Control of hyperglycemia among septic and nonseptic patients in the general intensive care unit

Background

Hyperglycemia is common among patients admitted to intensive care units (ICUs) and carries risk for various complications, especially sepsis, and prolonged ICU stay. Some studies have demonstrated that intensive insulin control of blood glucose reduced morbidity and mortality. The objective was to determine whether intensive or conventional insulin control of blood glucose in hyperglycemic septic and nonseptic ICU patients is correlated with the prognosis.

Materials and methods

Septic and nonseptic ICU patients with hyperglycemia were randomly assigned to a group treated intensively with insulin targeting glucose at 6.6–8.4 mmol/l or to a conventional insulin therapy group where glucose upon exceeding 12 mmol/l was controlled at 8.4–12 mmol/l. Rates of morbidity and mortality, hypoglycemic episodes and required insulin dosage were compared.

Results

A total of 89 patients were enrolled, including 27 patients with sepsis: 11 patients were treated with insulin intensively with a mean glucose of 8.3 mmol/l, while 16 patients received conventional insulin treatment with a mean of 10.3 mmol/l. Thirty nonseptic patients received intensive insulin treatment with a mean of 8.46 mmol/l and 32 nonseptic patients were treated conventionally with a mean of 10.4 mmol/l. Among septic patients, both groups were similar with respect to age and Acute Physiology and Chronic Health Evaluation scores. There was no significant difference between groups in the morbidity, including rates of new infection, renal and hepatic damage. There was a somewhat shorter ICU stay in the intensive treatment group. Both groups had similar ICU, inhospital and 28-day follow-up mortalities and similar rates of hypoglycemic episodes. The daily dosage of insulin was higher with the conventional treatment. Similar results were obtained among nonseptic patients between both groups, but septic patients had a longer total ICU stay and higher mortality.

Conclusion

Intensive insulin control of blood glucose at 8.4 mmol/l does not affect the mortality or morbidity of septic and nonseptic patients in intensive care, except for a somewhat shorter ICU stay. An increased insulin dosage in the conventional treatment group was attributed to the group's higher initial blood glucose, probably due to a higher prevalence of diabetes and associated insulin resistance and toxicity hyperglycemia.