Strict or Liberal Insulin Protocol Following Coronary Artery Bypass Graft (CABG) Surgery (SLIP)

• Operative death, major adverse cardiac events (death, myocardial infarction, re-vascularization), re-operation, Cerebrovascular accident, Deep Sternal Wound Infection, Prolonged Ventilatory Support, Acute Renal Failure, and prolonged inotropic support. [ Time Frame: 30 days ] [ Designated as safety issue: No ]
  

• The second pre-specified endpoint will be all-cause mortality at 90 days. [ Time Frame: 90 days ] [ Designated as safety issue: No ]
  

Hyperglycemia is commonly encountered following cardiac surgery, whether a patient has a history of diabetes or not. Hyperglycemia has been associated with increased perioperative morbidity and mortality; several studies have demonstrated that glycemic control utilizing insulin protocols improves operative mortality, lowers operative morbidity (mediastinitis, atrial fibrillation), and improves long-term survival. However, the optimal target for serum glucose has not been established in post-CABG patients.

Methods:

All CABG patients will be consented prior to surgery. Inclusion criteria for non-diabetic patient is a random fingerstick blood glucose (FSBG) above >150 mg/dL prior, during, or immediately following surgery. All patients with history of diabetes mellitus (Type 1 or Type II) will be immediately eligible for inclusion.

Following CABG surgery, if the patient was started intra-operatively on an insulin infusion, then that patient will be randomized to one of two treatment target groups: Group 1 [Blood Glucose (BG): 80 mg/dL-120 mg/dL] or Group 2 [BG: 121-180 mg/dL]. The randomization design will be a 1:1 allocation of patients between the two groups, with both diabetic and non-diabetic patients enrolled in both arms of the study. Patients will be maintained on an electronic-based protocol of intravenous insulin for a minimum of 72 hrs postoperatively. Patients remaining in the CVICU greater than 72 hrs will have their intravenous insulin continued until transfer to the step-down unit.

The Glucommander© will be programmed to adjust the insulin drip to one of these two target groups. The nursing staff will not be blinded to treatment group allocation. The primary endpoint with be a composite of operative death, major adverse cardiac events (MACE: death, myocardial infarction, re-vascularization), STS Defined Major Morbidity (re-operation, Cerebrovascular accident, Deep Sternal Wound Infection/Mediastinitis, Prolonged Ventilatory Support (> 24 hrs), Acute Renal Failure), and prolonged inotropic support. The pre-specified sub-group analysis will compare perioperative outcome of patients with diabetes vs non-diabetic patients.

Hypothesis:

Our hypothesis is that the perioperative outcome of Group 2 [BG: 121 - 180 mg/dL] will not be inferior to Group 1 [BG: 80-120 mg/dL]. We anticipate significantly more hypoglycemic events in Group 1.