Noninvasive Predictors of Pulmonary Hypertension in Interstitial Lung Disease Identified

Readily available and easy-to-execute noninvasive tests have the ability to predict pulmonary hypertension (PH) in interstitial lung disease (ILD) with a high probability, suggest new study findings published in Pulmonary Circulation. These approaches, wrote the researchers, can aid in determining which patients should proceed with invasive diagnostic methods and receive targeted treatment.

Currently, the standard for diagnosing PH in ILD involves right heart catheterization, which is both invasive and not widely accessible, creating an opportunity for noninvasive approaches. The researchers noted that their study is the first to assess the ability of noninvasive variables, both resting and exercise based, to identify PH in ILD.

“Mortality amongst ILD patients with concomitant PH remains unacceptably high,” wrote the researchers. “As advances in PAH-targeted therapeutics for PH‐ILD become more readily available, there will be an increasing need for early identification of patients likely to benefit from earlier institution of PAH‐directed therapy.”

Compared with the 22 patients with ILD alone, the 44 patients with PH-ILD demonstrated additional deficiency in pulmonary function test, echocardiographic, and noninvasive exercise testing parameters. For example, patients with PH-ILD showed reduced results on the 6-minute walk distance (6MWD) and heightened systemic O₂ desaturation. They also exhibited some abnormal gas exchange parameters on submaximum exercise testing.

The mean (SD) gas exchange derived estimate of pulmonary vascular capacitance (GX_CAP; area under the curve [AUC], 0.85 [0.04]; P < .0001) and delta end tidal carbon dioxide (delta ETCO₂; AUC, 0.84 [0.04]; P < .0001) were the strongest predictors of PH‐ILD in the patients with both PH and ILD. Other noninvasive variables that the researchers assessed, but that were less discriminatory, included diffusing capacity for carbon monoxide (DLCO), ventilatory inefficiency, forced vital capacity to DLCO (FVC/DLCO) ratio, 6MWD, and O₂ pulse.

Although sensitive, GX_CAP and delta ETCO2 and were not specific to help determine PH in ILD, leading researchers to perform a CART analysis that accounted for all variables, which showed that GX_CAP, estimated right ventricular systolic pressure (eRVSP) by echocardiogram, and FVC/DLCO ratio were predictive.

With this analysis, the AUC increased to 0.94, indicating that the decision tree incorporating the variables was superior to each individual variable.

GXCAP had the highest importance score. Patients who had a GX_CAP at or under 416 mL x mm Hg had an 82% probability of PH-ILD, patients with GX_CAP at or under 416 mL x mm Hg plus a high FVC/DLCO ratio had an 80% probability, and patients with GX_CAP at or under 416 mL x mm Hg plus an elevated eRVSP had a 100% probability.

Multivariate analysis showed that GX_CAP was also the only variable that was able to predict extrapolated maximum oxygen uptake, previously shown to be an important prognostic indicator in PAH, in both groups of patients.

“Taken together, GX_CAP can serve as an important adjunctive diagnostic tool to help screen ILD patients for PH, while also offering a pathophysiological reasoning for the differential exertional intolerance experienced by PH‐ILD and non‐PH ILD,” detailed the researchers.

Reference

Joseph P, Savarimuthu S, Zhao J, et al. Noninvasive determinants of pulmonary hypertension in interstitial lung disease. Pulm Circ. Published online January 27, 2023. doi:10.1002/pul2.12197