Continuous Glucose Monitoring and the Effect of Liraglutide in Cardiac Surgery Patients: A Substudy of the Randomized Controlled GLOBE Trial.

ABSTRACT

OBJECTIVES: The objectives of the current research were to evaluate the accuracy and reliability of continuous glucose monitoring (CGM) in patients undergoing cardiac surgery and assess the impact of preoperative liraglutide administration on perioperative glucose control as captured by CGM. DESIGN: This was a prospective, single-center, prespecified analysis of the GLOBE trial, a randomized controlled trial comparing preoperative liraglutide treatment to placebo in patients undergoing cardiac surgery. SETTING: The work took place at a single-center academic hospital in the Netherlands. PARTICIPANTS: Twenty-five patients undergoing cardiac surgery were recruited from the hospital's cardiac surgery department. INTERVENTIONS: Participants received the Dexcom G5 CGM system from the day before surgery until discharge from the intensive care unit after surgery. Additionally, participants were randomized to receive either preoperative liraglutide or placebo. MEASUREMENTS AND MAIN RESULTS: Arterial blood gas (ABG) glucose measurements were collected as a reference and matched to CGM readings to assess accuracy and reliability. In 240 paired CGM-ABG glucose measurements, the mean absolute relative difference was 14.4 ± 12.5%. Temporary sensor interruption occurred mainly intraoperatively (92% of patients). The median duration of intraoperative sensor interruption was 65 (48-95) minutes. Liraglutide increased glycemic time in range 72% versus 47% in the control group (absolute difference 25%, 95% confidence interval -41.4 to -8.9, p = .004). CONCLUSIONS: Despite intraoperative sensor interruption, CGM seems an accurate method for semi-invasive, real-time assessment of blood glucose levels. CGM can provide a detailed observation of the pre- and postoperative glycemic trajectory, demonstrating increased time in range following perioperative liraglutide treatment compared with placebo.