Change in practice patterns in the management of diabetic cardiac surgery patients.

Diabetes and elevated blood glucose (BG) levels > 11.1 mmol/L in the acute post-operative period have been identified as risk factors for surgical site infections (SSI) and nosocomial infections (Furnary, Zerr, Grunkemeir, & Starr, 1999; American College of Endocrinology consensus guidelines for glycemic control, 2002). Some studies have suggested that intensive insulin therapy reduced in-hospital mortality and that a continuous insulin infusion should be a standard of care for diabetic cardiac surgery patients (Furnary et al., 2003; Brown & Dodek, 2001). Our urban tertiary care teaching hospital initiated an insulin nomogram in the intensive care unit intending to more effectively control blood glucose (BG) levels in cardiac surgical patients. This cohort study compared glucose control and clinical outcomes in 53 diabetic cardiac surgery patients prior to the initiation of the insulin infusion and 50 patients following the implementation between October 2002 and April 2003. Results demonstrated target glucose control in the ICU was improved by 20% (p < .001) and mean BG was lower in the intervention group (p < .001). However, target glucose (6.1-10.0 mmol/L) was exceeded in 45% of patients in the intervention group, 65% in the control group as well as 42% of patients on the ward. The insulin nomogram is now initiated as soon as the BG is obtained immediately following patient transfer from the operating room (OR). There is more aggressive use of sliding scale insulin, and earlier resumption of pre-operative diabetic regimens on the ward.