Screening of Pulmonary Hypertension in Interstitial Lung Disease in the PHINDER Study: A Comparison of Physician Gestalt With Hemodynamics

Abstract Rationale: Pulmonary hypertension (PH) complicating interstitial lung disease (ILD-PH) has a poor prognosis. Despite significant morbidity associated with ILD-PH, there are no evidence-based screening recommendations for PH in the setting of ILD. The PHINDER study (NCT05776225) aims to develop a PH screening tool in patients with ILD. PH is confirmed in all patients by right-heart catheterization (RHC). This analysis compares clinicians’ gestalt of PH with hemodynamically confirmed PH. Methods: PHINDER is a prospective, multicenter study intending to enroll 200 ILD patients who are suspected to have PH. Patients receive study clinical assessments followed by right heart catheterization (RHC). Precapillary ILD-PH is defined by 2022 ESC/ERS guidelines: mean pulmonary arterial pressure >20 mmHg, pulmonary artery wedge pressure ≤15 mmHg, and pulmonary vascular resistance >2 Wood Units. Prior to the RHC, clinicians are required to document their gestalt-based suspicion of PH (low, medium, high). Results: Data were available for 58 patients at this interim analysis (Table 1). There were 8, 27, and 23 patients at low, medium, and high suspicion of PH, respectively. In these 3 groups, 4 (50%), 11 (41%), and 14 (61%) of patients respectively met the hemodynamic definition of precapillary ILD-PH. Conclusions: Findings suggest that patients with higher suspicion of PH (gestalt based) had lower 6-minute walk distance, increased supplemental oxygen use, increased right ventricle systolic pressure and decreased tricuspid annular plane systolic excursion (TAPSE) on echocardiography, and an enlarged pulmonary artery (PA) on HRCT. Increased forced vital capacity to diffusing capacity (FVC/DLCO) ratio, decreased DLCO, decreased TAPSE/systolic PA pressure ratio, and enlarged PA were more pronounced among patients with RHC-confirmed PH in all 3 suspicion groups, suggesting that these parameters may be useful to clinicians assessing risk of PH. Surprisingly, brain natriuretic peptide (BNP) and N-terminal proBNP do not seem prognostic of PH thus far in the study. Physical exam findings did not correlate well with either clinical suspicion of PH or RHC-confirmed PH. Even with low clinical suspicion (a situation in which clinicians may not obtain a RHC), PH may be missed in 50% of patients. Results of this analysis underscore the difficulty in using clinical gestalt alone to assess the risk of PH.