| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Departments of Anesthesiology (H.B.v.W., C.J.Z., J.v.D.), Endocrinology and Metabolism (E.F.), Intensive Care Medicine (E.d.J.), and Cardiac Surgery (B.A.d.M.), and Laboratory of Endocrinology and Radiochemistry (E.E.), Academic Medical Center, University of Amsterdam, Meibergdreff 9, 1105 AZ Amsterdam, The Netherlands; and Department of Internal Medicine (E.W.C.M.v.D.), Free University Medical Center, De Boelelaan 1117, 1007 MB Amsterdam, The Netherlands
Address all correspondence and requests for reprints to: Harry B. van Wezel, M.D., Ph.D., Department of Anesthesia, Academic Medical Center, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands. E-mail: H.B.vanWezel{at}amc.uva.nl.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Methods: Forty-four nondiabetic patients, scheduled for elective CABG, were randomized to either a control group (n = 22) receiving standard care or to a clamp group (n = 22) receiving additionally a perioperative hyperinsulinemic (regular insulin at a fixed rate of 0.1 IU·kg–1·h–1) normoglycemic (plasma glucose between 3.0 and 6.0 mmol·liter–1) clamp during 26 h. We measured the endocrine response of the hypothalamus-pituitary-adrenal (HPA) axis, the sympathoadrenal axis, and glucagon, as well as plasma glucose and insulin at regular intervals from the induction of anesthesia at baseline through the end of the second postoperative day (POD).
Results: There were no differences in clinical outcome between the groups. In the control group, hyperglycemia developed at the end of surgery and remained present until the final measurement point on POD2, whereas plasma insulin levels remained unchanged until the morning of POD1. In the intervention group, normoglycemia was well maintained during the clamp, whereas insulin levels ranged between 600 and 800 pmol·liter–1. In both groups, plasma ACTH and cortisol increased from 6 h after discontinuation of cardiopulmonary bypass onward. However, during the clamp period, a marked reduction in the HPA axis response was found in the intervention group, as reflected by a 47% smaller increase in area under the curve in plasma ACTH (P = 0.035) and a 27% smaller increase in plasma cortisol (P = 0.002) compared with the control group. Compared with baseline, epinephrine and norepinephrine increased by the end of the clamp interval until POD2 in both groups. Surprisingly, the area under the curve of epinephrine levels was 47% higher (P = 0.026) after the clamp interval in the intervention group as compared with the control group.
Conclusion: A hyperinsulinemic normoglycemic clamp during CABG delays and attenuates the HPA axis response during the first 18 h of the myocardial reperfusion period, whereas after the clamp, plasma epinephrine is higher. The impact of delaying cortisol responses on clinical outcome of CABG remains to be elucidated.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Nygren et al. (8) reported that insulin and glucose infusions in patients undergoing hip replacement surgery reduce the cortisol and glucagon responses during the first 2 h postoperatively. However, the maximal responses of the hypothalamus-pituitary-adrenal (HPA) axis and the sympathoadrenal system in patients undergoing CABG occur 6–24 h postoperatively (9, 10, 11, 12). This implies that the effect of insulin on the postoperative neurohumoral stress response and on its course after discontinuation of insulin infusion during CABG is presently unknown.
We recently reported that a hyperinsulinemic normoglycemic clamp can be applied safely and effectively throughout the entire perioperative period (>26 h) in patients during CABG (4). The aim of the present study was to use this technique to quantify the effect of a hyperinsulinemic normoglycemic clamp on the dynamics of perioperative neurohumoral stress responses in nondiabetic patients undergoing elective CABG.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Insulin and glucose infusions
After the insertion of a central venous access port and baseline blood sampling, hyperinsulinemic normoglycemic clamping was started and continued throughout the period of cardiopulmonary bypass (CPB) until 24 h after release of the aortic cross clamp (this is the beginning of the myocardial reperfusion period). Regular insulin (Actrapid, NovoNordisk, Kopenhagen, Denmark) was infused continuously at a fixed rate of 0.1 IU·kg–1·h–1. A separate mixture of 30% glucose (Baxter-Clintec Benelux SA, Bruxelles, Belgium), 80 mmol·liter–1 potassium chloride, and 60 mmol·liter–1 phosphate was infused at a variable rate adjusted to maintain blood glucose levels within a target range of 4.0–5.5 mmol·liter–1. This rather narrow range was used to prevent unexpected overshoot hypoglycemia or hyperglycemia during the clamping period. A wider range of blood glucose levels between 3.0 and 6.0 mmol·liter–1 was considered normoglycemic.
The infusion of glucose was started at a rate of 0.5 ml·kg–1·h–1. Glucose samples were taken 10, 15, and 30 min after the start of glucose and insulin infusions to obtain the initial direction and degree of change of plasma glucose levels under hyperinsulinemic conditions. Adjustments to the glucose 30% infusion rate were made as follows: if plasma glucose was less than 3.0 mmol·liter–1, the infusion rate of glucose 30% was increased by 10 ml·h–1, and one or more additional boluses of 10 ml of glucose 30% were administered, depending on the absolute value of plasma glucose. Plasma glucose levels were checked 5–15 min later. If required, the infusion rate of glucose 30% was increased again by 5–10 ml·h–1. If plasma glucose was between 3.0 and 4.0 mmol·liter–1, the infusion rate of glucose 30% was increased by 5–10 ml·h–1. The plasma glucose concentration was checked 15–30 min later. If plasma glucose was within the target range of 4.0–5.5 mmol·liter–1, the infusion rate of glucose 30% remained unchanged, unless there was a difference of more than 0.5 mmol·liter–1 in the plasma glucose level, compared with the previous measurement. In the latter situation, the glucose infusion rate was increased or decreased by 5–10 ml·h–1, and plasma glucose levels were checked 60 min later.
If plasma glucose was greater than 5.5 mmol·liter–1, the infusion rate of glucose 30% was reduced by 5–10 ml·h–1. Plasma glucose levels were checked 60 min later.
Anesthetic management
Calcium channel blockers and long-acting nitrates were given until the evening before surgery. ß-Adrenoceptor blocking agents were continued until the morning of surgery. Lorazepam 2–3 mg was given for premedication 2 h before surgery.
Anesthesia was induced with sufentanil 3 µg·kg–1 (Sufenta; Janssen-Cilag, Tilburg, The Netherlands) and propofol 50–100 mg (Fresenius Kabi, Den Bosch, The Netherlands). Pancuronium bromide 0.1 mg·kg–1 (Pavulon; Organon, Oss, The Netherlands) was given for muscle relaxation. Morphine 20 mg was given as a slow bolus injection. Anesthesia was maintained with a continuous infusion of propofol 2–5 mg·kg–1·h–1. The lungs were ventilated with air/oxygen (FiO2 = 0.5). After the induction of anesthesia, a flow-directed pulmonary artery catheter (Edwards Lifesciences, Irvine, CA) was inserted into the right internal jugular vein. Dexamethasone and 
2-adrenoceptor agonists were not used in any of the participating patients.
CPB
The CPB system was primed with 1850 ml of priming liquid consisting of Hemaccel 1000 ml (Behring, Malburg, Germany), 500 ml of Ringer’s lactate, 200 ml of aprotinin (Bayer, Leverkusen, Germany), 100 ml of a mannitol 20% solution and 50 ml of 8.4% sodium bicarbonate. After systemic heparinization (300 U·kg–1; Leo Pharmaceutical Products, Weesp, The Netherlands), CPB was initiated with cannulas placed in the ascending aorta and right atrium. Activated clotting time was kept above 480 sec. At a temperature of 30–33 C, flow rate was kept at 2.2 liters·min–1·m–2. For myocardial protection, patients received high-potassium cold crystalloid cardiolplegia (1000 ml; St. Thomas type II, containing potassium 20 mmol, calcium 2 mmol, magnesium 16 mmol, chloride 203 mmol, and procain hydrochloride 273 mg, given at a temperature of 4 C) during aortic cross clamping. The hematocrit during CPB was maintained above 20%. After termination of CPB, heparin was antagonized with protamine hydrochloride (ICN Pharmaceuticals Holland BV, Zoetermeer, The Netherlands). Residual volume from the extracorporeal circuit was infused into the patient after a cell-saving process.
Postoperative intensive care management
After surgery, patients were admitted to the intensive care unit (ICU) and treated according to a standardized clinical protocol. Fluid administration consisted of NaCl 0.9% and hydroxyethyl starch 6%, molecular mass 200 kDa (Haes-Steril, Fresenius Kabi, Den Bosch, The Netherlands). Throughout the ICU stay, a continuous infusion of glucose 5% (partial parenteral nutrition) was given at a rate of 30 ml·h–1 (1.5 g·h–1) to all patients through a central venous line. All patients in both groups stayed relatively long in the ICU as a result of the study design with continuation of hyperinsulinemic normoglycemic clamping for 24 h in the reperfusion period.
Blood samples and measurement points
Samples in the operating room and ICU were taken from the radial artery catheter. On the ward, blood samples were collected through venapuncture. At the following timepoints blood samples were collected for the measurement of glucose, insulin, glucagon, ACTH, cortisol, epinephrine, norepinephrine, and cortisol binding globulin (CBG): 1) after induction of anesthesia and the insertion of the pulmonary artery catheter, but before the start of clamping (baseline); 2) 45-min after start of clamping, but before CPB, and before heparinization (before CPB); 3) immediately after release of the aortic cross-clamp (reperfusion); 4) 2 h after release of the aortic cross-clamp (2-h reperfusion); 5) 6 h after release of the aortic cross-clamp (6-h reperfusion); and 6–9) on the first and second POD, both in the morning between 0600 and 0800 h and in the evening between 1800 and 2000 h (POD1 morning, POD1 evening, POD2 morning, and POD2 evening, respectively). Cardiac enzymes (CK-MB) were sampled on arrival in the ICU and every 3 h thereafter, until it was clear that levels were past the peak level.
Insulin was determined by a luminescence enzyme immunoassay (Immulite, Diagnostic Products Corporation, Los Angeles, CA). The intra- and interassay coefficients of variation were 3–5% and 6–9%, respectively. The detection limit was 2 mU·liter–1. Glucagon was determined by RIA (Linco Research, St. Charles, MO). The intraassay coefficient of variation was 3–5%. The interassay coefficient of variation was 9–13%. The detection limit was 15 ng·ml–1. ACTH was determined by an immunoluminometric assay (Nichols Institute, Los Angeles, CA). The intra- and interassay coefficients of variation were 3.7–4.6% and 5.1–5.4%, respectively. The detection limit was 1 ng·liter–1. Cortisol was determined by a luminescence enzyme immunoassay (Immulite, Diagnostic Products Corporation). The intra- and interassay coefficients of variation were 3.6–6.4% and 4.7–9.0%, respectively. The detection limit was 30 nmol·liter–1.
Both epinephrine and norepinephrine were determined by an in-house HPLC method. Norepinephrine and epinephrine were selectively isolated by liquid-liquid extraction and derivatized to fluorescent components with 1,2-diphenylethylenediamine. The fluorescent derivatives were separated by reversed phase liquid chromatography and detected by scanning fluorescence detection. For norepinephrine, the intraassay coefficient of variation was 6–8%, and the interassay coefficient of variation was 7–10%. The detection limit was 0.05 nmol·liter–1. For epinephrine, the intraassay coefficient of variation was 6–8%, and the interassay coefficient of variation was 7–12%. The detection limit was 0.05 nmol·liter–1.
CBG was measured by RIA (BioSource Europe S.A., Nivelles, Belgium). The intraassay coefficient of variation was 4–5%. The interassay coefficient of variation was 9–12%. The detection limit was 10 mg/liter, and the reference range was 18–55 mg/liter.
Plasma glucose was measured with a Rapidlab 865 blood gas analyzer (Bayer Diagnostics AG, Leverkussen, Germany).
Lactate
Plasma samples for assessment of lactate were deproteinized with perchloric acid and subsequently analyzed using established enzymatic techniques.
Statistical analysis
Demographic data were reported as means (± SD) for continuous variables in case of approximately normal distribution and as counts (proportions) for categorical variables.
Outcome parameters (glucose and hormone concentrations) were reported as mean ± SEM.
A standard t test was performed to analyze differences at baseline.
One-way analysis for repeated measurements within group for time effects was performed on outcome parameters. When a significant time effect was found (P < 0.05), means at different time points were compared with baseline value (with Bonferroni correction).
A standard t test was performed to analyze differences in areas under the curve (AUCs) between the control group and the hyperinsulinemic normoglycemic clamp group during and after the period of clamping for the different outcome parameters. P values < 0.05 were considered statistically significant. Data analyses were performed with SPSS 11.5 (SPSS, Inc., Chicago, IL).
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
The characteristics of the 44 patients are summarized in Table 1
.
|
. There were no statistically significant differences in clinical outcome between the groups.
|
There were no differences between the groups in perioperative infusion requirement (including glucose 30% infusion in the clamp group), urine production, and blood loss. The total fluid balances between the groups on the day of surgery and on the first POD were not significantly different. No details of fluid balances were analyzed on the second POD.
Plasma glucose concentrations
Plasma glucose levels are presented in Fig. 1B.
|
The AUC of glucose concentrations in the clamp group was significantly lower (P < 0.0001) during, but not after, the clamping period.
Efficacy of hyperinsulinemic normoglycemic clamping
During the clamping period, 599 glucose measurements were performed. One hundred thirteen datapoints (18.8%) were higher than 6.0 mmol·liter–1; 16 datapoints (2.7%) were less than 3 mmol·liter–1, whereas 2 datapoints (0.3%) were less than 2.2 mmol·liter–1.
In the control group, 388 glucose measurements were performed between baseline and 24 h after the start of myocardial reperfusion on POD1.
A total of 271 datapoints (70%) were greater than 6.0 mmol·liter–1. There were no datapoints < 4.0 mmol·liter–1 in the control group.
Glucose requirement in the hyperinsulinemic normoglycemic clamp group
In the clamp group, the mean (SD) glucose requirement was 99 ± 16 mg·kg–1·h–1 after 45 min of clamping (before CPB). However, at reperfusion (60 min later), the mean glucose requirement to maintain normoglycemia increased (P < 0.05) to 180 ± 33 mg·kg–1·h–1 and remained significantly elevated at all measurement points during the clamping period.
Lactate
The mean (± SD) plasma lactate concentration at baseline was 0.8 ± 0.3 mmol·liter–1 in the clamp group and 0.7 ± 0.3 mmol·liter–1 in the control group (not significant). The AUCs of plasma lactate during the clamping period and after clamping were not significantly different compared with the control group. However, there was a trend toward higher plasma lactate levels at reperfusion, 2-h reperfusion, and 6-h reperfusion in the clamp group.
Insulin
In the control group, seven patients were treated with insulin 1–2 IU·h–1 because plasma glucose levels were at least 11.0 mmol·liter–1 during the evening or night after surgery.
In the clamp group, the mean (± SD) time interval between the start of insulin infusion and onset of CPB was 58 ± 9 min.
Plasma insulin levels are presented in Fig. 1A
.
At baseline, there were no differences in insulin levels between the control and hyperinsulinemic normoglycemic clamp groups. In the clamp group, insulin levels ranged from 600 to 800 pmol·liter–1. After discontinuation of insulin infusion, plasma insulin levels quickly decreased to similar concentrations as found in the control group. On the evening of POD2, the insulin concentration in the hyperinsulinemic normoglycemic clamp group was significantly higher than at baseline (P < 0.001). In the control group insulin levels were significantly elevated when compared with baseline on the morning of POD1 (P < 0.001) and on the morning and evening of POD2 (P < 0.001).
ACTH
ACTH levels are reported in Fig. 2A. There were no differences in serum ACTH levels at baseline. Compared with the control group, the AUC of plasma ACTH was 47% smaller (P = 0.03) in the clamp group during the clamping period. In both groups, plasma ACTH levels peaked at 6 h after reperfusion and returned to baseline values on the morning of POD1. After discontinuation of clamping, ACTH levels remained in the lower normal range in both groups.
|
Cortisol levels are reported in Fig. 2B. There were no differences in cortisol levels at baseline. During the clamp, the AUC of cortisol levels was 27% lower (P = 0.002) in the clamp group, compared the control group. Compared with baseline, cortisol increased (P < 0.001) at 6 h after reperfusion and remained significantly elevated until the evening of POD2 in both groups.
Norepinephrine
Norepinephrine levels are reported in Fig. 3A. There were no significant differences in norepinephrine levels between the groups at any time during the study.
|
There were no differences in AUC of norepinephrine concentrations between the groups at any time during the study.
Epinephrine
Epinephrine levels are reported in Fig. 3B. The AUC of epinephrine concentrations was 46% greater (P = 0.026) in the clamp group after the clamping period, but not during the clamping interval.
There were no significant differences between the groups in epinephrine levels at baseline. Compared with baseline values, epinephrine levels increased (P < 0.01) at the start of reperfusion (end of CPB) in the control group and remained significantly elevated until the morning of POD2. In the clamp group, epinephrine levels increased 6 h after the start of reperfusion (P < 0.01) until the morning of POD2.
Glucagon
Plasma glucagon levels are presented in Fig. 4. At baseline, there where no differences in glucagon levels between the groups. After the start of hyperinsulinemic normoglycemic clamping and after discontinuation of clamping, the AUCs of plasma glucagon values were lower in the clamp group (29%, P = 0.026; and 27%, P = 0.014, respectively).
|
CBG
There were no significant differences in plasma CBG levels at baseline.
Compared with baseline, the plasma concentration of CBG decreased significantly (P < 0.001) at reperfusion in both study groups and remained significantly reduced until 6 h after the start of reperfusion in the clamp group and until the morning of POD1 in the control group. There were no differences in AUCs between the groups in the cortisol/CBG ratio at any time during the study.
CK-MB
CK-MB levels measured during the first 24 h postoperatively were not indicative of myocardial infarction (cutoff value, 80 µg·liter–1) in any of the reported patients. There were no significant differences in postoperative CK-MB levels between the control and hyperinsulinemic normoglycemic clamp group. All measured mean group levels were less than 30 µg·liter–1.
Hemodynamic parameters
There were no differences between the groups in measured or calculated hemodynamic parameters obtained at baseline, 2-h reperfusion, 6-h reperfusion, or on the morning of POD1.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
The reported endocrinological findings are not associated with differences in clinical outcome parameters between the groups (Table 2). This is probably due to the design of the protocol. To avoid confounding factors, including the need for prolonged inotropic stimulation with exogenous catecholamines, sedation, and ventilation, which would affect the reliability of endocrine assessment, we included only low-risk patients with a minimal probability of periperative complications. In these selected patients, we did not expect to improve clinical outcome by hyperinsulinemic normoglycemic clamping.
Although little is known about the local processes involved in activation of hypothalamic CRH and ACTH release from the anterior pituitary in surgical patients, there is ample evidence that pain, fever, hypothermia, and hypovolemia all result in robust HPA axis activation associated with elevated cortisol levels (14, 15). Both very high and low plasma concentrations of cortisol have been associated with adverse outcome in critical illness and surgery (6, 16, 17, 18, 19). Therefore, appropriate modulation of the perioperative response of the HPA axis may be clinically relevant. At the present time, it is not possible to differentiate between a HPA axis response that is adequate to meet stress conditions and an unfavorable overshoot of the cortisol response. One might argue that the cortisol levels reported in the present study were very low at baseline and pre-CPB, which would be true in awake, healthy subjects. But for the interpretation of cortisol concentrations in patients during CABG, one should take into account that those measurements were preceded by the induction of deep surgical anesthesia associated with significant suppression of stress responses in all patients. Furthermore, CABG is considered as a rapidly changing stress model (6), with different levels of surgical, thermal, and hemodynamic stimuli in the presence of changing levels of consciousness due to slowly diminishing concentrations of anesthetic agents, resulting in dramatic shifts in stress responses. Absolute values of stress hormone concentrations should therefore be judged from this perspective. The present study was not designed to compare absolute values of stress responses but rather to compare differences between patients receiving a clamp and those receiving standard care.
The large increase in cortisol levels as found in our control patients postoperatively is known to induce insulin resistance and hyperglycemia, which has been shown to be an independent risk factor in CABG and acute myocardial infarction (AMI) (2, 3, 7). Acute hyperglycemia has been associated with activation of inflammatory responses, increased infarct size, and reduced ischemic preconditioning possibly via endothelial damage, increased platelet aggregation, and activation of plasmatic coagulation factors (20). In addition, high concentrations of cortisol may further enhance endothelial dysfunction (21) via activation of the vascular endothelin system (22) and inhibition of the nitric oxide synthase (NOS) isoforms inducible NOS and endothelial NOS (23, 24). Thus, delayed and attenuated plasma cortisol levels and normal plasma glucose levels as found in the clamp group may reduce the risk of endothelial dysfunction and coagulopathy. This may be especially beneficial at that particular stage of surgery, because coronary revascularization is associated with mechanical damage to the endothelium of coronary and mammarian arteries, saphenous veins, as well as pulmonary blood vessels, which may trigger ongoing serious endothelial damage in the presence of hypercoagulability. In addition, it is well established in isolated heart models that postischemic myocardium requires predominantly glucose as a metabolic substrate (25). In the presence of cortisol induced insulin resistance (and reduced endogenous insulin secretion), myocardial glucose uptake will be impaired, which may prolong and deteriorate myocardial function after ischemic episodes as in CABG or AMI. Reduced myocardial glucose uptake has been associated with decreased GLUT4 levels, greater ATP depletion, and increased ischemic damage in the chronically infarcted rat heart (26). In the hyperinsulinemic normoglycemic clamp group, the infusion rate of glucose was 75% higher after extracorporeal circulation (moderate myocardial ischemia) than during the first 45 min of clamping, leading to a maximal systemic glucose requirement of 180 mg·kg–1·h–1 under normoglycemic conditions at that time. If systemic glucose uptake and utilization reflect myocardial glucose uptake and breakdown in the clinical human situation, this observation may be an explanatory factor in the reported beneficial effects of GIK infusion strategies in CABG and critical illness (1, 2, 3). Additional fuels used by postischemic myocardium are free fatty acids and lactate, which may both be influenced by insulin infusion (27, 28). In our study groups, systemic lactate levels were high at reperfusion, with no significant difference in AUCs, although there was a trend toward higher lactate levels at that time in the clamp group. Unfortunately, our study was not designed to quantify the systemic or myocardial utilization of these substrates.
Before discontinuation of the insulin and glucose infusions in the clamp group, the cortisol concentration increased to values not different from those found in the control group, followed by an overshoot effect after the clamping interval. When the AUC of cortisol levels for the entire study period was calculated, there was no significant difference between the groups. This emphasizes that hyperinsulinemic normoglycemic clamping induces a delay and attenuation in the cortisol response during the early reperfusion period selectively, whereas total cortisol production is not significantly altered. Thus, GIK infusion strategies may be only useful during the first 6–18 h of the myocardial reperfusion period. Future studies will aim at the association between clamping induced dampening of the HPA axis response to CABG and clinical outcome. Because a certain degree of stress response by the body is absolutely crucial to survive, this approach carries the risk of inducing relative adrenal insufficiency (19), which may negatively affect outcome, especially in high-risk patients.
The persistent postoperative dissociation between plasma concentrations of ACTH and cortisol found in both study groups starting on the morning of POD1 and lasting until the final measurement point on POD2 has been described previously in patients undergoing CABG and receiving general anesthesia without insulin infusion (10). This phenomenon has been attributed to high serum levels of norepinephrine, because norepinephrine may stimulate both adrenal medulla and cortex (29). Indeed, compared with baseline values, serum norepinephrine was high in both study groups postoperatively (Fig. 3A).
We can only speculate about the mechanism behind the delayed and attenuated cortisol response during the early postoperative period in our patients receiving glucose and insulin. Because there was a trend toward a lower (P = 0.067) AUC of the ACTH/cortisol ratio in the clamp group during the total interval of hyperinsulinemic normoglycemic clamping (data not shown), it appears that a lower sensitivity of the adrenal cortex for ACTH can be excluded as a factor explaining the delayed and attenuated HPA axis response to CABG. Thus, this effect cannot be explained solely by a peripheral action of clamping but suggests an alteration at the central, possibly hypothalamic, level of the HPA axis.
A direct central effect of insulin is in keeping with striking recent animal experimental studies by Rosetti’s group (30, 31, 32, 33). They demonstrated that adequate hypothalamic nutrient sensing and the hypothalamic action of insulin are required for the full inhibitory effect of systemic insulin on endogenous hepatic glucose production in rats. This process requires an intact insulin signaling cascade, activation of hypothalamic ATP-sensitive potassium channels, and an intact vagal system to convey central input to the liver (30, 31, 32, 33). Disruption of this complex regulation system of glucose homeostasis leads to hyperglycemia. Although the role of central hypothalamic insulin signaling as a determinant of central HPA axis regulation is presently unknown, it is possible that the supraphysiological concentrations of insulin in the hyperinsulinemic normoglycemic clamp group may have corrected this disruption, resulting in lower ACTH and cortisol concentrations. Likewise, it is not known whether these novel neuroendocrine pathways may explain the observed changes in plasma catecholamine levels (epinephrine) found in the present study. There is evidence from animal experimental studies that insulin signaling can increase the central sympathoadrenal response to hypoglycemia (34). We have no indication that (relative) hypoglycemia is responsible for the higher epinephrine levels and protracted HPA axis response after the clamping interval in our patients. Even after exclusion of all patients with two or more plasma glucose samples less than 4.0 mmol·liter–1 during clamping, the stress response patterns remained unchanged.
In both of our study groups, catecholamine levels were high at the start of reperfusion and during the postoperative period. It is not clear whether the higher epinephrine levels found in our patients after the clamp play a role in the improved outcome associated with the use of insulin and glucose infusion strategies in CABG. However, elevated endogenous epinephrine levels, in combination with the recently reported increase in expression of ß-1 adrenoreceptor mRNA by GIK in patients undergoing CABG (35) may explain the reported lower inotropic requirement associated with GIK infusions (2).
Finally, it is not surprising that the postoperative glucagon response was lower in our patients treated with exogenous insulin, considering the negative feedback effect of insulin on pancreatic glucagon production. This led to a lower glucagon/insulin ratio, perhaps inducing less catabolic activity during the stressful (post) surgical conditions (36). However, in addition to its metabolic effect, glucagon also has potent positive inotropic properties, which are mainly induced by its degradation product miniglucagon (37). Therefore, the significant reduction of glucagon levels in the hyperinsulinemic normoglycemic clamp group may not necessarily represent an advantage in patients with reduced myocardial contractility as in CABG or AMI.
Summarizing, the results of the present study indicate that a hyperinsulinemic normoglycemic clamp has a significant and differential impact on the neurohumoral stress response to CABG. The impact of delaying and attenuating HPA axis activation on clinical outcome of CABG remains to be elucidated.
|
Footnotes |
|---|
H.B.v.W., C.J.Z., E.d.J., E.W.C.M.v.D., J.v.D., E.E., B.A.d.M., and E.F. have nothing to declare.
First Published Online August 8, 2006
Abbreviations: AMI, Acute myocardial infarction; AUC, area under the curve; CABG, coronary artery bypass grafting; CBG, cortisol-binding globulin; CK-MB, cardiac enzymes; CPB, cardiopulmonary bypass; GIK, glucose-insulin-potassium; HPA, hypothalamus-pituitary-adrenal; ICU, intensive care unit; NOS, nitric oxide synthase; POD, postoperative day.
Received June 2, 2006.
Accepted July 28, 2006.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  R. S. Wiener, D. C. Wiener, and R. J. Larson Benefits and Risks of Tight Glucose Control in Critically Ill Adults: A Meta-analysis JAMA, August 27, 2008; 300(8): 933 - 944. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Zuurbier, F. J. Hoek, J. van Dijk, N. G. Abeling, J. C. M. Meijers, J. H. M. Levels, E. de Jonge, B. A. de Mol, and H. B. Van Wezel Perioperative hyperinsulinaemic normoglycaemic clamp causes hypolipidaemia after coronary artery surgery Br. J. Anaesth., April 1, 2008; 100(4): 442 - 450. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. D. Evans and Y. Niu Hypolipidaemic effects of high-dose insulin therapy Br. J. Anaesth., April 1, 2008; 100(4): 429 - 433. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
, P < 0.0001 within group.
