A Randomized, Double-blind, Placebo-controlled Crossover Pilot Study of the Effect of Metformin on Airway Glucose in COPD: The Metformin and Airway Glucose In COPD (MAGIC) Trial

Background and ObjectivePatients with COPD have elevated levels of airway glucose. This may increase the risk of bacterial infection leading to exacerbation and disease progression. We aimed to test whether treating COPD patients with metformin, an oral hypoglycaemic used in diabetes, reduced airway glucose. MethodsIn this randomized, double-blind, placebo-controlled crossover pilot study, we assigned patients with COPD to two 3-month periods of treatment with 1g metformin or placebo twice daily, followed by the alternate treatment after a 2-week washout period. Patients were required to be free of infection, antibiotic or oral steroid treatment in the 8 weeks prior to enrolment. The following were collected at baseline and monthly thereafter: sputum (lower airways sample), nasal/upper airway lining fluid samples using synthetic absorptive matrix (SAM), spirometry, and St Georges Respiratory Questionnaire (SGRQ) and COPD Assessment Test (CAT) scores. The primary outcome was a change in sputum glucose following 3 months treatment with metformin compared to 3 months of placebo. Results14 patients were randomised of whom 3 completed the study, mainly due to withdrawals for exacerbations (8/11 withdrawals). In the n=4 patients who completed 3 months metformin treatment, there were no significant changes in sputum or nasal glucose compared to baseline. Metformin did not affect COPD symptom scores or spirometry. ConclusionsTreatment with metformin in this underpowered pilot study did not significantly reduce airway glucose in COPD. Larger studies are required to definitively evaluate this. SUMMARY AT A GLANCEElevated airway glucose is associated with bacterial infection in chronic obstructive pulmonary disease (COPD) exacerbations. We tested whether metformin, a drug used to treat diabetes, reduced airway glucose in patients with COPD. Recruitment/retention difficulties led to an underpowered study, which observed no difference between metformin and placebo on airway glucose.