To Örebro University

Introduction: Chronic Obstructive Pulmonary Disease (COPD) is a heterogeneous disease with a variable mortality risk. A simple clinical algorithm has been validated for short-term mortality by Burgel et al. (ERJ 2017).

Aim: To study if Burgel’s clinical algorithm is valid to predict long-term mortality.

Methods: Data from two COPD cohorts, the Swedish PRAXIS Study (PS) (n=784, mean age (SD) 64.0 years (7.5), 42% males) and the Rotterdam Study (RS) (n=735, mean age (SD) 72 years (9.2), 57% males), with 9-years of follow-up data including mortality was used. The five clinical clusters were derived from baseline data on age, body mass index, dyspnea grade (mMRC), FEV 1 (%pred) and comorbidity (cardiovascular disease or diabetes). Mortality risk was estimated by unadjusted Cox models.

Results: The distribution of clinical clusters (1-5) was: 29%/45%/8%/6%/12% in PS and 23%/26%/36%/0/15% in RS. The cumulative proportion of deaths after 9-years of follow-up was highest among COPD clusters 1 (65%) and 4 (72%), and lowest among cluster 5 (10%) in the PS cohort. In RS, Cluster 1 (44%) had the highest cumulative mortality and cluster 5 (5%) the lowest. Compared to cluster 5, the meta-analysed hazard ratio (HR) (95%CI) for cluster 1 was 9.95 (6.52–15.19) and for cluster 4, 13.49 (6.41–28.38). The meta-analysed HR for clusters 2 and 3, compared with cluster 5, were: 2.80 (1.77 – 4.36) and 4.73 (3.02 – 7.42), respectively.

Conclusions: Burgel’s clinical clusters can be used to predict long-term mortality risk. Clusters 1 and 4 are associated with the poorest prognosis, cluster 5 with best prognosis and clusters 2 and 3 with intermediate prognosis in two independent COPD cohorts from Sweden and Netherlands.