Original Article
Persistent hyperglycemia in critically ill children

https://doi.org/10.1016/j.jpeds.2004.08.076Get rights and content

Objectives

To determine the prevalence and prognostic significance of hyperglycemia among critically ill nondiabetic children.

Study design

We performed a retrospective cohort study using point-of-care blood glucose measurements, hospital administrative databases, and a computerized information system; 942 nondiabetic patients admitted to our Pediatric Intensive Care Unit (PICU) from October 2000 to September 2003 were included. The prevalence of hyperglycemia was based on initial PICU glucose measurement, highest value within 24 hours, and highest value measured during PICU stay up to 10 days after the first measurement. Primary outcome was in-hospital death with PICU lengths of stay (LOS) as secondary outcome.

Results

Through the use of three cutoff values (120 mg/dL, 150 mg/dL, and 200 mg/dL), the prevalence of hyperglycemia was 16.7% to 75.0%. The relative risk (RR) for dying increased for maximum glucose within 24 hours >150 mg/dL (RR, 2.50; 95% confidence interval (CI), 1.26 to 4.93) and highest glucose within 10 days >120 mg/dL (RR, 5.68; 95% CI, 1.38 to 23.47). LOS was decreased for admission glucose >120 mg/dL and 150 mg/dL but increased for all threshold values for maximum glucose within 10 days.

Conclusions

Hyperglycemia occurs frequently among critically ill nondiabetic children and is correlated with a greater in-hospital mortality rate and longer LOS.

Section snippets

Methods

We conducted a retrospective cohort study by using data collected in the normal course of patient care. The study was reviewed and approved by our institution's Human Investigation Committee, which waived the need to obtain informed consent. The study examined patients cared for in the 11-bed multidisciplinary Pediatric Intensive Care Unit (PICU) within a not-for-profit private academic teaching hospital.

Results

During the study period, 1927 patients were admitted to the PICU; 44 patients had a diagnosis of diabetes mellitus. Of the patients without diabetes mellitus, 942 had at least one glucose measurement. There was a median of 2 glucose measurements per patient (IQR, 1 to 3 measurements) for the first 24 hours and a median of 1 glucose measurement per patient per day (IQR, 0.7 to 2 measurements) for the next 10 days after the first determination. Twenty-eight study patients had incomplete

Discussion

The finding that maximum glucose values on average exceed the initial glucose values suggests that estimates of the prevalence of hyperglycemia based on initial or admission glucose values may underestimate the likelihood that patients become hyperglycemic. In a study on the prevalence of hyperglycemia in critically ill children, Ruiz Magro et al17 reported that 50% of 353 patients without diabetes mellitus had initial glucose values >120 mg/dL.

The prevalence of hyperglycemia reported in this

References (23)

  • G.E. Umpierrez et al.

    Hyperglycemia: an independent marker of in-hospital mortality in patients with undiagnosed diabetes

    J Clin Endocrinol Metab

    (2002)
  • Cited by (266)

    • What's new in pediatric critical care?

      2024, Best Practice and Research: Clinical Anaesthesiology
    • Assisted Enteral Feeding of Exotic Companion Animals

      2024, Veterinary Clinics of North America - Exotic Animal Practice
    View all citing articles on Scopus
    View full text