Late-Breaking Abstract: Large Differences in Rate of Reported and Unreported Copd Exacerbations in Japanese Patients; Data from the Cosmos-J Trial

Background: The GOLD strategy proposes a COPD assessment framework based on symptoms and exacerbation risk;this study was conducted in Japanese patients with moderate-to-severe COPD using a protocol based upon monthly symptom assessment and occurrence of exacerbations (Betsuyaku et al Int J COPD 2013;8:453). Aims: To compare rate of physician-diagnosed COPD exacerbations (reported events) with unreported events identified using the Exacerbations of Chronic Pulmonary Disease Tool (EXACT). Methods: Patients were randomized to receive 24 weeks treatment with either salmeterol/fluticasone propionate 50/250μg BID or tiotropium 18μg QD and assessed every 4-weeks. Physician9s diagnosis of COPD exacerbations was made independently of any events recorded in the EXACT diary. Event rates were calculated using a negative binomial regression. Results: 406 patients participated; mean FEV 1 %predicted 59±13% and COPD assessment test score 12±6 at baseline. Three hundred sixty six subjects (90.1%) completed. Kaplan-Meier plots for time to first event between reported and unreported events showed differences in cumulative onset; EXACT detected 288 events (175subjects; 1.79/person-year 95%CI: 1.56-2.05), whereas physicians reported56 events (49subjects; 0.33/person-year; 95%CI: 0.24-0.44) - a 5.5 fold difference (post-hoc analysis). Six subjects with reported events were admitted to hospital. Conclusion: COPD exacerbations are reported to be relatively infrequent in Japanese patients (Suzuki et al ERJ 2013; 43:1289); this study suggests that under-reporting by patients or under-recognition by physicians may be factors contributing to this low rate. Funding: GSK (NCT01762800).