Abstract Rationale Patients with chronic pulmonary diseases may be at risk of clinical deterioration during air travel due to reduced cabin pressure. In-flight oxygen prescription is typically guided by hypoxic challenge testing (HCT), a procedure that is time-consuming, costly, and not universally available. Several predictive equations based on sea-level physiological parameters have been proposed as alternatives. This study aimed to compare the performance of these equations against HCT results in patients with different underlying lung disorders. Methods We retrospectively analyzed all patients who underwent HCT at our center between 2022 and 2024. Sea-level arterial blood gases, pulmonary function tests, and relevant clinical data were collected. Six published predictive equations (E1-E6) were used to estimate in-flight PaO2. These estimates were compared with measured PaO2 during HCT (PaO2-HCT) using correlation coefficients, sensitivity and specificity analyses. Intra-class correlation (ICC) was used to assess the homogeneity of predictive equations. Discriminative ability was assessed using receiver operating characteristic (ROC) curves. A positive HCT was defined as a PaO250 mmHg (or a SpO285% for patients with CF). Results Ninety-six patients were included 24 (25%) with COPD, 23 (24%) with ILD, 20 (21%) with CF, and 29 (30%) with other pulmonary diseases. Mean age was 59±17 years, 39 patients were male (58,2%), and mean DLCO was 62,7±22,7% predicted. Correlations between PaO2-HCT and predicted in-flight PaO2 were low to moderate but statistically significant (rho ranging from 0.37, p = 0.009 for E5 to 0.49, p 0.001 for E4). Across all patients, predictive equations demonstrated high specificity (80% for E2 to 98% for E6) but low sensitivity (7% for E6 to 64% for E2). All equations demonstrated low agreement (ICC(2,1)=0.54, 95%CI 0.30-0.72, p 0.001), indicating limited homogeneity of predicted PaO2 values. Areas under the ROC curve ranged from 0.66 95%CI 0.50-0.82 for E4 to 0.78 95%CI 0.63-0.92 for E1. Optimal predicted-PaO2 thresholds varied widely (55 mmHg for E5 to 69 mmHg for E4) and remained associated with low sensitivity. Results were similar when restricted to diagnosis subgroups. Conclusions In our patients with COPD, ILD, or CF undergoing HCT, currently available predictive equations for in-flight PaO2 showed good specificity but inadequate sensitivity when compared with measured PaO2-HCT. Reliance on these equations may underestimate the degree of in-flight desaturation, provide false reassurance, and lead to under-prescription of in-flight oxygen. Additional studies are needed to identify more accurate predictors of hypoxia at altitude and develop improved predictive models. This abstract is funded by: None
Beaupré et al. (Fri,) studied this question.