A nomogram model for predicting maternal cardiovascular complications and neonatal adverse outcomes in pregnant patients with pulmonary arterial hypertension.

Background: Pulmonary arterial hypertension (PAH) during pregnancy significantly increases maternal and fetal mortality risk. We developed nomogram prediction models from retrospective data to assess maternal cardiovascular risks and neonatal adverse outcomes.

Methods: Our study included 170 pregnant women, divided into training (70%) and validation (30%) sets. Predictors of outcomes were identified using logistic regression in the training set, and nomograms were constructed to predict maternal cardiovascular complications and neonatal adverse outcomes. Model performance was evaluated through internal validation.

Results: Predictors of cardiovascular complications included severe PAH (OR = 4.80), New York Heart Association (NYHA) classification ≥ III (OR = 25.94), ST-T changes (OR = 25.18), total bilirubin (OR = 1.49), albumin (OR = 0.87) and lactate dehydrogenase level (OR = 1.01). The nomogram showed high predictive accuracy with concordance indices of 0.96 and 0.91, areas under the ROC curve of 0.96 and 0.93. Neonatal outcome predictors included gestational age at termination (OR: 0.93), maternal platelet count level (OR: 0.99), and B-type natriuretic peptide level (OR: 1.00). The corresponding nomogram showed concordance indices in the training set and validation set were 0.92 and 0.73, respectively, with area under the ROC curve values of 0.92 and 0.73.

Conclusions: Nomogram models based on the above factors useful tools for predicting cardiovascular complications and neonatal adverse outcomes in pregnant women with PAH, potentially aiding in early detection and timely intervention. Further validation is needed to confirm their accuracy in broader clinical settings.