Hyperinsulinemic Normoglycemia Does Not Meaningfully Improve Myocardial Performance during Cardiac Surgery: A Randomized Trial

Abstract

Background: Glucose-insulin-potassium (GIK) administration during cardiac surgery inconsistently improves myocardial function, perhaps because hyperglycemia negates the beneficial effects of GIK. The hyperinsulinemic normoglycemic clamp (HNC) technique may better enhance the myocardial benefits of GIK. The authors extended previous GIK investigations by (1) targeting normoglycemia while administering a GIK infusion (HNC); (2) using improved echocardiographic measures of myocardial deformation, specifically myocardial longitudinal strain and strain rate; and (3) assessing the activation of glucose metabolic pathways.

Methods: A total of 100 patients having aortic valve replacement for aortic stenosis were randomly assigned to HNC (high-dose insulin with concomitant glucose infusion titrated to normoglycemia) versus standard therapy (insulin treatment if glucose >150 mg/dl). The primary outcomes were left ventricular longitudinal strain and strain rate, assessed using speckle-tracking echocardiography. Right atrial tissue was analyzed for activation of glycolysis/pyruvate oxidation and alternative metabolic pathways.

Results: Time-weighted mean glucose concentrations were lower with HNC (127 ± 19 mg/dl) than standard care (177 ± 41 mg/dl; P < 0.001). Echocardiographic data were adequate in 72 patients for strain analysis and 67 patients for strain rate analysis. HNC did not improve myocardial strain, with an HNC minus standard therapy difference of -1.2% (97.5% CI, -2.9 to 0.5%; P = 0.11). Strain rate was significantly better, but by a clinically unimportant amount: -0.16 s (-0.30 to -0.03 s; P = 0.007). There was no evidence of increased glycolytic, pyruvate oxidation, or hexosamine biosynthetic pathway activation in right atrial samples (HNC, n = 20; standard therapy, 22).

Conclusion: Administration of glucose and insulin while targeting normoglycemia during aortic valve replacement did not meaningfully improve myocardial function.

Figure 1

Time line of study protocol depicting the study intervention (administration of hyperinsulinemic normoglycemic clamp (HNC) vs. standard therapy; shaded area), surgical/anesthetic events, and collection of outcomes. TEE = transesophageal echocardiographic examination; CPB = cardiopulmonary bypass.

Figure 2

Consolidated Standards of Reporting Trials flow diagram. TEE = transesophageal echocardiographic examination.

Figure 3

Boxplots demonstrating changes in left (LV) and right (RV) ventricular strain and strain rate from beginning to end of surgery. Interquartile range (IQR, box), median (horizontal line), high and low values within 1.5 IQR (whiskers), outliers (circles), and mean (diamond) are shown. * P is less than the significance level of 0.025 for two group comparisons. HNC = hyperinsulinemic normoglycemic clamp.

Figure 4

Exploratory subgroup analysis of the difference (HNC minus standard therapy) and 97.5% CI of left ventricular systolic longitudinal strain and strain rate. LV = left ventricular; CABG = Coronary artery bypass grafting.

Figure 5

Boxplot demonstrating the distribution of laboratory measures before (pre-) and after (post-) aortic clamping on regulatory enzymes of the glycolytic/pyruvate oxidation pathway. If no interaction between time and treatment was found, we collapsed time and fit a main effect model. If interaction between time (pre- vs. post-) and treatment was significant (P<0.15), we compared groups at each time. Extreme outliers are not shown. Standard = standard therapy (open box); HNC = hyperinsulinemic normoglycemic clamp group (shaded box); GAPDH =glyceraldehyde 3-phosphate dehydrogenase; PDH= pyruvate dehydrogenase.

Figure 6

Boxplots demonstrating the distribution of laboratory measures before (pre-) and after (post-) aortic clamping examining adverse cellular and biochemical effects of hyperglycemia and cardioplegic arrest. Because there was no interaction with time, we collapsed time and fit a main effect model. Extreme outliers are not shown. Standard = standard therapy (open box); HNC = hyperinsulinemic normoglycemic clamp group (shaded box); TSP-1 =thrombospondin-1; O-GlcNAc =O-linked N-acetylglucosamine transferase; Egr-1 =Early growth response protein-1.