A Simplified Three-Item Clinical Score to Identify Exertional Hypoxemia in Fibrotic Interstitial Lung Disease: A Real-World Cohort Study

<b>Background:</b> Exertional oxygen desaturation (SpO₂ ≤ 88%) during the six-minute walk test (6MWT) is a key prognostic marker in interstitial lung disease (ILD), yet access to the test is often limited in clinical practice. Developing simple, bedside tools to identify patients at risk may support early risk stratification and guide clinical decision-making. <b>Methods:</b> We conducted a retrospective, real-world cohort study in a tertiary referral center between January 2024 and July 2025, including 150 patients, of whom 67.33% (101 patients) were using supplemental oxygen. Clinical and physiological data collected within 30 days of the 6MWT were analyzed. The primary outcome was exertional hypoxemia, defined as peripheral oxygen saturation (SpO₂) ≤ 88% at the end of the test. Four predictive approaches were evaluated: multivariable logistic regression, stepwise logistic regression, and a simplified clinical score (0-3). The simplified score assigned one point for each of the following: forced vital capacity (FVC) ≤ 61% predicted, diffusing capacity for carbon monoxide (DLCO) ≤ 53% predicted, and presence of chronic cough. Model performance was assessed by receiver operating characteristic (ROC) curves, sensitivity, specificity, predictive values, and risk stratification. <b>Results:</b> The simplified score demonstrated robust discriminative performance, comparable to more complex statistical models, with high sensitivity and acceptable specificity. A threshold of ≥2.0 points identified patients at high risk for exertional desaturation with 100% sensitivity and 0.66 specificity. Observed desaturation risk increased progressively across score categories: 17.1% for scores 0-1 (low risk), 58.6% for score 2 (intermediate risk), and 95.1% for score 3 (high risk). <b>Conclusions:</b> Compared with multivariable models, the simplified 0-3 clinical score-based on widely available variables (FVC ≤ 61%, DLCO ≤ 53%, and chronic cough)-maintained similar predictive performance (AUC 0.82) with greater operational simplicity. Owing to its high sensitivity and bedside applicability, it represents a promising screening tool for identifying patients at high risk of exertional desaturation, particularly when the 6MWT is unavailable.