The PVD-B65 risk score for interstitial lung disease and pulmonary hypertension: External validation using the PVDOMICS cohort

Background Validated prognostication tools for patients with pulmonary hypertension related to interstitial lung disease (PH-ILD) are lacking. Our group developed and internally validated a novel one-year mortality risk score for patients with pulmonary hypertension related to chronic lung disease, termed PVD-B65. Here, we aimed to externally validate the PVD-B65 risk score using the Pulmonary Vascular Disease Phenomics Study (PVDOMICS) cohort in patients with PH-ILD specifically. Materials and Methods We analysed patients with PH-ILD from the multicentre PVDOMICS cohort. We assigned each patient a PVD-B65 score with corresponding risk group. We assessed the effectiveness of the PVD-B65 score in predicting one-year mortality via receiver operator characteristic analysis and performed survival analysis comparing the three risk groups. We compared the performance of the REVEAL 2.0 and ESC/ERS 4-strata pulmonary arterial hypertension risk scores to the PVD-B65 score in predicting one-year mortality in PH-ILD. Results The PVD-B65 risk score effectively predicted one-year mortality in this PVDOMICS cohort of patients with PH-ILD (AUC=0.87, p<0.001). Kaplan-Meier survival analysis demonstrated significantly worse one-year survival probability in the high-risk group (logrank p<0.001). The REVEAL 2.0 (AUC=0.84) and ESC/ERS 4-strata risk assessment (AUC=0.81) risk scores were less effective in predicting one-year mortality in this cohort. Conclusion The PVD-B65 risk score was successfully externally validated within the PVDOMICS cohort, effectively stratifying and predicting one-year mortality in patients with PH-ILD and outperforming the REVEAL 2.0 and ESC/ERS 4-strata risk assessment scores.