Association of Single Breath Carbon Monoxide Diffusion Capacity with Exercise-Induced Oxygen Desaturation: Differences amongst Obstructive and Restrictive Respiratory Disorders

Background: While reduced DLCO is associated with impaired oxygenation, its relationship to exercise-induced oxygen desaturation (EID) across different lung disease phenotypes remains incompletely characterized. This study aimed to evaluate the association between DLCO and EID across obstructive, restrictive, and normal lung physiologies, and assess its utility and other pulmonary functions as identifying clinically significant EID. Methods: Of 600 patients screened, 196 adults underwent complete pulmonary function testing and exercise desaturation studies performed within six months. EID was defined as ≥4% decrease in oxygen saturation. Linear regression models assessed associations between desaturation and DLCO, adjusting for age, sex, ethnicity, height, smoking history, and disease classification. Receiver operating characteristic (ROC) curve analyses were performed to evaluate discriminatory ability of DLCO and other PFTs for predicting EID. Results: Higher DLCO values were significantly associated with smaller decreases in oxygen saturation during exercise. After adjustment, each 1-L increase in DLCO was associated with a 0.14% smaller desaturation (p < 0.001), and each 1% increase in percent predicted DLCO corresponded to a 0.035% smaller desaturation (p < 0.001). No significant interaction was observed between DLCO and lung disease classification or smoking history. DLCO moderately associated with desaturation (r = ？0.35, p < 0.001). ROC analyses showed modest discriminatory performance for DLCO, comparable to TLC and FVC; other pulmonary function variables performed less well. Conclusions: DLCO is inversely associated with EID across obstructive, restrictive, and normal lung function patterns. However, DLCO and other PFTs demonstrate only a modest association with clinically significant desaturation amongst disease categories.