Outcomes of a cardiothoracic intensive care web-based online intravenous insulin infusion calculator study at a Medical University Hospital.

BACKGROUND The purpose of this study was to examine whether a web-based, on-line intravenous insulin (IVI) infusion calculator (IVIIC) program for the delivery of IVI therapy in coronary artery bypass graft (CABG) patients was superior to a prior IVI protocol used in the cardiothoracic intensive care unit at our institution. METHODS The study included 97 CABG patients studied from October 2004 to February 2005 pre-protocol (type 2 diabetes) and October 2005 to February 2006 post-protocol (with or without type 2 diabetes). The IVIIC was programmed to resemble an algorithm whereby any patient with type 2 diabetes or a blood glucose (BG) greater than 120 mg/dL was started on IVI with an insulin sensitivity factor, a multiplier of 0.03. The calculator used the following mathematical formula: rate of insulin infusion/hour = (current BG - 60 mg/dL) x 0.03. RESULTS Pre- and post-protocol groups for patients with type 2 diabetes were similar in all demographics measured, including initial age, mean age, percentage female, and percentage African-American. Significant differences were observed between pre- and post-protocol groups in mean BG recorded over a 48-h period (P < 0.0001), percentage not at target within 48 h (P < 0.0001), mean hours to first BG between 80 to 120 mg/dL (P < 0.0001), mean hours to target (three consecutive BGs 80-120 mg/dL) (P < 0.0001), and hyperglycemic index (P < 0.0001). The incidence of hypoglycemia (percentage BG < 70 mg/dL) was not significantly increased in the post-protocol groups (P = 0.2581). CONCLUSIONS We conclude that the IVIIC is a safe nurse-driven protocol with excellent BG outcomes.