Beta‐blocker‐associated hypoglycaemia: New insights from a real‐world pharmacovigilance study
Carla Carnovale, Michele Gringeri, Vera Battini, Giulia Mosini, Elena Invernizzi, Faizan Mazhar, Francesco Bergamaschi, Mara Fumagalli, Gianvincenzo Zuccotti, Emilio Clementi, Sonia Radice, Valentina Fabiano
Abstract
AIMS: To investigate the statistical association between hypoglycaemia and β-blocker use and to define what patient and drug characteristics could potentially increase the risk for its occurrence. METHODS: We investigated the relationship between pharmacological parameters of β-blockers and the occurrence of hypoglycaemia by conducting a case/non case analysis using the Food and Drug Administration Adverse Event Reporting System database. Pharmacological properties that could represent a predictive factor for hypoglycaemia were analysed through a multilinear binary logistic regression (null hypothesis rejected for values of P < .05). We also performed a systematic review of clinical studies on this association. RESULTS: Of 83 954 selected reports, 1465 cases (1.75%) of hypoglycaemia were identified. The association was found statistically significant for nadolol (reporting odds ratio [95% confidence interval]: 6.98 [5.40-9.03]), celiprolol (2.35 [1.35-4.10]), propranolol (2.14 [1.87-2.46]) and bisoprolol (1.42 [1.25-1.61]). Paediatric cases (n = 310) showed a positive association with hypoglycaemia for long half-life drugs (odds ratio [95% confidence interval]: 2.232 [1.398-3.563]) and a negative association for β1-selectivity (0.644 [0.414-0.999]). Seven papers were included in the systematic review. Because of great heterogeneity in study design and demographics, hypoglycaemia incidence rates varied greatly among studies, occurring in 1.73% of the cases for propranolol treatment (n total participants = 575), 6.6% for atenolol (n = 30) and 10% for carvedilol (n = 20). CONCLUSION: Nadolol appears to be the β-blocker significantly most associated with hypoglycaemia and children represent the most susceptible sample. Furthermore, long half-life and nonselective β-blockers seem to increase the risk for its occurrence.