Specific causes of mortality in patients with chronic obstructive pulmonary disease – a prospective multicenter study.

<bold>Introduction:</bold> While factors associated with all-cause mortality in patients with chronic obstructive pulmonary disease (COPD) have been studied extensively, scientific data on causes of specific mortality is limited. <bold>Aim:</bold> Our aim was to analyze patterns of specific causes of mortality in patients with COPD in relation to clinical phenotypes. <bold>Methods:</bold> Prospective data of 781 patients from the Czech Multicenter Research Database of COPD (CMRD) was used. Survival analysis was performed using cumulative incidence, the Kaplan-Meier (KM) estimates and Cox model of proportional hazards. Survival estimates, median survival and hazard ratios (HR) were supplemented with 95% confidence intervals (CI). <bold>Results:</bold> Median age at inclusion was 66.6 years and FEV1 46.1% of predicted value. Main causes of mortality (n=461 deaths) included respiratory (46.6%), cardiovascular (22.8%) and cancer diseases (20.4%). Factors associated with increased risk of mortality included higher age, male sex, lower FEV1 and TLCO, higher BODE score and presence of emphysematous, frequent excacerbator or cachectic phenotype (all p˂0.05). Respiratory cause of death was associated with a presence of emphysematous, frequent excacerbator or cachectic phenotype (HR 1.4-1.76; all p˂0.05), while cardiovascular cause of death was associated with overlap of emphysematous and frequent excacerbator phenotypes (p=0.038). ACO appeared to be a protective factor of cardiovascular mortality (HR 0; p˂0.001). Finally, frequent exacerbators were less likely to die of a malignancy (HR 0.59; p=0.04). <bold>Conclusion:</bold> We concluded that specific causes of death differed with presence of certain phenotypes in the studied COPD population.