BACKGROUNDThe Global Initiative for Chronic Obstructive Lung Disease (GOLD) presented a new ABCD group classification of chronic obstructive pulmonary disease (COPD). We aimed to examine the association of spirometric classification and the new GOLD classification with exacerbations, and to compare symptoms in different ways.

METHODSWe investigated 848 patients with stable COPD from 24 hospitals. The annual frequencies of acute exacerbation and hospitalization were compared between the old and new classification. The symptom level was assessed using COPD assessment test (CAT) and modified British Medical Research Council (mMRC) questionnaire.

RESULTSA total of 848 patients were included in this study. According to spirometric classification, there were 32 patients of grade I (3.8%), 315 of grade II (37.1%), 366 of grade III (43.2%), and 135 of grade IV (15.9%). According to GOLD 2011 classification, there were 59 patients of group A (7.0%), 172 of group B (20.3%), 55 of group C (6.5%), and 562 of group D (66.3%). In spirometric classification, the annual frequencies of acute exacerbation and associated hospitalization were respectively 1 (0-3) and 0 (0-2) for grade I; 1 (0-5) and 0 (0-2) for grade II; 2 (0-6) and 1 (0-3) for grade III, and 3 (0-6) and 2 (0-3) for grade IV. In GOLD 2011, respectively 0 (0-3) and 0 (0-1) (group A), 1 (0-4) and 0 (0-3) (group B), 1 (0-5) and 0 (0-3) (group C), and 3 (0-6) and 1 (0-3) (group D). There were no significant difference between group B and C (Z = -1.347, P = 0.178; Z = -0.772, P = 0.440, respectively). The coincidence rate using mMRC = 1 and CAT = 10 as cutoff points was 86.6% (734/848, k = 0.706), compared with 77.9% (661/848, k = 0.60) using mMRC = 2 and CAT = 10.

CONCLUSIONSLung function test may be a better predictor of acute exacerbation and associated hospitalization of COPD. It is more appropriate to use mMRC = 1 as cutoff point for assessing COPD symptoms.