C93-05 Evaluating the Multidimensional Diagnostic Schema for Chronic Obstructive Pulmonary Disease in the Absence of Spirometry

Abstract Introduction Chronic obstructive pulmonary disease (COPD) is diagnosed based on the presence of airflow obstruction on postbronchodilator spirometry. It is increasingly recognized that individuals without airflow obstruction may also have substantial structural changes on imaging, as well as respiratory symptoms compatible with the presence of COPD. We recently developed a multidimensional diagnostic schema for COPD that incorporates symptoms, imaging abnormalities, and spirometry. This schema identifies additional individuals with COPD who do not have spirometric airflow obstruction. In this study, we aimed to evaluate the diagnostic performance of the multidimensional diagnostic schema for COPD when spirometry is not available. Methods We analyzed participants from the multicenter COPDGene cohort of current and former smokers aged 40-85 years. COPD was diagnosed either by (1) major diagnostic category: presence of the major criterion (airflow obstruction) and at least 1 of 5 minor criteria (emphysema or bronchial wall thickening on computed tomography, dyspnea, poor respiratory quality of life, and chronic bronchitis); or (2) minor diagnostic category: presence of at least 3 of 5 minor criteria. Participants were then classified using only the minor diagnostic category, assuming spirometry was not available. The discriminative accuracy of the minor diagnostic category was evaluated against the multidimensional classification using the area under the receiver operating characteristic curve (AUC). We also conducted stratified analyses by sex and race to evaluate the discriminative accuracy of the minor diagnostic category alone across demographic subgroups. Results Among 9,416 participants, 4,695 (49.9%) had COPD using the multidimensional schema. Use of the minor diagnostic category alone identified 3,482 (37.0%) participants. The discriminative accuracy of the minor diagnostic category alone for detecting COPD was 0.86 (95%CI 0.85-0.87). The sensitivity and specificity were 0.73 (95%CI 0.72-0.74) and 0.99 (95%CI 0.98-0.99), respectively, and positive and negative likelihood ratios were 59.4 (95%CI 45.9-76.7) and 0.27 (95%CI 0.26-0.29), respectively. The discriminative accuracy for COPD detection was consistent across sex: 0.86 (95%CI 0.85-0.86) in men and 0.86 (95%CI 0.85-0.87) in women, and race: 0.85 (95%CI 0.84-0.86) in non-Hispanic Whites and 0.89 (95%CI 0.88-0.90) in non-Hispanic Black individuals. Conclusion In a population of current and former smokers, in the absence of spirometry, those meeting minor diagnostic criteria are highly likely to have COPD. However, for those not meeting these criteria, spirometric testing is required to confidently rule out COPD. Therefore, the use of spirometric airflow obstruction as a major criterion for COPD in the multidimensional schema is justified. This abstract is funded by: This work was supported by NHLBI R01 HL151421. The COPDGene study (NCT00608764) is supported by grants from the NHLBI (U01HL089897 and U01HL089856), by NIH contract 75N92023D00011, and by the COPD Foundation through contributions made to an Industry Advisory Committee that has included AstraZeneca, Bayer Pharmaceuticals, Boehringer-Ingelheim, Genentech, GlaxoSmithKline, Novartis, Pfizer and Sunovion.