Objective

This retrospective observational study compares glycemic control in the cardiac surgery intensive care unit before and after implementation of an electronic glucose management software system (EndoTool), i.e., versus a paper protocol.

Setting

820-bed academic medical center and Level 1 trauma center.

Results

• Incidence of hypoglycemia <70 mg/dL decreased 84.2% with EndoTool, from 1.96% to 0.31% of blood glucose readings; incidence <60 mg/dL decreased 91.5%, from 0.94% to 0.08%; incidence <50 mg/dL decreased 96%, from 0.49% to 0.02%; and incidence <40 mg/dL decreased 100%, from 0.22% to 0.00%.
• Incidence of hyperglycemia >180 mg/dL decreased 55.7% with EndoTool, from 18.09% to 8.02% of blood glucose readings, and incidence >200 mg/dL decreased 61.7%, from 11.66% to 4.47%.
• Time to a target blood glucose <180 mg/dL decreased 34.1% with EndoTool, from 6.27 to 4.13 hours, while time to a target blood glucose <150 mg/dL decreased 38.8%, from 9.02 to 5.52 hours.
• Number of blood glucose measurements decreased 79.1% with EndoTool, from 158 to 33 per patient visit.
• Glucose variability with EndoTool (defined as average blood glucose after achieving goal) was 130.139 mg/dL ± 23.13 for a target <180 mg/dL and 127.56 mg/dL ± 21.14 for a target <150 mg/dL.
• Average time on intravenous insulin post-cardiac surgery with EndoTool was 46.73 hours ± 42.

Conclusion

Use of an electronic glucose management system (EndoTool) in a cardiac surgery intensive care unit delivers better control and better outcomes than a paper protocol, as evidenced by reduced hypoglycemia, hyperglycemia, time to target and number of blood glucose measurements as well as glucose variability and average time on intravenous insulin consistent with best practice aims.