The association of preoperative glycemic control, intraoperative insulin sensitivity, and outcomes after cardiac surgery.

Abstract

The impairment of insulin sensitivity, a marker of surgical stress, is important for outcomes. The aim was to assess the association between the quality of preoperative glycemic control, intraoperative insulin sensitivity, and adverse events after cardiac surgery. We conducted a prospective cohort study at a tertiary care hospital. Nondiabetic and diabetic patients scheduled for elective cardiac surgery were included in the study. Based on their glycosylated hemoglobin A (HbA(1c)), diabetic patients were allocated to a group with good (HbA(1c) <6.5%) or poor (HbA(1c) >6.5%) glycemic control. We used the hyperinsulinemic-normoglycemic clamp technique. The primary outcome was insulin sensitivity measurement. Secondary outcomes were major complications within 30 d after surgery including mortality, myocardial failure, stroke, dialysis, and severe infection (severe sepsis, pneumonia, deep sternal wound infection). Other outcomes included minor infections, blood product transfusions, and the length of intensive care unit and hospital stay. A total of 143 nondiabetic and 130 diabetic patients were studied. In diabetic patients, a negative correlation (r = -0.527; P < 0.001) was observed between HbA(1c) and intraoperative insulin sensitivity. Diabetic patients with poor glycemic control had a greater incidence of major complications (P = 0.010) and minor infections (P = 0.006). They received more blood products and spent more time in the intensive care unit (P = 0.030) and the hospital (P < 0.001) than nondiabetic patients. For each 1 mg x kg(-1) x min(-1) decrease in insulin sensitivity, the incidence of major complications increased (P = 0.004). In diabetic patients, HbA(1c) levels predict insulin sensitivity during surgery and possibly outcome. Intraoperative insulin resistance is associated with an increased risk of complications, independent of the patient's diabetic state.