The training cohort included 397 patients with COPD-PH, diagnosed using right cardiac catheterization, from four Chinese hospitals from December 2008 to July 2022, to develop a nomogram for predicting severe PH in COPD. The external validation cohort included 138 patients from two additional hospitals. Multivariate logistic regression analysis was used to analyse the risk factors of severe PH in patients with COPD, and multivariate Cox regression for the prognostic factors of COPD-PH. Internal validation was performed using bootstrapping and the external cohort. Performance was assessed using the calibration, area under the receiver operating characteristic, and decision analysis curves. Risks were stratified by prognosis nomograms. Kaplan?Meier curves were used for survival analysis.

Results: Tricuspid regurgitation velocity, right ventricular diameter, N-terminal pro-brain natriuretic peptide (NT-proBNP), New York Heart Association functional class, and sex were independent variables of severe PH in COPD and used for constructing a diagnostic nomogram with good discrimination. NT-proBNP, mean pulmonary arterial pressure, partial pressure of arterial oxygen, platelet, and albumin were independent prognostic factors of COPD-PH and used to create a predictive nomogram of overall survival rates. The radiomics nomogram was found clinically useful and the prognostic nomogram showed good predictive accuracy.

Conclusion: We established a nomogram for identifying severe PH in COPD and a prognostic nomogram for overall survival prediction of COPD-PH. These nomograms and accompanying web software can be effectively used in clinical practice.