Incident and recurrent hypoglycaemia with linagliptin and glimepiride over a median of 6 years in the CAROLINA cardiovascular outcome trial

Abstract

AbstractAimThe CAROLINA trial established non‐inferiority of linagliptin versus glimepiride for major adverse cardiovascular events in patients with relatively early type 2 diabetes at increased cardiovascular risk. In pre‐specified and post‐hoc analyses, we investigated treatment effects on total hypoglycaemic burden in CAROLINA.Materials and methodsPatients were randomized and treated with 5 mg linagliptin (n = 3014) or 1‐4 mg glimepiride (n = 3000) once daily added to standard care. Hypoglycaemia captured from investigator‐reported adverse events was analysed with Poisson and negative binomial regressions for the first and total (first plus recurrent) events, respectively. The influence of insulin initiation and glycated haemoglobin (HbA1c) change on the treatment effect for hypoglycaemia was also explored.ResultsOver 6.3 years median follow‐up, average HbA1c over time did not differ between linagliptin versus glimepiride (weighted mean difference [95% confidence interval]: 0.00%, [−0.05, 0.05]), nor did insulin initiation (18.6% vs. 19.2% of patients, respectively), whereas body weight was lower with linagliptin (−1.54 kg, [−1.80, −1.28]). Hypoglycaemia frequency was lower with linagliptin across all hypoglycaemia categories, including severe episodes. Rate ratios (95% confidence interval) for first and total events for investigator‐reported hypoglycaemia were 0.21 (0.19‐0.24) and 0.12 (0.10‐0.14), respectively, with 8.7 first and 60.8 total estimated events prevented/100 patient‐years with linagliptin versus glimepiride. These differences occurred during night‐time and daytime, and in subgroup analyses of total events. Treatment differences in hypoglycaemia were neither impacted by HbA1c changes nor insulin initiation.ConclusionsAcross the severity spectrum, linagliptin substantially reduced the hypoglycaemic burden versus glimepiride in patients with relatively early type 2 diabetes at increased cardiovascular risk.