Amino acid-enriched glucose-insulin-potassium infusion improves hemodynamic function after coronary bypass surgery. A...

The goal of this study was to assess the effects of a combination of glucose-insulin-potassium (GIK) and the amino acids aspartate and glutamate upon perioperative hemodynamics in coronary surgery patients with unstable angina and/or compromised left ventricular function.

Prospective, randomized, and double-blind clinical study.

Operating theatre and intensive care unit (ICU) of a university hospital.

44 coronary artery bypass graft (CABG) patients with unstable angina and/or compromised left ventricular function.

22 patients (group A) were given 1l of an infusion with 250g glucose, 100 I.U. fast-acting human insulin, 72 mmol potassium, 32 mmol magnesium, 20 mmol phosphate, 65 mmol aspartate, and 65 mmol glutamate, while another 22 patients (group C) were given 1l of an infusion with 50 g glucose, 72 mmol potassium, 32 mmol magnesium, and 8 mmol phosphate. The infusion rate was 1.2 ml/kg/h from the anesthesia induction onward to the commencement of cardiopulmonary bypass, when it was reduced to 0.8 ml/kg/h. When 11 had been infused, but not later than 4 a.m., the infusion was continued by giving 10% glucose at the same rate to both groups. Additional insulin (median: 14.2 I.U., range: 0-41.5) or saline was given during bypass to the A and C patients, respectively. A blood cardioplegia technique containing aspartate and glutamate was used in both groups.

At aortic cannulation, the cardiac index (CI) had increased from the pre-anesthetic level by 15.3% (mean) (SD: 31.7%) in group A and decreased by 7.7% (15.1%) in C patients, p = 0.0069. Also the changes in stroke index (SI; p = 0.022), left (LVSWI; p = 0.0037) and right ventricular stroke work index (RVSWI; p = 0.0097) were more favorable in group A. Despite longer aortic cross-clamp, p = 0.031, and perfusion times, p = 0.042, in A patients, the change in cardiac index was also better in this group after bypass: At decannulation, the difference between mean values was 31.8%, p = 0.0001, and at arrival in the ICU it was 16.1%, p = 0.028. The same was also seen 8 h postoperatively and on the 1st and 2nd postoperative mornings; p = 0.034, 0.040, and 0.037, respectively (Wilcoxon test). Favorable changes were seen for the A patients also regarding SI at decannulation (p = 0.0002) and after 8 h (p = 0.017); LVSWI at decannulation (p = 0.0002), at arrival in the ICU (p = 0.0023), and after 8 h (p = 0.0011); and RVSWI at decannulation (p = 0.0027), at the ICU (p = 0.021), after 8 h (p = 0.014), and on the 1st postoperative morning (p = 0.039). However, the response to a hemodynamic loading test (6% hydroxyethyl starch 5 ml/kg) was similar in the 2 groups, and there was no difference in the need for inotropic support.

Amino acid-enriched GIK infusion improves hemodynamic function in CABG patients with unstable angina and/or compromised left ventricular function.