Evaluation of glucose variability when converting from insulin infusion to basal-bolus regimen in a surgical-trauma intensive care unit

Abstract Purpose This study aimed to identify predictive factors resulting in glucose values greater than 200 mg/dL in patients with trauma transitioned from an insulin infusion to a basal-bolus subcutaneous insulin regimen. Materials and Methods Thirty-nine patients with trauma on goal enteral nutrition in the intensive care unit receiving an insulin infusion for at least 48 hours and transitioned to a basal-bolus regimen were retrospectively identified. Results Ten patients had hyperglycemic events after transition. Hyperglycemia was significantly associated with increased age (42 [17] years vs 56 [13] years, P = .02), admission glucose (128 [39] mg/dL vs 214 [91] mg/dL, P = .015), and insulin drip rate 48 hours before transition (87 [38] units/d vs 127 [49] units/d, P = .012). There was no difference between groups with respect to injury severity, demographics, or physiologic parameters. Multiple regression analysis revealed that increased age (odds ratio [OR], 1.215 [1.000-1.477]; P = .05), increased admission blood glucose (OR, 1.053 [1.006-1.101]; P = .025), and higher insulin infusion rates 48 hours before transition (OR, 1.061 [1.009-1.116]; P = .020) predisposed patients to severe hyperglycemic episodes. Conclusions Older patients with trauma and patients with higher blood glucose on admission are more likely to experience severe hyperglycemia when transitioned to basal-bolus glucose control. Higher insulin infusion rates at 48 hours before transition are also associated with severe hyperglycemia.