Comparison of chronic bronchitis definitions in COPD for predicting 10-year all-cause mortality.

BACKGROUND

Chronic bronchitis (CB) is a common phenotype of chronic obstructive pulmonary disease (COPD) associated with poor outcomes. This study aimed to compare the predictive performance of a CB definition based on the COPD Assessment Test (CAT-CB) with that of the classic definition for 10-year all-cause mortality in patients with COPD.

METHODS

We analyzed 3476 participants in a prospective, multicenter study, namely, the Korea COPD Subgroup Study (KOCOSS). CB was classified using both the classic ATS definition (cough + sputum ≥3 months per year for 2 consecutive years) and the CAT-CB definition (CAT item 1 [cough] > 2 and item 2 [sputum] > 2). CAT-CB was further stratified as mild (any item ≤3) or severe (both items >3). Mortality data were ascertained from national death records; follow-up was truncated at 10 years. Cox proportional hazards models were used to evaluate mortality risk.

RESULTS

At baseline, 354 patients (11.0 %) met the classic CB definition and 774 (22.6 %) met the CAT-CB definition. CAT-CB was independently associated with higher all-cause mortality, whereas classic CB was not. Mortality risk was confined to the severe CAT-CB subgroup, whereas patients with mild CAT-CB showed no significant excess risk. Individually, CAT-cough >2 and CAT-sputum >2 each predicted mortality; the classic subcomponents did not.

CONCLUSIONS

A CAT-based definition of CB better identifies COPD patients with increased risk of long-term mortality relative to the classic definition.