Background  

To evaluate the safety and efficiency of a protocol for glycemic control in intensive care unit (ICU) patients with neurovascular or head injury.

Methods  

Two cohorts of 50 consecutive patients admitted to the ICU with an admission diagnosis of neurovascular or head injury before and after protocol implementation were evaluated. All patients in the interventional cohort received insulin using a standardized intravenous insulin infusion protocol targeting blood glucose levels of 7–9 mmol/l. Efficiency (time to reach and time within target range), safety (hypoglycemia), and nursing compliance (protocol violations) were evaluated.

Results  

The median time to reach the target blood glucose range was shorter in the interventional cohort than the conventional cohort (5.0 h [0.5–20.5 h] vs. 12.9 h [1.3–90.3 h]; P < 0.001). More time was spent within target range in the interventional cohort than in the conventional cohort (36.4 ± 16.3% vs. 27.1 ± 19.0%; P < 0.001). The median prevalence of mild (<4.9 mmol/l) hypoglycemia (0 [0–1.11]% vs. 0.58 [0–2.79]%; P < 0.001) and moderate (<3.9) hypoglycemia (0[0–0.55]% vs. 0 [1–1.25]%; p < 0.001) was significantly lower in the interventional cohort.

Conclusions  

The intravenous insulin infusion protocol improved the safety and efficiency of glycemic control for ICU patients with neurovascular or head injury.