Lower rates of cardiovascular events and mortality associated with liraglutide use in patients treated with basal insulin: A DEVOTE subanalysis (DEVOTE 10)

Abstract

Aim: To compare the associations between concomitant liraglutide use versus no liraglutide use and the risk of major adverse cardiovascular events (MACE) and all-cause mortality among patients receiving basal insulin (either insulin degludec [degludec] or insulin glargine 100 units/mL [glargine U100]) in the Trial Comparing Cardiovascular Safety of Insulin Degludec versus Insulin Glargine in Patients with Type 2 Diabetes at High Risk of Cardiovascular Events (DEVOTE).

Materials and methods: Patients with type 2 diabetes and high cardiovascular risk were randomized 1:1 to degludec or glargine U100. Hazard ratios for MACE/mortality were calculated using a Cox regression model adjusted for treatment and time-varying liraglutide use at any time during the trial, without interaction. Sensitivity analyses were adjusted for baseline covariates including, but not limited to, age, sex, smoking and prior cardiovascular disease.

Results: At baseline, 436/7637 (5.7%) patients were treated with liraglutide; after baseline, 187/7637 (2.4%) started and 210/7637 (2.7%) stopped liraglutide. Mean liraglutide exposure from randomization was 530.2 days. Liraglutide use versus no liraglutide use was associated with significantly lower hazard rates for MACE [0.62 (0.41; 0.92)_95%CI ] and all-cause mortality [0.50 (0.29; 0.88)_95%CI ]. There was no significant difference in the rate of severe hypoglycaemia with versus without liraglutide use. Multiple sensitivity analyses yielded similar results.

Conclusions: Use of liraglutide was associated with significantly lower risk of MACE and death in patients with type 2 diabetes and high cardiovascular risk using basal insulin.

Keywords: cardiovascular disease; hypoglycaemia; insulin therapy; liraglutide; randomized trial; type 2 diabetes.

Figure 1

Analyses of major clinical outcomes by time‐varying liraglutide use. Full analysis set (all randomized patients). HRs presented are for time to the first confirmed event (in days), comparing concomitant liraglutide use with no concomitant liraglutide use. HRs are based on a Cox regression model adjusted for treatment and time‐varying liraglutide use at any time during the trial, without interaction. Thus, the analyses are adjusted for patients initiating, interrupting or discontinuing liraglutide treatment during the trial. For one patient who experienced an event occurring on the same day as liraglutide initiation, half a day was added to the day of the event. CI, confidence interval; HR, hazard ratio; MACE, major adverse cardiovascular events; N, number of events; Rate, events per 100 patient‐years of observation

Figure 2

Adjusted analyses of major clinical outcomes by time‐varying liraglutide use. Full analysis set (all randomized patients). HRs presented are for time to the first confirmed event (in days), comparing concomitant liraglutide use with no concomitant liraglutide use. HRs are based on a Cox regression model adjusted for treatment and time‐varying liraglutide use at any time during the trial alongside additional baseline factors and covariates, including age, sex, smoking status, race, diabetes duration, cardiovascular risk group, insulin treatment, HbA1c, BMI, systolic blood pressure, LDL cholesterol, HDL cholesterol, hepatic impairment category and renal impairment category, all without interaction. Thus, the analyses are adjusted for patients initiating, interrupting or discontinuing liraglutide treatment during the trial. For one patient who experienced an event occurring on the same day as liraglutide initiation, half a day was added to the day of the event. CI, confidence interval; HR, hazard ratio; MACE, major adverse cardiovascular events; N, number of events; Rate, events per 100 patient‐years of observation, BMI, body mass index; LDL, low‐density lipoprotein; HDL, high‐density lipoprotein