Comparative analysis of glycemic outcomes following implementation of a pharmacist‐led insulin management service in postoperative cardiothoracic surgery patients

Abstract Introduction Perioperative hyperglycemia is a predictor of mortality in cardiothoracic surgery (CTS) patients. Cardiothoracic surgery patients with diabetes have improved clinical outcomes with maintained glycemic management, including improved hospital length of stay (LOS), surgical site infection risk, and rates of morbidity and mortality. A pharmacist‐led insulin management service was implemented, granting pharmacists autonomy to adjust basal, bolus, and infusion insulin regimens to optimize glycemic management following CTS. Literature evaluating the impact on glycemic management in CTS patients following a pharmacist‐led insulin management service is limited. Objective The study objective was to compare postoperative glycemic management in CTS patients before and after implementation of a pharmacist‐led insulin management service. Methods This retrospective study included adult CTS patients with diabetes with a hemoglobin A1c (HbA1c) of 6.5% or greater who received at least 2 basal insulin doses following postoperative intravenous (IV) insulin infusion. The primary outcome was the percentage of blood glucose concentrations within the target range (70–180 mg/dL) until postoperative Day 7 or discharge, if sooner. Secondary outcomes were average daily blood glucose, time to discontinuation of postoperative IV insulin infusion, incidence of hypoglycemia and hyperglycemia requiring re‐initiation of an insulin infusion, hospital LOS, and postoperative surgical site infections within 3 months. Results A total of 200 patients were included, 100 in each group. Mean percent of glucose checks within target range was higher in the post‐implementation group compared with the pre‐implementation group (60.8 vs. 48.9, p = 0.0004). The post‐implementation group achieved a lower mean daily blood glucose (170 vs. 191, p < 0.0001) and spent fewer days on postoperative insulin infusion (1.24 vs. 1.85, p < 0.0001). There were no differences in the incidence of hypoglycemia, re‐initiation of insulin infusion, LOS, or surgical site infections. Conclusion Implementation of a pharmacist‐led insulin management service was a safe and effective strategy to improve glycemic management in postoperative CTS patients.