Number 29, 2006
Bypass surgery for coronary artery disease: a vanishing treatment?

Metabolic interventions during cardiac surgery: focus on glucose–insulin–potassium

Back to the Summary

Harry B. van Wezel
Department of Anesthesiology, Academic Medical Center, Amsterdam, The Netherlands
Correspondence: Dr Harry B van Wezel, Department of Anesthesiology, Academic Medical Center, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands.
Tel: +31 20 5669111; fax: +31 20 6979441; e-mail: H.B.vanwezel@amc.uva.nl

Introduction
Since the days of the first cardiac surgical operations in the previous century, myocardial preservation has been an essential component of the successful outcome of these procedures. Although many different techniques to achieve myocardial preservation and modulation have been described over the past 50 years, this review will focus on the use of glucose–insulin–potassium (GIK) in patients with (post) ischemic myocardium.
It has been known for decades that administration of insulin in patients who have experienced ischemic events may have positive effects on morbidity and mortality. In 1962, Sodi-Pallares et al [1] showed, in experimental animals and in patients with acute myocardial infarction, that infusion of GIK reduced electrocardiographic signs of ischemia, reduced ventricular ectopy, limited infarct size, and improved survival. After this classic publication, enthusiasm for the use of GIK was somewhat dampened by a report from the Medical Research Council of the UK stating that GIK treatment failed to show any positive effect on survival in patients with acute myocardial infarction. However, publications describing the beneficial effect of “the metabolic cocktail” in patients with acute myocardial infarction continued to appear regularly. In 1995, Malmberg et al [2] demonstrated that insulin–glucose infusion improved longterm prognosis in diabetic patients with acute myocardial infarction. In 1997, Fath-Ordoubadi and Beatt [3] described a meta-analysis of nine randomized placebo-controlled studies all using GIK in patients with acute myocardial infarction; they found that GIK therapy led to a highly significant (28%) reduction in mortality. In 1998, the Estudios Cardiologicos LatinoAmerica (ECLA) study group [4] showed significant GIK-induced reduction in mortality in patients with acute myocardial infarction who also underwent a reperfusion strategy. In 2003, van der Horst et al [5] showed that the use of GIK therapy as an adjunct to coronary angioplasty in patients with acute myocardial infarction led to reduced mortality in patients with Killip class 1 only. Surprisingly, two recent large studies (CREATE–ECLA and the Diabetes Mellitus, Insulin Glucose Infusion in Acute Myocardial Infarction-2 trial) failed to report any advantage of GIK in patients with acute myocardial infarction [6,7]. The rather unexpected results of these two studies have been discussed in several Editorials and have been explained to a large extent by flaws in the design and practical execution of the protocols [8,9]. However, in the cardiological field, the use of GIK has lost some of its “esteem”, and only very well designed clinical studies may lead to renewed acceptance of this old and controversial technique for the management of patients with acute myocardial infarction in the future.

Glucose–insulin–potassium in cardiac surgery
In the early days of cardiac surgery, GIK was used to induce cardioprotection and for weaning off bypass. However, the technique was more or less abandoned with the introduction of St Thomas’ cardioplegia and hypothermic cardiopulmonary bypass techniques in the mid-1970s and the subsequent availability of inotropic agents such as dopamine and dobutamine, and anesthetic agents with minimal cardiodepressant and vasodilatory effects such as the synthetic opioids.
In the 1980s, GIK therapy was “rediscovered” in the setting of cardiac surgery, for several reasons, including: an increase in the number of patients with unstable coronary syndromes (ie, severe myocardial ischemia before operation) requiring emergency coronary artery bypass grafting (CABG), the introduction of warm cardioplegia and cardiopulmonary bypass techniques (possibly allowing improved metabolic stimulation of normal myocardial enzyme function), and, more generally, because it appeared that the limits of adequate cardioprotection had been reached, especially in the growing cohort of elderly cardiac surgical patients with a history of severe, long-established coronary artery disease, chronic heart failure, and reduced contractile reserve before operation. This type of patient, in particular, frequently requires prolonged episodes of extracorporeal circulation for complicated coronary revascularization. In such patients, hemodynamic abnormalities and acute heart failure frequently develop after extracorporeal circulation.
Serious cardiovascular complications usually begin in the period after extracorporeal circulation. At that time, acute ventricular failure may develop – a condition probably caused by postischemic dysfunction or myocardial stunning. This phenomenon may thus be superimposed on pre-existing impaired ventricular function before operation.
The standard treatment consists of large doses of inotropic agents, glyceryl trinitrate, peripheral vasopressors, or combinations thereof, and intra-aortic balloon pumping. The use of β-adrenoceptor-stimulating therapy at high infusion rates, in particular, is associated with a number of undesirable side effects, including tachycardia and increased oxygen requirement of the (postischemic and dysfunctional) myocardium, and is often effective for only a limited period of time. The last of these may be the result of acute β-receptor downregulation, an increase in plasma lipids in the presence of high endogenous and exogenous catecholamine concentrations, insulin resistance, and a reduction in myocardial glucose uptake and utilization.
Merhige et al [10] performed a study to test the hypothesis that adrenergic stimulation suppresses myocardial uptake of glucose. They measured myocardial activity of [¹⁸F]2-fluoro-2-deoxyglucose (FDG) in glucose-loaded dogs, randomly studied during dopamine infusion, during insulin infusion, and during their combined infusion. They concluded that myocardial uptake of FDG was significantly decreased when animals were treated with dopamine, compared with treatment of the same animals with insulin (P < 0.03) or a combination of insulin and dopamine. The results demonstrated that dopamine inhibits the myocardial uptake of FDG by increasing the concentration of circulating free fatty acids, and that this inhibition can be reversed by insulin on the basis of substrate availability and competition [10]. These findings may have clinical importance in patients requiring long-term treatment with exogenous catecholamines.
During the past 10 years, the considerations described above led to renewed interest in the role of GIK therapy in patients undergoing CABG, especially after extracorporeal circulation and in the intensive care period. In 1995, Svedjeholm et al [11] used GIK successfully in an open, uncontrolled study in cardiac surgical patients with heart failure. They reported almost full recovery of hemodynamic performance in the majority of patients at 6 h after bypass. In 1997, Lazar et al [12] undertook a randomized placebo-controlled study in patients with unstable angina during urgent CABG, and described reduced inotropic requirement, improved cardiac index, and requirements for a shorter duration of intensive care and shorter total hospital stay, associated with GIK therapy. Also in 1997, Taegtmeyer et al [13] reported a retrospective analysis of cardiac surgical patients with impaired left ventricular function randomly treated with GIK or placebo. They concluded that an “aggressive” therapy of postischemic dysfunctional myocardium appeared to be beneficial when pharmacological and mechanical measures had failed to improve cardiac function.
In 2001, van den Berghe and coworkers [14] demonstrated that intensive insulin therapy reduced morbidity and mortality in critically ill patients. They showed that, in particular, a subgroup of cardiac surgical patients receiving intensive insulin therapy (blood glucose concentrations between 4.4 and 6.1 mmol/L), starting at the time of their arrival in the intensive care unit and continuing until they were discharged to the ward, benefited significantly. This pivotal study showed for the first time that mortality can be significantly improved in cardiac surgical patients using “metabolic modulation”.
In 2000 and in 2004, Lazar et al [15,16] showed that the use of GIK led to reduced perioperative morbidity in diabetic individuals undergoing CABG. This was an important finding, because the findings of a study involving more than 140 000 patients undergoing CABG have confirmed that diabetes mellitus is a significant risk factor for short-term morbidity and mortality in this group [17].
Summarizing the findings of these studies using GIK techniques in cardiac surgical patients, there appears to be a beneficial effect of this approach on postoperative morbidity and mortality. This should be sufficient reason to develop further effective and safe techniques for the intravenous delivery of GIK and to unravel its mode of action.

Mechanisms involved in the beneficial effect of glucose–insulin–potassium
In spite of the widespread use of different GIK strategies in cardiac surgery and acute myocardial infarction, the exact mode of action underlying this approach remains to be elucidated. GIK research initially focused on and demonstrated the ability of insulin to influence substrate flux through myocardial metabolic pathways and transmembrane signaling. Furthermore, GIK infusions were found to attenuate postischemic disturbances in lipid and glucose homeostasis, in the setting of CABG and in acute myocardial infarction [1–11]. Recent evidence from animal studies suggested that GIK has the potential to reduce the inflammatory response [18]. In patients with acute myocardial infarction, Chaudhuri et al [19] were the first to demonstrate the anti-inflammatory effects of insulin, as reflected by a reduction in the absolute increase in postinfarct concentrations of C-reactive protein and serum amyloid A. In 2005, Visser et al [20] showed for the first time in patients undergoing CABG that GIK, applied as a hyperinsulinemic normoglycemic clamp, has anti-inflammatory effects, as reflected by a significant reduction in postoperative C-reactive protein and serum amyloid A concentrations and a reduction in postoperative leukocytosis. This innovative approach to maintaining tight glycemic control may be useful, because acute (stress) hyperglycemia frequently develops in patients undergoing CABG, and it has been demonstrated in both rats and humans that proinflammatory cytokine concentrations are increased by acute hyperglycemia [21,22]. In addition, it has been demonstrated that stress hyperglycemia per se is associated with adverse outcome of CABG and acute myocardial infarction [23,24].
Future studies in (high-risk) cardiac surgical patients are required to assess the effect of different GIK infusion techniques on perioperative inflammatory control and its hypothetical association with reduced perioperative morbidity and mortality. This is especially important as, with increasing numbers of elderly and diabetic patients, the cohort of high-risk patients will grow during the next decades. Other fields of interest that may be beneficially affected by perioperative infusion of GIK include: the coagulation cascade, complement activation, neurohumoral stress responses, parameters reflecting insulin resistance (ketone bodies and lactate), the lipid profile, and leukocyte function.

REFERENCES