Now the next step after the study is really to think about a prospective study where we actually perform ECGs in all women, including women who would not have qualified for getting an ECG under a current obstetrical practice,” Lau said. “We need to identify how ECG-estimated cardiorespiratory fitness captures risk of pregnancy-related cardiovascular complications across an entire sample of individuals who are less sick.”
Related Content: