ICU patients have higher mortality with blood glucose level above 190 mg/dL

A glucose level above 190 mg/dL (10.5 mmol/L) is significantly linked with a higher risk of in-hospital mortality among critically ill patients, according to new results of an observational study.

Critically ill adult patients who were expected to be in the ICU for at least three days were prospectively enrolled, and all patients had a continuous glucose monitor (CGM) placed. Patients with diabetic ketoacidosis, a hyperosmolar hyperglycemic state, less than 24 hours of CGM data, and ICU readmissions and patients taking greater than 4 g of acetaminophen were ultimately excluded from the study.

Patients were managed using a standard ICU glucose protocol with a blood glucose target of 140 to 180 mg/dL (7.8 to 10.0 mmol/L), with blood glucose levels measured hourly to every four to six hours based on clinical need. IV insulin was given by a micropump at a predetermined level and was appropriately adjusted based on blood glucose levels. Clinicians were blinded to the CGM data throughout the study, but researchers subsequently used the data to determine the threshold at which hyperglycemia increased mortality risk in critically ill patients. Findings were published by Diabetologia on April 3.

Of the 293 patients included (198 male, 95 female; mean age, 68 years), 23.5% had pre-existing diabetes. There were 66 in-hospital deaths throughout the study period, 49 in patients without diabetes and 17 in patients with diabetes. After adjusting for covariates, researchers found that only time above a glucose level of 190 mg/dL (10.5 mmol/L) was significantly associated with in-hospital mortality risk. Data showed that as patients' glucose control ranges increased from 190 to 250 mg/dL (10.5 to 13.9 mmol/L), in-hospital mortality hazards also incrementally increased with every 10% rise. Risk of in-hospital mortality increased by 12.1% (hazard ratio [HR], 1.121; 95% CI, 1.003 to 1.253) per absolute 10% decrease in time in the range of 70 to 190 mg/dL (3.9 to 10.5 mmol/L).

Time in range has become popular as a glucose measure with increasing use of CGMs, the study authors noted, and these results suggest "the optimal upper glucose range for defining [time in range] in critically ill patients may be set near 10.5 mmol/l (190 mg/dl)," they wrote. More research is needed to determine whether targeting an even lower level would reduce mortality risk, they added.

One limitation to the study is that the CGM device used may be less accurate in the hypoglycemia range, so more research is needed to determine the optimal lower thresholds for target glucose ranges, the study authors said. They noted that, in practice, glucose ranges should be defined based on patient characteristics and treatment purpose.