Objectives: To compare the effectiveness and safety of a paper-based and a computerized algorithm used for tight glycemic control.
Setting: Academic pediatric intensive care unit.
Design: Retrospective cohort study.
Patients: Two groups of nondiabetic critically ill children with persistent hyperglycemia (blood glucose ≥140 mg/dL [ ≥7.8 mmol/L] for at least 2 hrs) were included.
Intervention: One group of patients' blood glucose was controlled at 90-119 mg/dL (5.0-6.6 mmol/L) using the Yale Insulin Infusion Protocol (YIIP), a paper-based protocol. Another group of patients' blood glucose was controlled at 80-110 mg/dL (4.4-6.1 mmol/L) with eProtocol insulin (ePi), a computerized decision support tool.
Measurements and main results: The effectiveness of the protocols was compared using percentages of blood glucose values within target range and glucose variability index. A safety comparison was made using hypoglycemia rates at ≤40 mg/dL (≤2.2 mmol/L), ≤50 mg/dL (≤2.8 mmol/L), and ≤60 mg/dL (≤3.3 mmol/L). Forty-two patients and 12 patients were included in the YIIP and ePi groups, respectively. The percent of values in range was lower in the YIIP group (33%) compared with the ePi group (41%) (p < .001). Mean glucose variability index was comparable in the two groups (18.7 ± 8.9 mg/dL/hr [1.0 ± 0.5 mmol/L/hr] for the YIIP group and 14.4 ± 7.6 mg/dL/hr [0.8 ± 0.4 mmol/L/hr] for the ePi group; p = .111). Hypoglycemia rates were statistically similar in both groups. In the YIIP group, 10% of patients and in the ePi group, 25% of patients had blood glucose ≤40 mg/dL (≤2.2 mmol/L) (p = .168).
Conclusion: YIIP is less effective but is as safe as ePi in achieving tight glycemic control. We are awaiting the results of two multicenter trials designed to determine the survival benefit of tight glycemic control in children. Further studies are needed to determine the clinical significance of the different glucose metrics in critically ill patients.