The influence of the exacerbator phenotype on mortality in chronic obstructive pulmonary disease (COPD)

Abstract

Abstract Background. Frequent exacerbations in chronic obstructive pulmonary disease (COPD) accelerate lung function decline, yet whether the ‘frequent-exacerbator’ phenotype, defined as two or more treated exacerbations in the previous year, independently increases mortality remains uncertain. This study aimed to investigate the association between the exacerbator phenotype and all-cause mortality. Methods. This retrospective cohort study analysed 3,926 spirometry-confirmed COPD patients from the multinational COPD Cohorts Collaborative International Assessment (3CIA). Exposure was defined as ≥ 2 hospital-treated exacerbations in the previous year. A total of 973 participants (24.8%) met this criterion, giving an approximately 1:3 exposure-to-control ratio. Multivariable logistic and Cox proportional hazards models estimated adjusted odds ratios (OR) and hazard ratios (HR) for death, controlling for age, sex, body mass index (BMI), smoking status, FEV₁% predicted and dyspnoea severity (mMRC). Exacerbation frequency was further analysed as a time-updated continuous variable. Results. Frequent exacerbators had lower mean FEV₁ % predicted (46.7% vs 52.9%; p < 0.0001), higher median mMRC score (2 vs 1; p < 0.0001) and shorter follow‑up (44 vs 52 months; p < 0.0001). Kaplan-Meier analysis showed worse unadjusted survival among frequent exacerbators (median 111 vs 137 months; log-rank p < 0.0001). However, after adjusting for confounders, the frequent exacerbator phenotype was not independently associated with increased mortality (OR 0.95, 95% CI 0.80 to 1.13, p = 0.56; HR 1.10, 95% CI 0.96 to 1.25, p = 0.18). Each additional exacerbation was associated with a 6% increase in mortality hazard (HR 1.06, 95% CI 1.02–1.09, p = 0.004). Independent predictors of mortality included advancing age (HR 1.06 per year, 95% CI 1.05–1.07; p < 0.0001), male sex (HR 1.41, 95% CI 1.21–1.65; p < 0.0001), higher dyspnoea severity (HR 1.30 per mMRC point, 95% CI 1.23–1.37; p < 0.0001), lower lung function (HR 0.98 per 1% increase in FEV₁% predicted, 95% CI 0.97–0.98; p < 0.0001), and lower BMI (HR 0.98 per 1 kg/m² increase, 95% CI 0.97–0.99; p = 0.001). The odds of being classified as a frequent exacerbator increased with disease severity (GOLD 4 vs GOLD 1: OR 1.54, 95% CI 1.05 to 2.25; P = 0.027). Conclusion. The exacerbator phenotype reflects more advanced disease but does not independently predict mortality after adjustment for confounders. Future research should validate a dynamic two-year exacerbator definition and integrate biomarkers, such as blood eosinophils, into risk models. Exacerbation history should prompt a comprehensive reassessment and optimisation of treatable traits rather than be viewed as a stand-alone prognostic factor.