Evaluation of an updated insulin infusion protocol at a large academic medical center

Purpose. The performance of an updated insulin infusion protocol was evaluated at a large, academic medical center. Methods. A retrospective medical record review was performed after a one-month run-in period for all patients at a large, academic, tertiary care medical center in whom the insulin infusion per protocol was administered from January 1 through February 28, 2014. Data were evaluated to determine the median blood glucose (BG) level, time to achieve BG in the target range, number of BG checks per patient per day, time elapsed between each BG check, and the frequency of hypoglycemia (BG concentration of <= 70 mg/dL). Results. A total of 170 patients were included. The median preinfusion BG was 244 mg/dL (interquartile range [IQR], 204-304 mg/dL), which decreased to a median of 168 mg/dL (IQR, 147.5-199.5 mg/dL) when the protocol was utilized. However, 70 patients (41%) had a median BG concentration of >= 180 mg/dL, and 25 patients' (15%) BG value remained above 180 mg/dL. The median time to achieve the goal BG value was 4.2 hours (95% confidence interval, 3.2-5.1 hours). BG checks were performed a median of every 2.1 hours (IQR, 1.4-2.3 hours). Hypoglycemia was rare, occurring in only 2 (1.2%) patients. Conclusion. The median BG with an updated insulin infusion protocol approached the upper limit of the target BG range, and 41% of patients had a median BG above the goal range. Protocol specifications for the frequency of BG monitoring were not commonly followed, but the frequency of hypoglycemia was extremely low.