To elucidate risk factors for hyperglycemia in medical intensive care unit (MICU) patients without prior history or laboratory evidence of impaired glucose handling.
Convenience sample of MICU admissions at a tertiary care center with glycosylated hemoglobin (HbA1c) less than 6.5%, length of stay anticipated to be at least 48 hours, APACHE II score at least 12, and no history of diabetes. Data collection included all blood glucose measurements within 120 hours of MICU admission, as well as times and quantities of norepinephrine, corticosteroids, and carbohydrates during the same interval.
51 patients were studied. APACHE II was 22±7 (mean±sd), HbA1c was 5.4±0.6%, initial glucose at MICU admission was 144±105 mg/dl, peak glucose was 240±126 mg/dl, and average glucose was 154±51 mg/dl. Glucose exceeded 110 mg/dl during 74±31% of each patient’s first 120 hours of intensive care; only 2 patients never exceeded 110 mg/dl. HbA1c was significantly correlated with the fraction of time blood glucose was abnormally high (P<0.01, R2=0.15). Multiple regression found age, HbA1c, steroids, and carbohydrate administration independently associated with hyperglycemic time (P<0.05 for each, R2=0.49). Each 1% increase in HbA1c increased hyperglycemic time by 12±5.5%. Body mass index, APACHE II, norepinephrine dose, gender, and sepsis were not associated with hyperglycemic time by univariate or multivariate analyses.CONCLUSIONS: Even in patients without evidence of diabetes or baseline impaired glucose tolerance, hyperglycemia is common during critical illness. Time exposure to hyperglycemia is correlated with both acute stressors, including steroid and carbohydrate loads, and baseline glucose regulation, as characterized by HbA1c. Patients with low HbA1c levels are less disposed to hyperglycemia during severe illness than patients with higher, but still normal, HbA1c.
Among nondiabetics, hyperglycemia during critical illness is a function of both acute stressors and a predisposition indicated by hemoglobin glycosylation.
C. Cely, None.