EATING AND GLYCEMIC CONTROL AMONG CRITICALLY ILL PATIENTS RECEIVING CONTINUOUS INTRAVENOUS INSULIN

Objective: Consensus guidelines recommend that intensive care unit (ICU) patients with blood glucose (BG) levels >180 mg/dL receive continuous intravenous insulin (CII). The effectiveness of CII at controlling BG levels among patients who are eating relative to those who are eating nothing by mouth (nil per os; NPO) has not been described. Methods: We conducted a retrospective cohort study of 260 adult patients (156 eating, 104 NPO) admitted to an ICU between January 1, 2014, and December 31, 2014, who received CII. Patients were excluded for a diagnosis of diabetic ketoacidosis or hyperglycemic hyperosmolar nonketotic syndrome, admission to an obstetrics service, or receiving continuous enteral or parenteral nutrition. Results: Among 22 baseline characteristics, the proportion of patients receiving glucocorticoid treatment (GCTx) (17.3% eating, 37.5% NPO; P<.001) and APACHE II score (15.0 +/- 7.5 eating, 17.9 +/- 7.9 NPO; P = .004) were significantly different between eating and NTO patients. There was no significant difference in the primary outcome of patient-day weighted mean BG overall (153 +/- 8 mg/dL eating, 156 +/- 7 mg/dL NPO; P = .73), or day-by-day BG (P = .37) adjusted for GCTx and APACHE score. Surprisingly, there was a significant difference in the distribution of BG values, with eating patients having a higher percentage of BG readings in the recommended range of 140 to 180 mg/dL. However, eating patients showed greater glucose variability (coefficient of variation 23.1 +/- 1.0 eating, 21.2 +/- 1.0 NPO; P = .034). Conclusion: Eating may not adversely affect BG levels of ICU patients receiving CII. Whether or not prandial insulin improves glycemic control in this setting should be studied.