Background: Intensive insulin treatment is associated with an increased risk of hypoglycemia, so strict glycemic monitoring is essential. The best type of sample for identifying hypoglycemia remains under debate.
Objectives: To establish the number of hypoglycemic events in intensive care patients relative to insulin administration method and the method used to collect the blood sample.
Methods: Retrospective descriptive study lasting 6 months. Hypoglycemia was defined as a blood glucose level less than 80 mg/dL (mild: 50-79 mg/dL, severe: <50 mg/dL), measured with a bedside glucometer and blood from the arterial catheter or fingerstick, in critically ill patients who require insulin administered subcutaneously (with sliding scales) or via continuous intravenous perfusion (intense infusion protocol with a nurse-managed insulin therapy algorithm).
Results: Analysis of the 6636 samples from 144 critically ill patients revealed 188 mildly hypoglycemic samples (2.8%) and 3 severely hypoglycemic samples (0.04%). The prevalence of mild hypoglycemia was greater when insulin was administered intravenously (3.2%) rather than subcutaneously (2.3%; P = .04). Among patients receiving insulin intravenously, hypoglycemia was found more often in arterial (4.5%) than in capillary (2.8%) blood (P = .01). The prevalence of hypoglycemia in capillary blood samples did not differ significantly between subcutaneous (2.3%) and intravenous (2.8%) insulin therapies (P = .21).
Conclusions: With a target blood glucose level of 110 to 140 mg/dL, few hypoglycemic events are detected in critically ill patients, regardless of whether insulin is administered intravenously or subcutaneously. Analysis of solely arterial samples may yield a higher prevalence of hypoglycemia than otherwise.