AimsPrevious large-scale, retrospective studies have shown increased mortality in heart failure (HF) patients using b2-agonists (B2As). We further examined the relationship between B2A use and mortality in a well-characterized population by adjusting for natriuretic peptide levels as a measure of HF severity. Methods and resultsThis was a retrospective cohort study of patients attending an HF Disease Management Programme with mean follow-up of 2.9 + 2.4 years. Chart review confirmed B2A use, dose and duration of use, and documented pulmonary function evaluation. The primary endpoint was the effect of B2A use compared with no B2A use on mortality using unadjusted and adjusted Kaplan2Meier survival curves. Data were available for 1294 patients (age 70.6 + 11.5 years) of whom 64% were male and 22.2% were taking B2As. b2-Agonist users were older, more likely to be male, to have smoked, to have chronic obstructive pulmonary disease (COPD) and asthma, and less likely to take beta-blockers. Multivariable associates of mortality included: B-type natriuretic peptide (BNP), coronary artery disease, age, and beta-blocker use. Unadjusted mortality rates for B2A users were found to be significantly higher than non-B2A users [hazard ratio (HR) 1.304, 95% confidence interval (CI) 1.030-1.652, P ¼ 0.028]. However, when adjusted for age, sex, medication, co-morbidity, smoking, COPD, and BNP differences, overall mortality rates were similar [HR 1.043, 95% CI (0.771 -1.412), P ¼ 0.783]. ConclusionUnlike previous reports, this retrospective evaluation of B2A therapy in HF patients shows no relationship with longterm mortality when adjusted for population differences including BNP. Large, prospective studies are required to define the risk/benefit ratio of B2As in patients with heart failure.--