Pulmonary hypertension (PH) is a progressive cardiopulmonary disease characterized by elevated pulmonary artery pressure and vascular remodeling, which leads to right heart failure and increased mortality. Despite advances in diagnostics, risk stratification remains limited due to the disease's heterogeneity. This study aims to develop and validate a dynamic risk prediction model for PH by integrating multimodal data-including echocardiography, Cardiac MRI, PET-MR, ECG, biomarkers, and clinical features-using advanced machine learning algorithms. The study will establish a prospective cohort of PH patients to explore predictive markers, stratify prognosis, and provide a scientific basis for early warning and individualized management.
Adults aged 18 years or older
Pulmonary artery systolic pressure (PASP) ≥35 mmHg as estimated by echocardiography
Provided written informed consent
Severe hepatic or renal insufficiency
Malignancy under active treatment
Severe infection
Active autoimmune disease
The study population includes adult patients (≥18 years) undergoing transthoracic echocardiography. Participants with a pulmonary artery systolic pressure (PASP) ≥35 mmHg on echocardiography will be included as suspected pulmonary hypertension cases. This cohort represents a real-world population at risk for or with early-stage pulmonary hypertension, suitable for developing dynamic risk prediction models based on multimodal data integration.
Cohort
Prospective
Not listed
Time to clinical worsening
Defined as any of the following: hospitalization for PH, escalation of therapy, 6MWD decrease \>15%, WHO-FC worsening, or death. Measured from baseline.
All-cause mortality
Death from any cause during follow-up, as confirmed by medical records or death registry.
Composite risk score performance (AUC)
Area under the ROC curve for the multimodal model predicting adverse outcomes.
Changes in NT-proBNP levels
Evaluate biomarker dynamics and predictive value.
Hospitalization rate for PH-related causes
Frequency of hospital admissions due to PH complications.
Change in Tricuspid Annular Plane Systolic Excursion (TAPSE) Measured by Transthoracic Echocardiography
TAPSE (mm) will be measured via transthoracic echocardiography to evaluate longitudinal right ventricular systolic function over time.
Longitudinal changes in health-related quality of life (HRQoL) among patients with suspected or confirmed pulmonary hypertension
Health-related quality of life (HRQoL) will be assessed using validated instruments such as the EQ-5D-3L and/or the emPHasis-10 questionnaire, both commonly used in pulmonary hypertension research. The emPHasis-10 is a disease-specific, patient-reported outcome measure developed for individuals with pulmonary hypertension, covering domains such as breathlessness, fatigue, social limitation, and psychological burden. Changes in HRQoL scores will be evaluated over time and correlated with clinical events (e.g., hospitalization, WHO functional class deterioration), imaging parameters (e.g., RV function), biomarkers (e.g., NT-proBNP), and model-predicted risk strata. Analyses will include repeated measures ANOVA or mixed-effects modeling to examine within-subject longitudinal chan
Area Under the ROC Curve (AUC) of the Multimodal Risk Prediction Model
The area under the receiver operating characteristic curve (AUC) will be calculated to evaluate the discrimination performance of the multimodal dynamic risk prediction model for adverse outcomes.
Harrell's Concordance Index (C-index) of the Multimodal Risk Prediction Model
The concordance index (C-index) will be computed to assess the predictive accuracy of the multimodal risk prediction model.
Calibration Slope of the Multimodal Risk Prediction Model
Calibration slope will be estimated to measure agreement between predicted and observed risks in the dynamic risk prediction model.
Click on the lead sponsor to view all of their trials.
