Validation of handheld spirometry in Interstitial Lung Diseases (ILD) as part of the European ILD Registry (eurILDreg)

Abstract

Background: Interstitial lung diseases (ILD) are a heterogeneous group of diseases characterized by frequent irreversible destruction of the alveolar structure in the lungs. ILDs are often progressive in nature, leading to an incremental worsening of dyspnoea, reduced exercise tolerance, and a reduction in the quality of life over time. Despite novel treatment approaches, most forms are severely life-limiting, and the course of the disease is highly disparate, rendering current monitoring regimes for ILDs inadequate. While regular spirometry monitoring is the standard approach for tracking disease progression, handheld monitoring has been shown to cater more effectively to the variable progression of ILDs. This study aimed to validate mobile over-App handheld spirometry and home monitoring to effectively monitor lung function in ILDs. Using a novel artificial intelligence (ArtiQ) based algorithm, lung function data was assessed in real time, providing immediate feedback on measurement quality. Patient adherence, correlation to established on-site spirometry measurements, and longitudinal FVC changes were evaluated in a European multi-center study as part of the European ILD Registry (eurILDreg). Results: Data from 59 eurILDreg patients who provided consent for handheld monitoring were considered. Weekly adherence for handheld monitoring decreased to 81.1% and 73.5% after 3 months, 52.0% and 45.5% after 6 months and 36.8% and 29.4% at 12 months for handheld spirometry and exercise testing respectively. Correlation to on-site FVC was calculated at r=0.90 (p<0.001), reliability at 0.89 (p<0.001) and agreement with a bias of -0.177 L (upper/ lower limit 0.712 L and -1.067 L respectively). Correlation to on-site FEV1 was calculated at r=0.87 (p<0.001), reliability at 0.77 (p<0.001) and agreement with a bias of -0.373 L (upper/ lower limit 0.395 L and -1.142 L respectively). Direction of longitudinal FVC change was corroborated between handheld and on-site spirometry in 60% of cases (t= - 0.912, p=0.386, 95% CI: - 6.00ml/week to 2.60 ml/week). Conclusions: We reported acceptable and comparable adherence pattern to previous studies when adjusted for the methodology. Similarly, agreement and correlation with current state of the art spirometry supports its use, a testament to the value of employing ArtiQ algorithms. Also, our novel approach regarding ArtiQ gave substantial insight into the direction of home monitoring for detecting disease progression and AEs whilst integrating well with the current gap of literature.