Abstract 12444: Outcomes After Starting Beta-Blockade Within a Clinically Sensible Period Before Major Elective Noncardiac Surgery

INTRODUCTION: Guidelines recommend that perioperative beta-blockade be started days to weeks before surgery. Nonetheless, all randomized trials except for the controversial DECREASE trials started treatment ≤1 day before surgery, while most observational studies did not distinguish between long-term beta-blockade versus beta-blockers started for perioperative reasons. We thus conducted a population-based cohort study of the effectiveness of beta-blockade started within a clinically sensible period (8-60 days) before surgery. METHODS: Following research ethics approval, we conducted a cohort study of patients (≥66 years) who underwent major elective noncardiac surgery from 2003 and 2012 in Ontario, Canada. Propensity-score methods were used to form a matched cohort that reduced important differences between patients who started beta-blockers 8-60 days before surgery versus controls (no beta-blockers within 1 year before surgery). We measured the association of beta-blockade with 30-day (death, MI, stroke) and 1-year (death) outcomes post-surgery. Subgroup analyses were performed based on Revised Cardiac Risk Index class and history of prior CAD. RESULTS: The cohort included 4268 beta-blocked patients and 154,357 controls. Metoprolol (median daily dose 50 mg) was prescribed to 36% of beta-blocked patients, atenolol (median 25 mg) to 26%, and bisoprolol (median 5 mg) to 37%. In the matched cohort (n=8492), beta-blockade was not associated with death (RR 0.96; CI 0.70-1.32), MI (RR 0.92; CI 0.72-1.17), and stroke (RR 1.31; CI 0.68-2.52) at 30-days, or death at 1-year (Figure). Associations with outcomes did not differ significantly across subgroups. CONCLUSIONS: Outcomes were not altered in patients who start perioperative beta-blockade within a clinically sensible period before surgery. A large randomized trial is needed to determine if the continued use of perioperative beta-blockade in clinical practice is justified.