Number 27, 2005
Metabolic approach in heart failure

Glucose–insulin–potassium infusion enhances myocardial wall motion in patients with chronic ischemic heart failure

Back to the Summary

Lucas J. Klein¹, Linda (C.)M. C. van Campen¹, Thomas H. Marwick^2
1Department of Cardiology, VU University Medical Center, Amsterdam, The Netherlands ²Department of Medicine, Princess Alexandra Hospital, University of Queensland, Brisbane, Australia
Correspondence: Lucas J. Klein, VU University Medical Center, Department of Cardiology, Room 6D120, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands.
Tel: +31 204442244, fax: + 31 204442446, e-mail: lj.klein@vumc.nl

Introduction
Hyperinsulinemic euglycemic clamping is a procedure that is frequently used in combination with [¹⁸F]2-fluoro-2-deoxyglucose (FDG) positron emission tomography scanning to determine viability in patients with heart failure caused by ischemic heart disease. The procedure was first described by DeFronzo et al [1] as a method to determine insulin secretion and resistance, but was adapted to enhance the uptake of glucose and FDG in myocardial tissue. The method closely resembles the metabolic modulation by infusion of glucose–insulin–potassium (GIK) that has been used in several studies in patients with acute myocardial infarction. Fath-Ordoubadi and Beatt [2] performed a meta-analysis of studies using GIK before the era of thrombolysis or percutaneous coronary intervention. Other trials using GIK in myocardial infarction were the Diabetes Mellitus and Insulin Glucose Infusion in Acute Myocardial Infarction (DIGAMI) study [3] and the pilot Estudios Cardiologicos LatinoAmerica (ECLA) trial [4]. The meta-analysis and both the DIGAMI study and the ECLA trial showed a benefit (reduction in mortality) from GIK in patients with myocardial infarction. However, the mechanism responsible for this reduction in mortality is unclear.
Several authors have observed that administration of insulin to the (ischemic) heart increases systolic function and results in an increased relaxation [5–9]. Dobutamine, a synthetic catecholamine, increases myocardial systolic function and is frequently used in dysfunctional myocardium in ischemic heart disease to determine contractile reserve, which is a predictor of recovery of function after revascularization [10]. Because insulin enhances myocardial systolic function in dysfunctional myocardium after ischemia, it may be possible that it enhances segmental myocardial wall motion to an extent similar to that achieved with the infusion of low-dose dobutamine (LDD).

Patients and methods
In total, 75 patients, referred to either the VU University Medical Center or Princess Alexandra Hospital for evaluation of viability, were included in the study. Patients were eligible for inclusion if they had ischemic heart disease with left ventricular dysfunction. Exclusion criteria were left bundle branch block, pacemaker rhythm, presence of serious ventricular tachyarrhythmias, atrial fibrillation, and the presence of a poor acoustic window. Patients gave informed consent before inclusion in the study and the Ethics Committees of both hospitals approved the study. Patients underwent standard LDD echocardiography and GIK echocardiography on two separate occasions.

Echocardiography
Standard echocardiograms were obtained with standard parasternal long-axis and short-axis views, in addition to apical four-, three- and two-chamber views. Echocardiograms were obtained at baseline and after LDD and GIK infusions.

Dobutamine procedure
Patients underwent LDD echocardiography according to a standardized procedure, starting with 5µg/kg per min for 5min, and then increasing to 10µg/kg per min. Further evaluation for ischemia followed the standard of the Princess Alexandra Hospital.

Glucose–insulin–potassium procedure
Patients received an infusion of insulin 100mU/kg per h plus glucose 20% with KCl 40mmol/L at a target rate of 1.8mL/kg per h (6mg/kg per min; rate adjusted for instantaneously determined blood glucose concentrations) during a period of 1h (VU University Medical Center) or 30mL/h of a solution of 10% glucose, 80 units of insulin and potassium 40mmol/L for 4h (Princess Alexandra Hospital; for a 75kg individual: insulin 32mU/kg per h and glucose 0.67mg/kg per min).

Echocardiographic analysis
Echocardiograms were judged by experienced observers according to a 13-segment (6 basal, 6 mid, 1 apical) model [11] on a 3-point scale: 0=normokinesia, 1=hypokinesia, 2=a/dyskinesia. For each dyssynergic segment, the response to LDD or GIK was determined. Wall motion score (WMS) was calculated as the sum score of visualised segments in each patient; a higher score implies a worse CV function. Wall motion score index (WMSI) was calculated as WMS divided by the number of segments visualized and is less dependent on acoustic window.
Comparisons were made between baseline and each intervention, and between the results of the interventions. Total agreement and Cohen's kappa coefficient [12] were calculated.

Results
A sample image of the response to dobutamine and insulin is shown in Figure 1. In all, 973 segments were suitable for analysis. Figure 2a shows the improvement (decrease) in WMS achieved with each intervention: 3.1±2.1 (from 11.4±4.4 to 8.0±4.5; P<0.0001) during LDD and 3.1±2.2 (from 11.4±4.4 to 8.2±4.3; P <0.0001) during GIK. Likewise, WMSI decreased, by 0.25±0.16 (from 0.87±0.35 to 0.62±0.35; P <0.0001) during LDD and by 0.23±0.17 (from 0.86±0.34 to 0.63±0.35; P<0.0001) during GIK. The number of segments exhibiting an improvement in response to LDD and GIK was also significant (Figure 2b): with LDD, at baseline 7.2±2.6 segments were dysfunctional and improvement was found in 3.0±1.9; in concert, with GIK, at baseline 7.4±2.6 segments were dysfunctional and there was improvement in 2.8±2.1 (P=NS for the number of improving segments).

Discussion
This study of patients with chronic left ventricular dysfunction and coronary artery disease has shown that infusion of GIK leads to increased segmental myocardial wall motion (decreased WMS and WMSI). It provided an opportunity to study contractile reserve by means of metabolic modulation similar to that which occurs in myocardial infarction. There was a high agreement with the contractile reserve determined by LDD infusion.
Other authors also have observed an effect of GIK on myocardial wall motion [13–16]. In a study of patients from a category similar to that in our present study, Cottin et al [13] observed improvement in systolic myocardial function (determined by echocardiography) after 20min of infusion of GIK at a rate of insulin 300mU/kg per h. Twenty and forty minutes after discontinuation of the infusion, improvement in wall motion was still present. No comparison was made with LDD. Also using echocardiography, Yetkin et al observed an improvement in wall motion in patients with chronic left ventricular dysfunction, in patients with recent myocardial infarction [14], and in patients with chronic coronary artery disease and left ventricular dysfunction [15], using insulin 100µU/kg per h. Alan et al [16] observed a significant increase in ejection fraction, prolongation of diastolic filling period, decrease in wedge pressure and decrease in defect score by resting technetium-99m (^99mTc)-sestamibi scintigraphy after administration of GIK at a rate of insulin 300mU/kg per h over 24h.
In the studies using GIK to enhance myocardial wall motion, many different procedures have been used, varying from that of the DIGAMI study (also used in the Princess Alexandra Hospital patients in this study: for a 75kg man, insulin 32mU/kg per h) to a high-dose insulin regimen (300mU/kg per h) for shorter (20min) or longer (24h) periods. In the studies under discussion here, no changes in heart rate or blood pressure were observed, which may be of advantage in patients who have suffered myocardial infarction. In canine papillary muscle preparations, incubation with insulin led to an increased contractile force that counteracted the addition of propanolol – an effect that could be reproduced by coronary injection of insulin [6]. Future studies will be needed to demonstrate whether echocardiographic assessment of the response to short-duration infusion of high-dose GIK can predict recovery of function. The benefit of GIK in patients with myocardial infarction remains to be demonstrated in one or more large, prospective trials, especially because treatment of myocardial infarction has been changed dramatically by the introduction of thrombolysis and primary percutaneous intervention, and a number of these studies were performed in the prethrombolytic era.

Mechanism of action
Infusion of GIK results in a reduction in circulating free fatty acids [17–22], often to less than the myocardial uptake threshold concentration, which has been determined to be between 100 and 200µmol/L [17]. Myocardial uptake of free fatty acids is therefore minimized [17–22], uptake of glucose and lactate increases [17–22], and the myocardial respiratory quotient changes towards 1.0 [18,19] (in the fasting state, when free fatty acids are the predominant fuel, the myocardial respiratory quotient is around 0.7). In most studies, a change in oxygen consumption was not observed [18,19,21], and only one study has related oxygen consumption to a clinical determinant of oxygen consumption (pressure–volume–area [PVA] loops) and observed a reduced “normalized“ oxygen consumption, especially in the unloaded heart [23].
Altered availability of substrate is exploited in anaerobic glycolysis, providing ATP to improve intracellular calcium homeostasis [24,25]. Although a direct effect of insulin on calcium metabolism in human myocardium has not yet been established, circumstantial evidence suggests that there will be an effect, as insulin-like growth factor seems to improve contractile function in cardiomyocytes isolated from failing human hearts [26,27]. Furthermore, the transition from free fatty acid to glucose metabolism will result in a more efficient use of oxygen, providing more ATP per molecule of oxygen used [28].
In this study, the concentration of glucose was kept greater than 4.0mmol/L (preferably more than 5.0mmol/L), and thus a catecholamine-induced increase in myocardial wall motion as a result of hypoglycemia was unlikely to be present. Furthermore, catecholamine concentrations measured in patients at the Princess Alexandra Hospital did not exhibit such an increase (data not shown).
The mechanism of action of LDD on myocardium with decreased wall motion in coronary artery disease is believed to be partly attributable to the increase in myocardial blood flow, which provides more oxygen and energy supply to the middle and outer layer of the myocardium and enables them to increase contractility. After the infusion of GIK, ^99mTc-sestamibi or ^99mTc-tetrofosmin scintigraphy show decreased defect scores, suggesting increased myocardial perfusion [6,29]. This was confirmed by a study in which insulin was injected directly into the coronary arteries in patients with coronary artery disease, with a significant 10% increase in myocardial blood flow and a transition from free fatty acid metabolism to glucose metabolism without an increase in myocardial oxygen consumption [30]. In another study, myocardial blood flow increased and oxygen extraction decreased, together with a decrease in coronary vascular resistance [18].

Conclusion
Infusion of GIK enhances segmental wall motion and can be used to detect myocardial contractile reserve in patients with chronic coronary artery disease, with the potential advantage over low-dose dobutamine echocardiography that it does not cause an increase in blood pressure or heart rate during the infusion. ?

Acknowledgment
The authors wish to thank Vince R. Khoury for his support with collecting patient data.

Back to the Summary

REFERENCES

Back to the Summary