Identification of five clusters of comorbidities in a longitudinal Japanese chronic obstructive pulmonary disease cohort

Abstract

Background and objectives: Patients with chronic obstructive pulmonary disease (COPD) frequently suffer from various comorbidities. Recently, cluster analysis has been proposed to examine the phenotypic heterogeneity in COPD. In order to comprehensively understand the comorbidities of COPD in Japan, we conducted multicenter, longitudinal cohort study, called the Keio COPD Comorbidity Research (K-CCR). In this cohort, comorbid diagnoses were established by both objective examination and review of clinical records, in addition to self-report. We aimed to investigate the clustering of nineteen clinically relevant comorbidities and the meaningful outcomes of the clusters over a two-year follow-up period.

Material and methods: The present study analyzed data from COPD patients whose data of comorbidities were completed (n = 311). Cluster analysis was performed using Ward's minimum-variance method.

Results: Five comorbidity clusters were identified: less comorbidity; malignancy; metabolic and cardiovascular; gastroesophageal reflux disease (GERD) and psychological; and underweight and anemic. FEV1 did not differ among the clusters. GERD and psychological cluster had worse COPD assessment test (CAT) and Saint George's respiratory questionnaire (SGRQ) at baseline compared to the other clusters (CAT: p = 0.0003 and SGRQ: p = 0.00046). The rate of change in these scores did not differ within 2 years. The underweight and anemic cluster included subjects with lower baseline ratio of predicted diffusing capacity (DLco/VA) compared to the malignancy cluster (p = 0.036).

Conclusions: Five clusters of comorbidities were identified in Japanese COPD patients. The clinical characteristics and health-related quality of life were different among these clusters during a follow-up of two years.

Keywords: Chronic obstructive pulmonary disease; Cluster analysis; Comorbidity; Phenotype; Quality of life.