Objective

This study compares glycemic outcomes with a traditional paper protocol to a computerized glucose management system (EndoTool) for patients receiving intravenous insulin.

Setting

207-bed not-for-profit community hospital.

Results

• Incidence of hypoglycemia <70 mg/dL decreased 89% with EndoTool, from 6.9% to 0.76% of blood glucose readings
• Incidence of hypoglycemia <40 mg/dL was .01% of blood glucose readings with EndoTool.
• Incidence of hyperglycemia >180 mg/dL decreased 54% with EndoTool, from 26.3% to 12.1% of blood glucose readings, and incidence >300 mg/dL decreased 68%, from 4.9% to 1.55% of blood glucose readings.
• Percentage of blood glucose readings in target increased 35% with EndoTool, from 63.6% to 86.1%.

Conclusions

Use of a computerized glucose management system (EndoTool) achieves substantially better glucose control than a traditional paper protocol, with lower incidence of both hypoglycemia and hyperglycemia and a greater percentage of blood glucose readings in target.