Controlling intraoperative hyperglycemia in the cardiac surgical patient using the computerized electronic insulin dose calculator

Perioperative hyperglycemia is associated with increased morbidity and mortality. We report the findings of our quality improvement project on the use of an electronic insulin dosing calculator (EIC) to reduce intraoperative hyperglycemia in a cohort of cardiac surgical patients. A pilot and a modified EIC were sequentially implemented in adult patients undergoing cardiopulmonary bypass (CPB) procedures. Outcome measures (average blood glucose (BG) before, during and after CPB), process measures (percent EIC implementation) and balancing measures (incidence of hypoglycemia and average insulin doses utilized) were compared between baseline (no-protocol), pilot and modified EIC periods. 248, 142, and 239 patients were enrolled during the baseline, pilot and modified EIC phases, respectively. The EIC was implemented in 97% (138 out of 142 patients) during the pilot phase and in 86% of patients (206 out of 240 patients) during the modified EIC phase. Average BG levels on CPB were highest at baseline compared with pilot and modified EIC phases (mean, SD, 209 mg/dl,+/-45), vs. 189.8,+/-38 mg/dl vs. 188 mg/dl, +/-40) mg/dl, respectively, p = 0.002). Mean BG values were highest at baseline compared to pilot and modified EIC phases at end of procedure (182.4, +/-49 mg/dl, vs. 122, +/-30 mg/dl, vs.123.3, mg/dl, +/-38.3, vs., respectively, p = < 0.001) and at the ICU (164 ,+/-44.4, mg/dl, vs. 140, +/-37.9 mg/dl, vs. 143.4, +/-36 mg/dl, respectively p = < 0.001). Hypoglycemia was significantly lower during EIC phase compared with baseline (1% vs. 7%, p = 0.008). Less insulin was used during the pilot and modified EIC phases compared to baseline (15.2, +/-10.3, U vs. 15.8, +/-10.2 U, vs. 31.6 +/-20.27 U, respectively, p = 0.006). These preliminary findings suggest EIC effectiveness in reducing intraoperative hyperglycemia in patients undergoing CPB.