Prediction and prognosis of adverse maternal and foetal/neonatal outcomes in pulmonary hypertension: an observational study and nomogram construction

Abstract

Background Pregnant women with pulmonary hypertension (PH) have higher mortality rates and poor foetal/neonatal outcomes. Tools to assess these risk factors are not well established. Methods Predictive and prognostic nomograms were constructed using data from a “Development” cohort of 420 pregnant patients with PH, recorded between January 2009 and December 2018. Logistic regression analysis established models to predict the probability of adverse maternal and foetal/neonatal events and overall survival by Cox analysis. An independent “Validation” cohort comprised data of 273 consecutive patients assessed from January 2019 until May 2022. Nomogram performance was evaluated internally and implemented with online software to increase the ease of use. Results Type I respiratory failure, New York Heart Association functional class, N-terminal pro-brain natriuretic peptide $$\ge$$ ≥ 1400 ng/L, arrhythmia, and eclampsia with pre-existing hypertension were independent risk factors for maternal mortality or heart failure. Type I respiratory failure, arrhythmia, general anaesthesia for caesarean section, New York Heart Association functional class, and N-terminal pro-brain natriuretic peptide $$\ge$$ ≥ 1400 ng/L were independent predictors of pulmonary hypertension survival during pregnancy. For foetal/neonatal adverse clinical events, type I respiratory failure, arrhythmia, general anaesthesia for caesarean section, parity, platelet count, fibrinogen, and left ventricular systolic diameter were important predictors. Nomogram application for the Development and Validation cohorts showed good discrimination and calibration; decision curve analysis demonstrated their clinical utility. Conclusions The nomogram and its online software can be used to analyse individual mortality, heart failure risk, overall survival prediction, and adverse foetal/neonatal clinical events, which may be useful to facilitate early intervention and better survival rates.

Analysis of risk factors and construction of nomograph model for critical condition of patients with hypertension during pregnancy

“…Nomogram model, based on core diagnostic indicators, can be used for risky patient assessment and early disease diagnosis. It has been used effectively for various conditions, including prediction of adverse outcomes in pregnant women and fetuses with pulmonary arterial hypertension [ 19 ], radiomics-based prediction of the risk of preeclampsia in patients with hypertension during pregnancy [ 13 ], screening for new-onset hypertension in rural Chinese population [ 20 ], prediction of occult hypertension and occult uncontrolled hypertension based on clinical characteristics of single outpatient visit [ 21 ], etc. In this study, after integrating the risk factors of the multifactorial analysis results, the constructed visualized nomogram model had an AUC value of more than 0.8 in both internal and external validation, indicating good consistency and discrimination of the nomogram model herein.…”

Maternal and fetal/neonatal outcomes of pregnancies complicated by pulmonary hypertension: a retrospective study of 154 patients

“…In this study, the inclusion criteria were as follows: all subjects had undergone cardiac echocardiography, met the diagnostic criteria of echocardiography, had PH values determined according to the clinical diagnostic criteria, and had a PASP confirmed to be more than 35 mmHg by echocardiography. 11 , 12 , 13 The exclusion criterion was patients with elevated PASP caused by outflow tract obstruction/pulmonary stenosis.…”