Perioperative hyperglycemia in cardiac surgery is associated with wound infection, acute kidney injury, and death, yet the intravenous insulin used to treat it is itself coupled to hypoglycemia. Glucagon-like peptide-1 receptor agonists lower glucose only when glucose is elevated and might therefore separate glycemic control from hypoglycemic risk. We searched PubMed/MEDLINE, Scopus, Web of Science, and CENTRAL through September 2025 for randomized controlled trials of perioperative subcutaneous liraglutide versus placebo or insulin-based usual care in adults undergoing cardiac surgery. Two reviewers screened, extracted data, and applied the Cochrane Risk of Bias 2 tool independently; certainty was graded using GRADE. Effects were pooled by inverse variance, fixed-effect when I2 was below 50%. Seven reports of 4 trials, enrolling 446 randomized patients, were eligible. Liraglutide did not alter 30-day mortality (1/161 vs 3/160; risk ratio [RR] 0.42, 95% confidence interval [CI], 0.06-2.81), the composite of any postoperative complication (68/129 vs 76/132; RR 0.92; 95% CI, 0.74-1.14), cardiac adverse events (RR 1.08; 95% CI, 0.83-1.40), hypoglycemia (8/165 vs 9/166; RR 0.85; 95% CI, 0.34-2.13), or postoperative nausea and vomiting (RR 3.01; 95% CI, 0.26-35.27).
