Advanced Echocardiographic Assessment Predicts 1-Year Mortality in Critically Ill Patients With Pulmonary Arterial Hypertension

Background: Despite advancements in diagnosing and managing pulmonary arterial hypertension (PAH), critically ill patients with PAH experience high mortality and current risk scores offer limited utility for risk stratification. Objectives: The purpose of this study was to evaluate whether echo-derived right heart metrics improve risk prediction for in-hospital and 1-year mortality in critically ill PAH patients. Methods: We analyzed PAH patients admitted to the intensive care unit from January 2010 to December 2020, with follow-up through January 2025. Echocardiographic assessments included right atrium (RA) area and phasic function, tricuspid annular plane systolic excursion (TAPSE), fractional area change, regional and global right ventricular (RV) free wall strain (RVFWS), and RV to pulmonary artery (PA) systolic pressure (PASP) coupling ratios. Cox regression evaluated in-hospital and 1-year mortality, and a random survival forest (RSF) model identified nonlinear predictors of 1-year mortality. Results: Of 102 patients, 77 (20 idiopathic PAH, 45 connective tissue disease-associated PAH, 12 other subtypes) had adequate echocardiograms. Cox regression revealed significant associations between in-hospital mortality and TAPSE and RA enlargement. Impaired TAPSE, RVFWSglobal, and RV-PA coupling predicted a higher 1-year mortality. Kaplan-Meier analysis identified prognostic thresholds for TAPSE, TAPSE/PASP, RVFWS, and RVFWS/PASP. RSF modeling identified TAPSE, left ventricular diastolic filling parameters, left ventricular end-diastolic volume, and fractional area change/PASP as significant mortality predictors, with a time-varying area under the curve of 0.84. Conclusions: In critically ill PAH patients, echocardiographic markers were independently associated with mortality. A novel RSF model identified key variables for risk prediction, emphasizing the value of comprehensive echocardiographic assessment in this high-risk population.