Cardiovascular dysfunction is the leading cause of pregnancy-related mortality in the United States [1], [2]. When compared to White women, Black women have worse cardiovascular outcomes after accounting for socioeconomic factors and medical comorbidities. In a 2015 California study of 2,741,220 women, Black women comprised only 5 % of births but nearly 40 % of cardiovascular dysfunction-related deaths, and the most common conditions associated with mortality were cardiomyopathy, hypertension, and obesity [3]. The incidence of cardiovascular dysfunction during pregnancy has been increasing over time and parallels the rise in pregnancy-related deaths in the United States, which has risen from 7.2 deaths per 100,000 live births in 1987 to 22.3 deaths per 100,000 live births in 2022 [4], [5]. Despite this increased mortality risk, the ability to easily identify at risk patients is lacking. Few biomarkers have been associated with cardiomyopathy or the development of postpartum hypertension. Routine lab tests are generally poor indicators of impending critical disease in the postpartum period, and when these tests are abnormal end-organ disease has already manifested.

In previous studies, elevated levels of Activin A, a ligand for the activin receptor type 2 (ACTRII), was pathogenic in the development of postpartum cardiac dysfunction among preeclamptic women [6], [7], [8], [9], [10], [11]. Follistatin-like 3 (FSTL3) is a downstream target of ACTRII signaling and is upregulated when this pathway is activated by the ACTRII receptor binding with Activin A. In non-pregnant rodent models, increasing circulating Activin A increased cardiac signaling and FSTL3 expression and worsened cardiac function, and blockade of ACTRII signaling decreased FSTL3 levels and improved cardiac function [12]. In pregnant murine models, serum levels of Activin A increased with each subsequent pregnancy and correlated with worsening heart failure. Inhibition of Activin A in these models lead to decreased levels of FTSL3 and improved symptoms of heart failure [13]. FSTL3 expression is predominantly localized to the cardiac myocyte and vascular endothelium. Taken together, these data suggest that FSTL3 expression may also be pathogenic in the development of postpartum cardiovascular dysfunction defined as postpartum hypertension and postpartum cardiomyopathy [14], [15], [16].

Based on these findings, it was hypothesized that antepartum FSTL3 levels would be associated with postpartum cardiovascular dysfunction among pregnant women. To test this hypothesis, we evaluated the association between FSTL3 levels and postpartum cardiovascular dysfunction using data collected from an observational cohort study. In addition, the association between antepartum FSTL3 levels and severe maternal and neonatal morbidity was assessed. As a secondary analysis, we evaluated the associated between Activin A levels and cardiovascular dysfunction and severe maternal morbidity.