Effect of beta blocker use and type on hypoglycemia risk among hospitalized insulin requiring patients

Background Although beta blockers could increase the risk of hypoglycemia, the difference between subtypes on hypoglycemia and mortality have not been studied. This study sought to determine the relationship between type of beta blocker and incidence of hypoglycemia and mortality in hospitalized patients. Methods We retrospectively identified non-critically ill hospitalized insulin requiring patients who were undergoing bedside glucose monitoring and received either carvedilol or a selective beta blocker (metoprolol or atenolol). Patients receiving other beta blockers were excluded. Hypoglycemia was defined as any glucose < 3.9 mmol/L within 24 h of admission (Hypo(1day)) or throughout hospitalization (Hypo(T)) and any glucose < 2.2 mmol/L throughout hospitalization (Hypo(severe)). Results There were 1020 patients on carvedilol, 886 on selective beta blockers, and 10,216 on no beta blocker at admission. After controlling for other variables, the odds of Hypo(1day,) Hypo(T) and Hypo(severe) were higher for carvedilol and selective beta blocker recipients than non-recipients, but only in basal insulin nonusers. The odds of Hypo(1day) (odds ratio [OR] 1.99, 95% confidence interval [CI] 1.28, 3.09, p = 0.0002) and Hypo(T) (OR 1.38, 95% CI 1.02, 1.86, p = 0.03) but not Hypo(severe) (OR 1.90, 95% CI 0.90, 4.02, p = 0.09) were greater for selective beta blocker vs. carvedilol recipients in basal insulin nonusers. Hypo(1day), Hypo(T), and Hypo(severe) were all associated with increased mortality in adjusted models among non-beta blocker and selective beta blocker recipients, but not among carvedilol recipients. Conclusions Beta blocker use is associated with increased odds of hypoglycemia among hospitalized patients not requiring basal insulin, and odds are greater for selective beta blockers than for carvedilol. The odds of hypoglycemia-associated mortality are increased with selective beta blocker use or nonusers but not in carvedilol users, warranting further study.