Small Airway Disease in Pulmonary Hypertension-Additional Diagnostic Value of Multiple Breath Washout and Impulse Oscillometry

Abstract

Airways obstruction is frequent in patients with pulmonary hypertension (PH). Small airway disease (SAD) was identified as a major contributor to resistance and symptoms. However, it is easily missed using current diagnostic approaches. We aimed to evaluate more elaborate diagnostic tests such as impulse oscillometry (IOS) and SF₆-multiple-breath-washout (MBW) for the assessment of SAD in PH. Twenty-five PH patients undergoing body-plethysmography, IOS and MBW testing were prospectively included and equally matched to pulmonary healthy and non-healthy controls. Lung clearance index (LCI) and acinar ventilation heterogeneity (S_acin) differed significantly between PH, healthy and non-healthy controls. Likewise, differences were found for all IOS parameters between PH and healthy, but not non-healthy controls. Transfer factor corrected for ventilated alveolar volume (TLCO/VA), frequency dependency of resistance (D5-20), resonance frequency (F_res) and S_acin allowed complete differentiation between PH and healthy controls (AUC (area under the curve) = 1.0). Likewise, PH patients were separated from non-healthy controls (AUC 0.762) by D5-20, LCI and conductive ventilation heterogeneity (S_cond). Maximal expiratory flow (MEF) values were not associated with additional diagnostic values. MBW and IOS are feasible in PH patients both providing additional information. This can be used to discriminate PH from healthy and non-healthy controls. Therefore, further research targeting SAD in PH and evaluation of therapeutic implications is justified.

Keywords: impulse oscillometry; lung clearance index; multiple breath washout; pulmonary hypertension; small airway disease.

Figure 1

Diagnostic performance: Receiver operating curve (ROC) analysis for conventional and novel lung function parameters in patients with pulmonary hypertension (PH) vs. healthy controls (A,B) as well as patients with PH vs. non-healthy controls (C,D). Solid black lines indicate best performing parameter of the respective panel. Abbreviations are identical with Table 2. (A) Overall good performance of individual parameters selected for inclusion in the generalized linear model. (B) Parameters not selected for inclusion in the generalized linear model. (C) Generalized linear model with improved diagnostic performance as compared to individual parameters. (D) Parameters not selected for inclusion in the generalized linear model.