Liraglutide to Improve corONary haemodynamics during Exercise streSS (LIONESS): a double-blind randomised placebo-controlled crossover trial

Abstract Background Glucagon-like peptide-1 receptor (GLP-1R) activation may improve myocardial performance in the context of ischaemia, independent of glycaemic control, in individuals with and without type 2 diabetes mellitus. Methods The LIONESS trial was a single-centre randomised double-blind placebo-controlled crossover study to determine whether prolonged GLP-1R activation could improve exercise haemodynamics in chronic stable angina patients. Eligibility criteria comprised angiographic evidence of obstructive coronary artery disease (CAD) and an abnormal baseline exercise tolerance test (ETT) demonstrating > 0.1 mV of planar or downsloping ST-segment depression (STD). Those randomised to active agent started with a 1-week run-in phase of 0.6 mg liraglutide daily, an established injectable GLP-1R agonist, followed by 1 week of 1.2 mg liraglutide, after which patients performed a week 2 ETT. Patients then self-administered 1.8 mg liraglutide for a week before completing a week 3 ETT. The placebo arm received visually and temporally matched daily saline injections. Participants then crossed over to a 3-week course of saline injections interspersed with a week 5 ETT and week 6 ETT and vice versa. Co-primary endpoints were rate pressure product (RPP) at 0.1 mV STD and magnitude of STD at peak exercise. Results Twenty-two patients (21 without diabetes) were randomised. There was no significant difference between saline versus liraglutide in the co-primary endpoints of RPP achieved at 0.1 mV STD (saline vs. liraglutide 1.2 mg p = 0.097; saline vs. liraglutide 1.8 mg p = 0.48) or the degree of STD at peak exercise (saline vs. liraglutide 1.2 mg p = 0.68; saline vs. liraglutide 1.8 mg p = 0.57). Liraglutide did not cause symptomatic hypoglycaemia, renal dysfunction, acute pancreatitis or provoke early withdrawal from the trial. Liraglutide significantly reduced weight (baseline 88.75 ± 16.5 kg vs. after liraglutide 87.78 ± 16.9 kg; p = 0.0008) and improved the lipid profile (mean total cholesterol: at baseline 3.97 ± 0.88 vs. after liraglutide 3.56 ± 0.71 mmol/L; p < 0.0001). Conclusion Liraglutide did not enhance exercise tolerance or haemodynamics compared with saline placebo during serial treadmill testing in patients with established obstructive CAD. It did, however, significantly reduce weight and improve the lipid profile. Trial Registration ClinicalTrials.gov Identifier NCT02315001. Retrospectively registered on 11th December 2014.