Factors Associated with Inhaled Corticosteroids (ICS) and Long-Term Oxygen Therapy (LTOT) in Chronic Obstructive Pulmonary Disease (COPD)

Abstract

Abstract Background: Chronic obstructive pulmonary disease (COPD) affects millions globally, with inhaled corticosteroids (ICS) and long-term oxygen therapy (LTOT) representing cornerstone interventions. However, their prescription patterns and impact on survival in large real-world cohorts remain underexplored, with significant gaps between clinical trial evidence and routine clinical practice.

Methods: This retrospective secondary analysis examined data from the COPD Cohorts Collaborative International Assessment (3CIA), comprising 17,843 COPD patients from 22 international cohorts across seven countries (2008-2014). Three objectives were addressed: describing baseline characteristics stratified by LTOT and ICS use; investigating associations between treatments and GOLD stage classification using ordinal logistic regression; and assessing treatment associations with all-cause mortality using binary logistic regression. Complete case analysis was employed, with statistical significance set at p<0.05.

Results: LTOT data were available for 3,987 patients (22.4%) and ICS data for 8,987 patients (50.4%). Both treatment groups demonstrated significantly higher baseline mortality rates (LTOT: 33.4% vs 11.2%, p<0.001; ICS: 21.9% vs 9.5%, p<0.001) and distinct clinical profiles, with former smokers predominating in treated groups. Paradoxically, both treatments were associated with decreased odds of higher GOLD stages (LTOT: OR=0.283, p<0.001; ICS: OR=0.392, p<0.001). In multivariable mortality analysis, neither LTOT (OR=0.725, p=0.398) nor ICS (OR=0.925, p=0.799) showed significant associations with all-cause mortality. The Charlson comorbidity index emerged as the strongest mortality predictor (OR=1.355 per point increase, p<0.001), while current smoking nearly tripled mortality risk (OR=2.442, p=0.016).

Conclusions: Despite higher baseline mortality in treatment users, neither ICS nor LTOT demonstrated mortality benefits in adjusted analyses, reflecting confounding by indication where treatment prescription serves as a disease severity marker. Comorbidity burden and smoking status were stronger mortality predictors than respiratory treatments. These findings emphasize the importance of comprehensive patient assessment, appropriate treatment selection, and smoking cessation in COPD management.