This study investigates how the recently published ERS/ATS guidelines for lung function interpretation (Stanojevic 2021) affect patient classification compared to the previous (Pellegrino 2005). 1325 subjects admitting to secondary respiratory practice (244 healthy, 1081 disease) were classified by both guidelines via an algorithm (based on pattern, severity, and bronchodilator response (BDR)). We observed limited impact in the pattern description: 100/1325 subjects are classified to a novel category for the lung function (Nonspecific category, previously obstructive with a normal FEV1/FVC, Fig 1A), and 14 for the diffusion (Fig 1C). Z-scores to define severity introduced a large shift in classes ranging from mild, moderate to severe obstruction/restriction, with milder labels based on FEV1 and TLC, but opposite shifts when DLCO was taken. The new BDR is stricter with 27% of the significant subjects changing to insignificant BDR, particular with lower FEV1 (Fig 1D). Figure. Comparison of 2005 and 2021 guidelines for lung function test interpretation applied on 1325 different subjects. Panel A/Lung function patterns; Panel B/FEV1/TLC severity Panel C/Diffusion severity; D/ Bronchodilator response. The 2021 guidelines add more categories to distinguish lung function impairments. The severity labels substantially change and BDR is stricter in a respiratory disease population. The impact on current disease management should be investigated.

Interstitial lung disease (ILD) patients are often misdiagnosed or diagnosed late. Earlier diagnosis allows earlier access to correct treatment, consequently improved survival and quality of life. We investigated if AI software can detect patterns of ILD on spirometry earlier to formal diagnosis. All subjects from UK Biobank with ILD as cause of death but without clinical ILD diagnosis at the time of spirometry, were selected (N=109). Spirometry data and subject demographics (gender, age, BMI, race, smoking) were input to an AI clinical decision support software (ArtiQ.PFT), which calculates a probability of having present one of lung diseases. Time between last recorded spirometry and clinical ILD diagnosis was 3.8±1.9 years, with 1.1±1.2y more to ILD death. AI detected ILD in 27% of subjects (N=29), 3.8±2.1 years prior to ILD diagnosis through standard care. 66% (N=19) of these subjects had normal lung function, 34% (N=10) were restrictive. In remaining 73% (N=80), AI most commonly pointed to no disease (39%) or was uncertain (18%). Spirometry in the group where AI identified ILD were significantly different from those where ILD was not detected [FVC%pred 76±12 vs. 87±17 (p=0.003), FEV1%pred 80±13 vs. 84±21 (p=0.36), PEF%pred 103±25 vs. 87±29 (p<0.0001), FEV1/FVC 0.82±0.05 vs. 0.75±0.09 (p=0.007)]. There was no difference in time to mortality, nor in time between spirometry and clinical diagnosis among groups. In one-quarter of patients, AI detected ILD many years before clinical diagnosis. Most had lung function within normal ranges, suggesting that AI may detect ILD prior to standard interpretation. With further validation, merging AI-support and regular spirometry testing may lead to earlier ILD diagnosis.