Objective

This observational study compares the safety and efficacy of replacing a paper-based protocol with a computer-guided glucose management system (EndoTool) for the treatment of post-operative hyperglycemia in a cardiovascular intensive care unit.

Setting

874-bed community teaching hospital.

Results

• Mean time spent free of clinical hypoglycemia <=70 mg/dL increased 3 days or 107.1% with EndoTool, from 2.8 to 3.8 days.
• Incidence of clinical hypoglycemia <=70 mg/dL decreased 40.5% with EndoTool, from 31.1% to 18.5% of patients.
• Mean time spent free of severe hypoglycemia <=40 mg/dL increased 5.9 days or 536% with EndoTool, from 1.1 to 7.0 days.
• Incidence of severe hypoglycemia <=40 mg/dL decreased 30% with EndoTool, from 2.0% to 1.4% of patients.
• Based on adjusted hazard ratios, patients treated using a paper-based protocol were 6 times as likely to experience clinical hypoglycemia as patients treated using EndoTool and more than 7 times as likely to experience severe hypoglycemia.

Conclusions

A computer-guided glucose management system (EndoTool) as a treatment modality facilitates attainment of appropriate glucose goals with significantly reduced incidence of hypoglycemia and improved patient safety in contrast to a paper-based protocol.