Introduction and objectives Fluticasone propionate/salmeterol (FP/SAL), can be delivered by metered-dose inhaler (MDI) or dry powder inhaler (DPI). The choice of device may affect adherence to and effectiveness of treatment. Although only 1000 mcg/day DPI is licensed for the treatment of COPD in the UK, the MDI and lower doses are regularly used in real-world practice. The aim of this study was to compare the effectiveness and safety of FP/SAL MDI or DPI at two doses (500 and 1000 mcg/day) in COPD patients.

Methods Historical, matched cohort study using the Optimum Patient Care Research Database in patients with COPD, aged ≥35 years and initiating with FP/SAL via either MDI or DPI. Conditional Poisson regression and conditional logistic regression were used respectively to compare the rate of moderate/severe COPD exacerbations and the odds of diagnosis of pneumonia and diabetes mellitus (including anti-diabetic drug prescriptions) between MDI and DPI during one year outcome period. Models were adjusted for the respective baseline values of the outcome variable of interest where possible. Addition of LAMA therapy during the outcome period was compared using conditional logistic regression.

Results 472 and 1172 patients initiated on FP/SAL at 500 mcg/day and 1000 mcg/day, respectively. The rate of moderate/severe COPD exacerbations was significantly lower for patients prescribed MDI versus DPI initiating with 500 mcg/day of FP/SAL; no significant difference was observed between MDI and DPI for those initiating at 1000 mcg/day. There were no significant differences in the odds of diagnosis of diabetes mellitus or pneumonia between MDI and DPI, irrespective of the initiation dose of FP/SAL (Table 1). LAMA prescription during the outcome period was significantly lower for patients prescribed MDI versus DPI initiating at 1000 mcg/day; no significant difference was observed between MDI and DPI for those initiating at 500 mcg/day.

Conclusions This study showed greater reduction in exacerbations for patients using MDI than those using DPI when initiating with FP/SAL 500 mcg/day; no differences in exacerbation reduction and safety were seen for FP/SAL at 1000 mcg/day. Fewer patients using MDIs than DPIs at 1000 mcg/day were prescribed LAMAs, suggesting less need for treatment intensification.