Relationship between hypoglycemia and mortality in critically ill children

Objectives: To determine the prevalence of hypoglycemia in critically ill nondiabetic children and the association of hypoglycemia with mortality and worsening organ function in critically ill children. Design: Retrospective cohort study with matched-cohort analysis. Setting: Academic pediatric intensive care unit. Patients: A total of 899 nondiabetic patients <18 yrs old admitted to the pediatric intensive care unit for >1 day with at least one blood glucose measurement. Forty-two patients with a blood glucose level of <50 mg/dL (<2.8 mmol/L) were matched with 126 nonhypoglycemic patients. Interventions: None. Measurements and Main Results: Hypoglycemia, based on point-of-care blood glucose measurements, occurred in 2.2% (<40 mg/dL [<2.2 mmol/L]) to 7.5% (<60 mg/dL [<3.3 mmol/L]) of the patients. Hypoglycemia was more common in patients on mechanical ventilation and/or vasopressor support. Severity of hypoglycemia correlated with an increased mortality rate. The highest odds ratio of mortality was 4.49 (95% confidence interval [CI], 1.69-11.96; p < .01) at a blood glucose level of < 40 mg/dL (<2.2 mmol/L). In the matched analysis, hypoglycemia was an independent risk factor for mortality. The unadjusted, covariate-adjusted, and propensity score-adjusted odds ratios of mortality were 3.69 (95% CI, 1.78-7.68; p < .01), 4.16 (95% CI, 1.53-11.32; p < .01), and 8.45 (95% CI, 1.75-40.86; p < .01), respectively. Hypoglycemia was associated with worsening organ function in the covariate-adjusted model (odds ratio, 2.37; 95% CI, 1.12-5.01; p = .02) but not in the unadjusted and propensity-score adjusted models. Conclusions: Hypoglycemia is common in critically ill children. It is associated with increased mortality rates in critically ill nondiabetic children. Our data suggest that hypoglycemia is also associated with worsening organ function. Hypoglycemia may merely be a marker of severity of illness. Further investigations are needed to establish the mortality risk with hypoglycemia due to insulin compared to spontaneous hypoglycemia. (Pediatr Crit Care Med 2010; 11: 690-698)