Predictive efficacy of PASP combined with NT-proBNP for outcomes in pregnant women with severe cardiovascular disease: a single-centre retrospective observational study

Objective To analyze maternal complications and pregnancy outcomes in pregnant women with severe cardiovascular disease (CVD), and explore the predictive value of pulmonary artery systolic pressure (PASP) and N-terminal pro-B-type natriuretic peptide (NT-proBNP) for these outcomes. Method A retrospective observational study was conducted on pregnant women with severe CVD who were admitted to our hospital between December 2019 to August 2025. Demographic data, delivery mode, laboratory and imaging examination results, and post-partum follow-up data were collected. The primary outcome was CVD-related readmission or death during follow-up. Results A total of 158 patients (168 pregnancies) were included, with median follow-up of 1.8 (interquartile range 3.0) years. Among them, 86 cases achieved successful delivering, including 14 vaginal deliveries and 72 cesarean sections. Compared with the PASP < 50 mmHg group, the PASP ≥ 50 mmHg group had a significantly higher medical abortion rate (34.5% vs. 12.2%), and a notably lower average birth weight (2247.5 g vs. 3042.5 g, P = 0.007). ROC curve analysis revealed that the combined PASP + NT-proBNP index exhibited the highest predictive efficacy for the primary outcome (AUC = 0.748, P = 0.001), followed by NT-proBNP alone (AUC = 0.724, P = 0.003) and PASP alone (AUC = 0.664, P = 0.028). Kaplan-Meier survival analysis demonstrated that pregnant women with elevated PASP (≥ 44.5 mmHg; HR = 3.456, 95% CI: 1.367–8.734; P = 0.009) or NT-proBNP (≥ 241.25 pg/mL; HR = 6.863, 95% CI: 2.432–19.370; P < 0.001) had a significantly increased risk of cardiovascular-related hospitalization or death during follow-up. Conclusion PASP may serve as a reference indicator for guiding perinatal risk assessment in pregnant women with severe CVD, while NT-proBNP could help optimize the long-term prognostic prediction of this population. The combined application of these two parameters might provide a preliminary basis for risk stratification, which may assist in clinical decision-making regarding perinatal management and long-term prognosis monitoring.