| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2005;80:511-517
© 2005 The Society of Thoracic Surgeons
a Department of Anesthesiology, Division of Intensive Care, University of Oulu, Oulu, Finland
b Department of Internal Medicine, University of Oulu, Oulu, Finland
c Department of Surgery, University of Oulu, Oulu, Finland
Accepted for publication March 4, 2005.
* Address reprint requests to Dr Koskenkari, Department of Anesthesiology, Box 21, Oulu FIN-90029, Finland (Email: juha.koskenkari{at}fimnet.fi).
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
METHODS: Forty elective patients scheduled for combined aortic valve replacement and coronary artery bypass surgery were randomly assigned to receive either high-dose insulin treatment (short-acting insulin 1 IU·kg–1 ·h–1 with 30% glucose 1.5 mL·kg–1 ·h–1 administered separately) or control treatment (saline). The blood glucose levels were maintained within a targeted range by adjusting the rate of glucose infusion in the treatment group and by short-acting insulin bolus doses in the control group.
RESULTS: The lactate clearance was faster (p = 0.046), and the lactate levels (p = 0.016), blood glucose levels (p < 0.001), and free fatty acid levels (p < 0.001) were lower in the insulin group postoperatively. Besides, there was lesser need for dobutamine support (p = 0.013) and a trend toward better cardiac indices. Insulin treatment increased the respiratory quotient (p < 0.001), but there were no differences between the groups with regard to systemic oxygen consumption or energy expenditure measured by indirect calorimetry. The average glucose uptake in the insulin group was 7.1 g/kg in 24 hours (28 kcal·kg–1 ·day–1).
CONCLUSIONS: The high-dose insulin treatment was associated with lower blood glucose levels, better preserved myocardial contractile function, and less need for inotropic support, and hence led to lower lactate levels postoperatively. The protocol is safe, but requires strict control of blood glucose level.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
However, although high-dose GIK is primarily intended to optimize myocardial glucose uptake and to overcome insulin resistance, it also has a major impact on systemic metabolism. The nonphysiologically high insulin dose often requires a large amount of glucose to be infused to maintain normoglycemia. There is some evidence that a high caloric intake may have detrimental effects on the body nutritional status, and be associated with an adverse outcome during the intensive care unit (ICU) stay [7]. A large carbohydrate load has been shown to place a burden on the respiratory system [8], and it may lead to hypermetabolism and heat production postoperatively. There might also be a risk that high-dose GIK treatment increases the whole-body metabolic drive and thereby increases oxygen consumption in this already hypermetabolic patient group [9].
The goal of this study was to evaluate the metabolic and hemodynamic effects of high-dose GIK treatment after combined aortic valve replacement and coronary artery bypass grafting surgery and to study the effects of large glucose and insulin doses on systemic metabolism postoperatively.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Anesthesia
Anesthesia, perfusion, and ICU care were standardized. Anesthesia was induced with a fentanyl bolus and propofol and alfentanil infusions and maintained with propofol and alfentanil infusions and isoflurane. Muscle relaxation was achieved with pancuronium. All patients received dexamethason (Decadron, MSD, Espoo, Finland) 60 to 80 mg and aprotinin (Trasylol; Bayer, Leverkusen, Germany) 2 million IU intravenously after the induction of anesthesia and 2 million IU in the perfusion priming solution.
Cardiopulmonary Bypass
Cardiopulmonary bypass was carried out by using roller pumps with nonpulsatile flow and a membrane oxygenator, and moderate systemic hypothermia was targeted (31°C). The cardiopulmonary bypass circuit was primed with Ringer acetate (1,500 mL), 15% mannitol (250 mL), and heparin (75 mg). Pump flow was maintained at 2.4 L·min–1
·m–2, and perfusion pressure was kept at 50 to 70 mm Hg. After the initiation of cardiopulmonary bypass, hypothermic (25°C) blood cardioplegic solution was delivered continuously in a retrograde fashion through the coronary sinus and venous grafts whenever possible. If cardioplegia was discontinued for better visualization, the total duration of ischemia was recorded. The cardioplegic temperature was raised to 36°C at least 5 minutes before the removal of the aortic cross clamp.
Hemodynamic Support
All the patients were primarily weaned off cardiopulmonary bypass without inotropic or vasopressor agents, but dobutamine and noradrenaline were used as needed. During the ICU stay, the goals of hemodynamic support were to maintain the cardiac index above 2.2 L·min–1
·m–2 and the mixed venous oxygen saturation above 58%. These goals were achieved with dobutamine after an adequate preload, afterload, and heart rate optimization. Hypotension (mean arterial pressure below 65 mm Hg) was treated with noradrenaline. Bradycardia (heart rate below 65) was treated with temporary ventricular pacing if the cardiac index was not sufficient.
Intensive Care Unit Care
After arrival into the ICU, the patients were covered with a thermoblanket (Warm Touch; Mallinckrodt Medical, St. Louis, Missouri), and the peripheral (hallux) and central temperatures were recorded (SG catheter tip). Propofol (1 mg·kg–1
·h–1) and oxycodone (0.03·kg–1
·h–1) infusions were continued in the intensive care unit until the patient was considered ready for extubation. The criteria for extubation were central temperature above 36.0°C, adequate spontaneous ventilation, cooperation, stable hemodynamics, and absence of significant bleeding. The patients were transferred into a surgical ward in line with our normal practice, namely, when their vital signs were stabilized and they no longer needed ventilatory or hemodynamic support.
Experimental Protocol
The high-dose GIK therapy consisted of insulin (1·kg–1
·h–1; Insulin Actrapid; Novo Nordisk A/S, Bagsvaerd, Denmark) and glucose (0.45 g·kg–1
·h–1, 30% glucose solution 1,000 mL with 20 mmol KCL and 20 mmol Mg) infusions administered separately. Both were started after the induction of anesthesia, and the glucose infusion rate was adjusted according the blood glucose levels, targeted at 108 to 180 mg/dL. The blood glucose levels were controlled every 15 minutes in the GIK group during the first hour and after that every half hour until the end of the operation. The control group was started on 0.9 % saline (1.5 mL·kg–1
·h–1, 1,000 mL with 20 mmol KCL and Mg 20 mmol), and their blood glucose levels were checked once an hour during the operation, and 4 to 8 IU of short-acting insulin was administered intravenously whenever the blood glucose levels exceeded 180 mg/dL. After the admission into the ICU, arterial blood glucose levels were checked at 0, 1, 2, 3, 4, 6, 8, 10, 12, 16, 20, 24, and 28 hours in both groups. Withdrawal of insulin infusion was begun 6 hours after the admission into the ICU at a rate of 5 IU per hour, and the glucose infusion rate was decreased gradually according to the blood glucose levels.
Laboratory Data
Arterial blood glucose levels were measured with the ACCU-CHEK Sensor (Roche Diagnostics, Indianapolis, IN) whole-blood bedside strip test. Lactate (Rapidlab 865, Ciba Corning Diagnostics, Halstead, Essex, England) and free fatty acid (FFA [JC-2401 PC; Shimazu, Kyoto, Japan]) levels were measured after the admission into ICU, 6 hours after the admission, and on the first postoperative day. Mixed venous saturation levels were measured after the admission into ICU, 4 and 8 hours after admission, and on the first postoperative morning. The plasma potassium levels were measured at 2-hour intervals during the first 12 hours and at every fourth hour after that, until the 32nd hour. Plasma potassium concentrations were maintained within a range from 3.0 to 5.0 mmol/L, and supplemental potassium bolus doses were administered when necessary. Cardiac troponin-I (AxSYM system; Abbot Laboratories, Abbot Park, IL) levels were measured at 0 and 6 hours after the operation and on the first postoperative morning. All the other tests were performed using the routine laboratory methods.
Hemodynamic and Physiologic Parameters
Cardiac index was measured using a thermodilution method, and systemic vascular resistance (SVRI) and pulmonary vascular resistance indices were calculated using the standard formulas. Hemodynamic measurements were recorded before the induction of anesthesia, after perfusion, at 0, 4, and 8 hours after the ICU, and on the first postoperative morning. The mean and total doses and the length of administration of inotropic and vasopressor agents and the use of an intra-aortic balloon pump during the operation and the ICU stay were recorded. Atrial fibrillation and severe arrhythmias were recorded. Energy expenditure and oxygen consumption (VO2), carbon dioxide production (VCO2), and respiratory quotient (RQ = VO2/VCO2) were measured by indirect calorimetry by sidestream measurement of the expired and inspired gases (Deltatrac; GE Healthcare Technologies, Helsinki, Finland) during the time the patients were intubated or until the first postoperative morning.
Clinical Data
Time to extubation, ICU stay, and hospital stay were recorded for all patients. Postoperative morbidity (cardiac, renal, pulmonary, neurologic, gastroenterologic) was also recorded. Myocardial infarction was defined as a new Q wave, depression of R wave, or new regional wall motion abnormalities in echocardiography combined with a rise of serum cardiac troponin-I above 60 µg/L.
Statistical Analysis
Summary measurements are expressed as mean and SD or as median and 25th to 75th percentiles. The groups were compared by t test, or by Mann-Whitney U test when the t test assumption (approximate normality) was not met. Area under the curve was calculated for the average administration rate (µg·kg–1
·min–1) of dobutamine and noradrenaline over the operating theatre and ICU period up to the second postoperative morning. Analysis of variance for repeated measurements was utilized for repeatedly measured data. Two-sided p values are reported. The analyses were performed using SPSS (versions 10.0.7 and 12.0.1; SPSS, Chicago, Illinois).
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
|
Blood Glucose Levels
Blood glucose levels are presented in Figure 1a. Three hypoglycemic blood glucose level values were recorded for three different patients at one measurement point in the GIK group during the ICU stay (56, 61, and 67 mg/dL), but they were rapidly corrected by adjusting the rate of glucose infusion.
|
Hemodynamics
Cardiac indices and systemic vascular resistance indices (SVRI) are presented in Figure 2a and 2b. There were no remarkable differences in heart rate and cardiac filling pressures (central venous pressure, pulmonary artery wedge pressure). Thirteen patients (65%) in the GIK group and 17 (85%) in the control group were treated with dobutamine (p = 0.3), and the mean area under the curve was 4.6 (SD 8.1) in the GIK group and 21.8 (SD 30.2) in the control group (p = 0.014). Twenty patients in the GIK group and 17 (85%) of the controls received norepinephrine (p = 0.2), and the mean area under the curve was 360 (SD 568) in the GIK group and 405 (SD 660) in the control group, respectively (p = 0.7). Three patients (15%) in the GIK group and 12 (60%) in the control group required temporary pacing after perfusion (p = 0.008), and the corresponding figures during the ICU stay were 0 vs 6 (30%; p = 0.02).
|
Plasma Potassium and Phosphate Levels
Plasma potassium levels are presented in Figure 3. There were 3 patients with low potassium levels (P-K< 3.0 mmol/L) in the GIK group, and the values were corrected according to the protocol with supplemental potassium bolus doses.
|
Myocardial Injury and Biochemical Markers
Four patients (20%) in the GIK group and 2 (10%) in the control group had peak cardiac troponin-I values above 60 µg/L. There was 1 myocardial infarction in the GIK group and 2 in the control group. The incidence of atrial fibrillation did not differ significantly between the groups during the hospital stay (GIK 60% vs control 70%).
Indirect Calorimetry and Caloric Intake
The values of systemic metabolism measured by indirect calorimetry are presented in Table 3. The mean value for daily glucose uptake was 7.1 g·kg–1 (28 kcal·kg–1) (range, 4.2 to 11.3 g·kg–1), in the GIK group during and after the operation. The average glucose infusion rate during 24 hours after the initiation of GIK treatment varied from 2.9 to 7.9 mg· kg–1
·min–1.
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Lactate Levels
This study was started before Van Den Berghes and colleagues [5] demonstrated the importance of strict glycemic control during ICU stay, and the targeted blood glucose levels do not represent our current practice. Although increased systemic lactate concentrations are presented to be a marker of inadequate tissue perfusion, the most obvious reason for the faster normalization of lactate levels in the high-dose insulin group was the lower blood glucose levels. It is unlikely that the slower lactate clearance in the control group was due to systemic hypoperfusion, as no differencies with regard in mixed venous saturation levels were detected. Similar results were also demonstrated by Lazar and colleagues [10] in their GIK study of diabetic coronary artery bypass graft patients. There is some evidence that large amounts of lactate can be produced in inflammatory processes without evident tissue hypoxia by acceleration of aerobic glycolysis and, hence, increased pyruvate and lactate production [11, 12] On the other hand, one of the specific actions of insulin is the stimulation of glucose oxidation, and it is possible that the faster normalization of lactate levels in the GIK group was due to increased pyruvate dehydrogenase activity [13] or altered free fatty acid availability caused by insulin.
Hemodynamic Effects
Glucose-insulin-potassium has been shown to improve cardiac function and to decrease the need for inotropic support in several studies [14, 15], but the mechanism underlying the insulin-mediated increase in myocardial function has not been clearly defined. Also according to this study, dobutamine treatment could be discontinued sooner for the GIK-treated patients because of their better preserved cardiac indices after the operation. Insulin has also been shown to have dose-dependent vasodilatory effect on the systemic vasculature [16], but in this study, no significant differences were seen here in systemic or pulmonary vascular resistance indices, in the need for vasopressor medication or in cardiac filling pressures (central venous pressure, pulmonary artery wedge pressure). However, the better maintained cardiac indices and the lesser need for inotropic support might have contributed to the faster lactate clearance in the GIK group.
Free Fatty Acid and Potassium Levels
The activated neuroendocrine stress response and the effect of heparin have been demonstrated to lead to elevated levels of circulating FFAs, which have been shown to have deleterious effects on the postischemic myocardium [17] and to increase the occurrence of arrhythmias [18]. Insulin brings down the FFA levels and reduces myocardial FFA uptake, but even the profound suppression of FFAs in the GIK group did not lead to any differences in the incidence of atrial fibrillation between the groups (GIK 60% vs control 70%), unlike in the studies of Lazar and associates [10]. The glucose regimen that was used contained a moderately small amount of potassium (30% glucose solution 1,000 mL with 20 mmol KCL) to avoid rebound hyperkalemia after the withdrawal of insulin infusion, and this might have had some impact on the incidence of atrial fibrillation in the GIK group.
Myocardial Injury
There are several mechanisms by which GIK administration has been described to protect an ischemic and reperfused myocardium [19, 20], but in this study, unexpectedly, there were more patients in the GIK group with high cardiac troponin-I levels. Some myocardial damage may be inevitable during major cardiac procedures, and it remains unclear as to whether the more abundant myocardial enzyme release in the GIK group is a consequence of poorer myocardial protection or other events, such as increased glucose metabolism.
Glucose Uptake
Despite this, the high-dose GIK regimen is principally intended to overcome insulin resistance and to optimize the myocardial glucose uptake [2], as the major target organs for its metabolic effects are the liver and skeletal muscle. Patients scheduled for cardiac surgery are usually given 50 to 100 g (200 to 400 kcal) of glucose in electrolyte solutions perioperatively, but this therapy often requires large amounts of glucose to be infused to avoid hypoglycemia, even after the discontinuation of insulin infusion. In this study, the patients with GIK received an average glucose amount of 7.1 g·kg–1
·24 h–1 (28 kcal·kg–1
·day–1), which clearly exceeds the current recommendations for perioperative administration of carbohydrates [7]. However, we did not detect any adverse events associated with that. The average glucose infusion rate was near the values reported by Szabo and coworkers [21] (mean glucose uptake 4.2 mg·kg–1
·min–1) and Nilsson and associates [22] (peak glucose uptake 7.0 mg·kg–1
·min–1) with high-dose GIK. The differences in the glucose infusion rates between the patients probably reflect the different degrees of insulin resistance and stress response.
Metabolic Monitoring
In this study, the higher respiratory quotient levels in the GIK group obviously indicate increased glucose utilization and oxidation with no increase in energy and oxygen consumption, and not even CO2 production differed significantly between the groups during the first hours of ICU stay. Similar results were also obtained by Hiesmayr and coworkers [23], when they compared high-dose insulin (insulin 1.5 IU·kg–1
·h–1 with glucose 0.5 g·kg–1
·h–1) and low-dose dobutamine after coronary surgery and showed that a high insulin dose had no effect on systemic oxygen consumption and energy expenditure. However, the amount of CO2 produced when fuel is burned may be clinically important in patients with problems in ventilator weaning. The differences with regard to core temperatures after the admission into ICU were most probably associated with hypermetabolism and thermogenesis induced by the increased glucose oxidation [24].
There are several potential limitations to the current study, and the results should hence be interpreted with caution. The sample size was small, but the study was powered to the metabolic endpoints. The study was randomized and controlled but not blinded because of the hypoglycemia risk associated with the high insulin dose used. In conclusion, the GIK protocol was safe, but required strict control of the blood glucose levels, and it is hence difficult to administer outside an intensive care unit or an operating theater.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. Ellger, Y. Debaveye, I. Vanhorebeek, L. Langouche, A. Giulietti, E. Van Etten, P. Herijgers, C. Mathieu, and G. Van den Berghe Survival benefits of intensive insulin therapy in critical illness: impact of maintaining normoglycemia versus glycemia-independent actions of insulin. Diabetes, April 1, 2006; 55(4): 1096 - 1105. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
