Document resource
Introduction A decline in forced vital capacity (FVC) exceeding 10% is a well-established prognostic marker in interstitial lung disease (ILD). 1 The 2021 Global Lung Initiative (GLI) reference values for relaxed vital capacity (RVC) yield a more severe disease classification than the 2012 GLI FVC values in idiopathic pulmonary fibrosis.2 This study examines how these differing reference equations influence ILD severity classification and progression monitoring.Methods Retrospective spirometry analysis in ILD patients (August 2022–2023, n=169). For each patient, we recorded%predicted values and z-scores, and disease severity was graded using the American Thoracic Society/European Respiratory Society and the Association for Respiratory Technology & Physiology criteria. Paired t-tests were used to analyse differences. We compared baseline classifications and lung function changes between patients evaluated with a consistent predictive model and those with alternating models between visits. Differences in%predicted values and z-scores were then examined to assess the impact of switching models. Clinical Governance Approval was obtained.Results GLI 2012 produced a mean FVC%predicted of 88.1% (z-score –0.79), significantly higher than the 82.6% (z-score –1.31) from GLI 2021. Consequently, more patients were classified as moderately to severely impaired. Consistent model use led to minor%predicted reductions (2.4% and 2.0% for GLI 2012 and GLI 2021, respectively). Longitudinal analysis revealed larger shifts when switching from GLI 2012 to GLI 2021; FVC%predicted decreased by 6.59% with this switch, compared to a 2.19% reduction with the alternative order. Similar trends were observed in z-score differences. Baseline differences and changes observed with model switching were approximately twofold greater in non-Caucasian patients.Conclusions The choice of GLI reference model significantly affects ILD severity classification and progression monitoring. Our findings indicate that the 2021 GLI values may be more sensitive in detecting both early and advanced impairment, highlighting the benefit of considering both RVC and FVC in longitudinal monitoring. To avoid systematic bias and misinterpretation, clinicians should consistently use the same predictive model when tracking ILD patients.References Eisa, et al. Rheumatology 2024;63. Donovan, et al. ERJ. 2023;62.