Pulmonary hypertension optimal screening in interstitial lung disease (PHOS-ILD): a scoring system for the detection of pulmonary hypertension in patients with interstitial lung disease

Abstract

Pulmonary hypertension (PH) is a common comorbidity of interstitial lung disease (ILD). Timely detection may improve outcomes through implementation of PH specific treatments. Retrospective, multicenter study aiming to identify features that predict PH in patients with ILD. In the derivation cohort (ILD cohort—University of South Florida), we identified predictors of PH in ILD and created a scoring system using the rounded value of odds ratio for each significant parameter. Correlation with hemodynamic parameters was investigated in the derivation and validation cohort (Idiopathic Pulmonary Fibrosis cohort—Cleveland Clinic). We included 164 patients with ILD that underwent right heart catheterization (RHC) [derivation cohort: n = 118, median age (years, 95% CI): 69.0 (65.9–71.0), female gender: 65 (54.6%), median mean pulmonary arterial pressure (mPAP) (mmHg, 95% CI): 25.0 (23.0–27.0), median pulmonary vascular resistance (PVR) [Wood units (WU), 95% CI]: 3.6 (3.0–4.3), validation cohort: n = 46, median age (95% CI): 63.0 (60.0–66.0), female gender: 17 (37%), median mPAP (mmHg, 95% CI): 20.5 (18.0–27.2), median PVR (WU, 95% CI): 3.1 (2.2–3.4)]. Parameters and optimal cut-offs that significantly predicted precapillary PH in the derivation cohort were: age>42 years, heart rate recovery (HRR) at 1 min ≤ 15 bpm in 6-min walk test (6 MWT), 6-min walk distance (6 MWD) ≤ 270 meters, B-type natriuretic peptide (BNP) > 136 pg/mL, pulmonary artery to ascending aorta diameter ratio (PA/AA) > 1 in high-resolution computed tomography (HRCT), main PA diameter > 29 mm in HRCT, right ventricular systolic pressure > 42 mmHg on echocardiogram. We created a PH detection model that ranged from 0 to 28. The optimal cut-off threshold for the detection of PH was > 14 (80% sensitivity, 68% specificity, AUC: 0.78). A simplified scoring system without BNP and main PA diameter (range: 0–21) maintained the same AUC. The simplified scoring system exhibited positive correlation with mean pulmonary artery pressure both in the derivation (rho = 0.47, p < 0.0001) and validation cohort (rho = 0.36, p = 0.013). This study demonstrates that incorporation of demographic, functional, imaging, and echocardiographic characteristics can optimize the early identification of precapillary PH in patients with ILD, paving the way to structured screening strategies. Timely detection of PH might lead to timely implementation of emerging PH specific treatments in patients with ILDs.