- Email: [email protected]
The effect of perioperative glucose control on postoperative insulin resistance
Clinical Nutrition 31 (2012) 676e681
Contents lists available at SciVerse ScienceDirect
Clinical Nutrition journal homepage: http://www.elsevier.com/locate/clnu
Original article
The effect of perioperative glucose control on postoperative insulin resistanceq Christina Blixt a, b, *, Christian Ahlstedt a, b, Olle Ljungqvist c, Bengt Isaksson b, d, Sigridur Kalman a, b, Olav Rooyackers a, b a
Dept of Anaesthesia and Intensive Care, Karolinska University Hospital, Huddinge, Sweden Dept of Clinical Science, Intervention and Technology (CLINTEC), Karolinska Institutet, Huddinge, Sweden c Dept of Surgery, Örebro University Hospital, Örebro, Sweden d Division of Surgery, Karolinska University Hospital, Huddinge, Sweden b
a r t i c l e i n f o
s u m m a r y
Article history: Received 22 November 2011 Accepted 22 February 2012
Background & aims: Postoperative insulin resistance and the consequent hyperglycemia affects clinical outcome. Insulin sensitivity may be modulated by preoperative nutrition, adequate pain management and minimal invasive surgery. This study aims to disclose the impact of perioperative glucose control on postoperative insulin resistance. Methods: Twenty patients scheduled for elective open hepatectomy were enrolled in this prospective, randomized study. In the treatment group (n ¼ 9) insulin was administered intravenously to keep blood glucose between 6 and 8 mmol/l during surgery. The control group (n ¼ 8) received insulin if blood glucose >14 mmol/l. Insulin sensitivity was measured by a hyperinsulinemic normoglycemic clamp (0.8 mU/kg/min), performed on all patients both on the day before surgery and immediately postoperatively. Plasma cortisol, insulin and C-peptide were measured. Results: There was a significant difference in mean glucose value during surgery. In the control group 8.8 mmol/l (SD 1.5) vs. 6.9 mmol/l (SD 0.4) in the treated group, p ¼ 0.003. In the control group insulin sensitivity decreased to 21.9% 16.2% of the preoperative value and in the insulin treated group to 46.8 15.5%, p < 0.005. Insulin levels were significantly higher in the treatment group as well as consequently lower C-peptide levels. Conclusions: This trial revealed a significant difference in postoperative insulin resistance in the group treated with insulin during surgery. Ó 2012 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
Keywords: Liver resection Hyperinsulinemic normoglycemic clamp Abdominal surgery Cortisol Stress hormones
1. Introduction Hyperglycemia is often observed in surgical and critically ill patients. The changes that are seen are similar to that of a patient with type 2 diabetes and are the result of a transient insulin resistance, characterized by enhanced hepatic gluconeogenesis and glycogenolysis and a compromised peripheral insulin-dependent glucose uptake.1 Hyperglycemia is associated with an increased morbidity and mortality both in surgical and critically ill patients. Perioperative hyperglycemia has been associated with a higher risk for morbidity such as infections, myocardial infarction, acute
Abbreviations: M, total glucose disposal, mg/kg/min; M%, difference between pre-and postoperative M-value expressed as percentage. q Poster presentation, ESPEN conference, Nice, 2010. * Corresponding author. Dept of Anaesthesia and Intensive Care, Karolinska University Hospital, Sweden. Tel.: þ46 8 58580000; fax: þ46 8 7795424. E-mail addresses: [email protected], [email protected] (C. Blixt).
renal failure, and neurological damage.2e4 The significance of hyperglycemia during critical illness and especially the relevance of normalizing glucose levels has been a hot topic since the studies from van den Berghe and colleagues,5,6 where a significantly reduced risks of both morbidity and mortality in ICU-patients was shown when glucose was normalized with intensive insulin treatment. Post-hoc analysis of this latter study also showed that normoglycemia rather than the insulin dose given was responsible for these positive effects.7 However, several later studies have not been able to reproduce these results3,8,9 and the issue is still heavily debated. Studies in the perioperative period are less abundant.10e12 Whether controlling perioperative glucose levels have an effect on postoperative outcome is not entirely clear. Studies investigating this have mostly been focused on cardiac surgery with diabetic subjects.8,13e16 From these studies it was suggested that poor perioperative metabolic control is associated with increased risk of postoperative complication such as wound infection and cardiac events. Even effects on mortality have been reported.17
0261-5614/$ e see front matter Ó 2012 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved. doi:10.1016/j.clnu.2012.02.011
C. Blixt et al. / Clinical Nutrition 31 (2012) 676e681
Our thesis in this prospective study was that perioperative metabolic management in the form of normalizing glucose levels can be advantageous also in abdominal surgery. Postoperative insulin resistance was used as the outcome measure since we know that it is related to clinical outcome from other studies.18,19 We hypothesized that controlling glucose level between 6 and 8 mmol/ l during surgery may lower the degree of postoperative insulin resistance in comparison to a level >10 mmol/l. We investigated the effect in major open abdominal surgery. 2. Methods 2.1. Patients Patients scheduled for elective partial hepatectomy due to neoplastic condition were enrolled in this randomized prospective study. Twenty patients studied per protocol were planned. Inclusion criteria were patients scheduled for open laparotomy, over 18 years of age and with no contraindications for epidural anaesthesia. Patients with known diabetes mellitus or medication with corticosteroids were excluded. The study design was reviewed and approved by the regional ethics committee in Stockholm. The subjects were informed about the purpose and the nature of the study and written informed consent was given before inclusion. 2.2. Study design Insulin sensitivity was measured by a normoglycemic hyperinsulinemic clamp, as described below, the day before surgery and immediately after surgery. Patients were randomized to perioperative insulin treatment (with a glucose target of 6e8 mmol/l) or to a control group. The control group was treated accordingly to current clinical practice. The day before scheduled operation, the patient arrived to the hospital at 1 pm, fasted for a minimum of 4 h. A normoglycemic hyperinsulinemic clamp was performed as described in detail below. At the day of surgery the patient arrived to the operating theatre at 8 am after fasting since midnight. The patients received oxycodone (OxycontinÒ) as premedication, intra- and postoperative analgesia was managed with a thoracic epidural catheter that was inserted at Th 6e9. After a test dose with Bupivacain epinephrine 5 mg/ml (3e5 ml) (Marcain adrenalinÒ, Astra Zeneca) and an epidural bolus dose of fentanyl (50 mg) (Fentanyl, BraunÒ), an epidural infusion containing Bupivacain 1 mg/ml, epinephrine 2 mg/ml and fentanyl 2 mg/ml (12e15 ml/h) was started. General anaesthesia was induced with Propofol-Lipuro (BraunÒ) and fentanyl and maintained with Sevoflurane (Sevoran, AbbottÒ). Atracurium (HamelnÒ) was used for muscle relaxation. All antibiotics given were diluted in sodium chloride or sterile water. All patients received a central venous catheter and an arterial catheter which were inserted after induction of anaesthesia. The blood samples for the study were taken from the arterial line. Continuous infusion of glucose 25 mg/ml, 1 ml/kg/h, started as soon as the central venous line was inserted, following the clinical routine at the hospital. Glucose was measured every 10 min using a bedside glucose monitor (Hemocue Glucose 201þÒ, HemocueAB, Ängelholm, Sweden). In addition, hourly plasma samples were obtained and stored at 80 C for later plasma glucose analyses in the laboratory. Before induction the patient was randomized, with a sealed, opaque envelope, either to the insulin treatment group or to the control group. When insulin treatment was required, the infusion (50 IU ActrapidÒ in 50 ml saline ¼ 1 IU/ml) was started. Insulin was given either as a peripheral infusion or a central infusion as soon as the central catheter was inserted. An initial glucose level higher
677
than 8 mmol/ml, or gradually increasing values higher than 7 mmol/l in two following samples, was considered an indication for insulin treatment. The infusion was started at 2 IU/h and then adjusted according to the current glucose level. In the control group glucose was targeted at 10e14 mmol/l. Insulin was only administered (by infusion or by intermittent boluses) when glucose levels exceeded 14 mmol/l. After surgery the patient was transferred to the postoperative ward for observation over the next 12e24 h. The infusions stayed unchanged until the postoperative hyperinsulinemic normoglycemic clamp was performed. The postoperative clamp was started as soon as the epidural anaesthetic effect could be assessed and the patient appeared circulatory stable, i.e. 60e120 min after extubation. Samples for plasma cortisol, insulin and C-peptide analyses were obtained at the start and the end of the clamping procedures and at the start of anaesthesia, at the start and at the end of the liver resection phase. These samples taken in pre-chilled EDTA tubes, centrifuged within 30 min (2700 rpm/min, 10 min in 4 C, Universal 32 R Hettich ZentrifugenÒ, Andreas Hettich GmbH & Co. KG, Tuttlingen, Germany) and stored at 80 C until analyses. Data for circulation, saturation, blood losses, blood transfusion, extension and time of surgery were collected. Also the rate of glucose-, norepinephrine- and insulin infusions was recorded. Crystalloids (Ringer-AcetateÒ 2 ml/kg) and colloids (Hydroxyethyl-starch, VoluvenÒ, 2 ml/kg) were given as intraoperative fluid replacement. Blood loss during surgery was replaced with crystalloids, colloids and/or blood products. 2.3. Hyperinsulinemic normoglycemic clamp Two peripheral vein catheters were inserted for infusions and blood sampling. Two plasma glucose values 10 and 5 min before the start of the clamp were used as a baseline mean value. During the clamp, plasma glucose was measured every 5 min. The clamp was initiated by a bolus of insulin (1.1 U/m2) and then maintained by a constant insulin infusion (0.8 mU/kg/min) and a simultaneous variable infusion of glucose (200 mg/ml) to maintain normoglycemia. Albumin (400 mg) and potassium (32 mmol) were added to the insulin infusion. The goal level for steady state was set to 0.5 mmol/ml of the baseline mean value. Total glucose disposal during hyperinsulinemic clamp is expressed as M (mg/ kg/min). A high M-value represents a high insulin sensitivity. For safety reason, potassium was measured before and after the clamp using a venous blood sample analyzed on a blood-gas analyzer (ABL 800 Flex, Radiometer, Denmark). The clamp was repeated at the postoperative ward approximately 1 h after arrival. The preoperative glucose level was used as a target. For practical reasons arterial samples were drawn in the postoperative clamp. At the end of the 2 h clamp the insulin infusion was interrupted and thereafter the glucose infusion continued. After 30 min the infusion of glucose was stopped when a control sample showed normoglycemia. The clamp was continued for 2 h. Steady state condition was assumed to be obtained after 60 min. Steady states for both the glucose infusion and glucose levels during the second hour were evaluated blinded by a person not involved in performing the clamps and unaware of the randomization. A period of at least 30 min with steady state for both glucose infusion and glucose levels was identified by this person and used for calculations. Clamps with no steady state during at least 30 min were excluded. This validation was done before any group calculation was performed. The amount of glucose given during the steady state period was used to calculate a mean M-value for whole body sensitivity to insulin (mg glucose/kg/min).
678
C. Blixt et al. / Clinical Nutrition 31 (2012) 676e681
2.4. Analysis Plasma glucose was measured with a bedside glucose analyzer (Hemocue Glucose 201þÒ, HemocueAB, Ängelholm, Sweden) both during surgery and clamping. Plasma glucose was analyzed on an automatic analyzer (Konelab 20, Thermo Scientific, Jönköping, Sweden) using a GOD-POD analysis (Thermo Fisher Scientific, Vantaa, Finland). Plasma cortisol, insulin and C-peptide were analyzed using an ELISA-based standard analyzing kit (IMMULITEÒ 1000 Immunoassay System, SiemensÒ, Illinois, USA). 2.5. Statistics Ten patients in each group were analyzed per protocol. Two patients in the treatment group and one patient in the control group were excluded due to suboptimal insulin clamping, assessed blinded. Both Student’s T-test and ANOVA were used to analyze the glucose and insulin sensitivity data. For the hormone analyses several values were out of range and therefore non-parametric KruskaleWallis ANOVA was used for comparison between the two groups.
4. Discussion
Between May 2008 and January 2010, 22 patients scheduled for elective open hepatectomy were included and 10 þ 10 patients were studied per protocol. Two patients were excluded in the intraoperative period due to unexpected corticosteroid treatment. Ten patients were planned in each group. Due to poor quality of the hyperinsulinemic normoglycemic clamping (i.e. unable to reach steady state) another three patients were excluded (assess blinded), one patient in the treatment group and two patients in the control group. The two cohorts were statistically similar, there were no difference in BMI, age, operation time, blood loss or blood products given (see Table 1). The indication for surgery was metastasized liver cancer for all patients in the treatment group (9/9). In the control group 6/8 had metastasized liver cancer, one patient was proven not to have cancer after the microscopic evaluation of the biopsy and one patient had primary liver cancer. 4 out of 9 patients in the treatment group had insulin infusion started after induction of surgery. No patient received insulin treatment prior to induction of anaesthesia. During surgery 9 U of insulin was given (mean value, range 4e25 U). Table 1 Distribution of baseline characteristics. Treated (n ¼ 9)
Control (n ¼ 8)
6/3
8/0
Mean SD
Mean SD
p-value
Age (yrs) BMI (kg/m2) Comorbidity IHD Hypertonia Lung Kidney Blood loss (ml) Blood products RPB (units) Op time (min) Hepatectomy > 2 segments Hepatectomy < 2 segments
64 4.3 25.7 4.8
68 9.9 27.5 2.8
0.35 0.37
1 1 0 0 1219 1103 1.9 214 67 5 4
0 4 0 0 1303 961 2.1 242 70 4 4
0.87
M preop (mg/kg/min) M postop (mg/kg/min) M% difference pre-post
6.94 2.9 3.32 1.7 46.8 15.5
7.01 2.5 1.63 1.7 21.9 16.2
0.94 0.06 <0.005
0.4
This randomized trial showed that continuous glucose control during open liver resection maintains higher insulin sensitivity immediately after surgery. The majority of studies on perioperative glucose control have been focused on cardiovascular surgery and predominantly diabetic subjects. The research has shown a positive effect on the risk of developing myocardial infarction in the postoperative period. The present study, on major upper abdominal surgery, was not powered to show any impact on postoperative morbidity or mortality. Nevertheless, it revealed an effect of glucose control on postoperative insulin resistance. We used a hyperinsulinemic normoglycemic clamp to evaluate insulin sensitivity, which is regarded as the golden standard.20 In the group where glucose levels were kept at a level between 6 and 8 mmol/l the postoperative insulin resistance was significantly less. Small but significant differences in plasma levels of insulin, but primarily C-peptide were seen. This latter finding is expected, since the endogenous insulin production is affected during exogenous insulin infusion. Perioperative glucose
20
Blood glucose (mmol/l)
3. Results
Gender (m/f)
All patients received norepinephrine during surgery at different rates (approximately 0.02e0.1 mg/kg/min). After surgery 5/8 in the control group received norepinephrine (0.02e0.1 mg/kg/min) and in the treatment group 5/9 received norepinephrine (0.02e0.05 mg/ kg/min). Glucose levels during surgery were significantly different between the two groups (see Fig. 1), with mean glucose values in the control group of 8.8 mmol/l 1.5 (range 6.4e11.6) and in the treatment group of 6.9 0.4 (range 6.3e7.5) (p ¼ 0.003; t-test). In the control group the preoperative M decreased from 7.0 2.3 to 1.7 1.7 whereas in the treated group the preoperative M decreased from 7.2 2.0 to 3.4 1.6 (p ¼ 0.056; ANOVA). When the postoperative insulin sensitivity was expressed as percentage of preoperative value, the control group retained 21.9 16.2% and the treated group 46.8 15.5% (p < 0.005; t-test) of the preoperative value (Fig. 2). The circulating cortisol levels increased during surgery, but this increase seemed to be less severe in the treated group (p ¼ 0.06; KruskaleWallis, Table 2). As expected, the treatment group had a higher mean insulin level during surgery as well as postoperatively while the mean C-peptide level diminished under the same period (p ¼ 0.02; KruskaleWallis, Table 2).
control treatment
15
10
5
0
Operating time (min) Fig. 1. Perioperative glucose measurement with individual values presented. The control group had significantly higher glucose levels during surgery. The control group had a mean glucose value of 8.8 mmol/l 1.5 (range 6.4e11.6) compared to the treatment group which had a mean glucose value of 6.9 0.4 (range 6.3e7.5) (p ¼ 0.003; t-test). (D ¼ control group, - ¼ treatment group).
C. Blixt et al. / Clinical Nutrition 31 (2012) 676e681
100 80
M%
60
40 20
0
treatment group
control group
Fig. 2. Percentage of retained insulin sensitivity post-surgery (M%), individual and mean values are presented. The treatment group kept 46.8 15.5% (SD) compared to the control group which retained 21.9 16.2% (SD) of the preoperative value, (p < 0.005; t-test).
control also affected stress hormones during surgery. Our findings did not show a statistical difference in cortisol levels in the treatment group, but merely a trend to a lower mean cortisol level during surgery was observed. This pilot study aimed to reveal any difference in insulin resistance due to glucose control during major upper abdominal surgery. The patients have been subjected to liver surgery and were non-diabetics. To our knowledge, there are a few studies on major upper abdominal surgery with the focus on glucose control and insulin resistance, but they are mostly from open cholecystectomies in the 1990’s.21e25 Insulin resistance has been shown to have a negative effect on patients’ outcome. A large descriptive study on the impact of insulin resistance showed that postoperative complications were reduced for every 1 mg/kg/min in decrease in insulin sensitivity,26 making the small difference observed in our study clinically relevant. Reduced insulin resistance can be achieved by minimizing the surgical trauma, giving adequate pain relief and by shortening the fasting time. All these limit the sympathetic stress response. Laparoscopic colonic surgery has been shown to have a positive
Table 2 Stress hormone levels in the two groups. Treated Mean SD Cortisol (mg/Dl) Preoperative clamp-before Preoperative clamp-after Start operation Pre resection Post resection Postoperative clamp-before Postoperative clamp-after Insulin (mIU/mL) Preoperative clamp-before Preoperative clamp-after Start operation Pre resection Post resection Postoperative clamp-before Postoperative clamp-after C-peptide (ng/mL) Preoperative clamp-before Preoperative clamp-after Start operation Pre resection Post resection Postoperative clamp-before Postoperative clamp-after
Control Mean SD
12.8 8.59 11.42 24.29 24.15 27.15 20.79
6.64 1.9 2.95 12.36 13.76 13.94 8.23
11.9 9.16 14.67 28.85 32.88 36.01 33.19
4.95 4.59 10.99 13.15 11.2 11.68 14.83
10.65 106.05 9.98 50.44 47.06 115.69 166.31
11.62 33.67 6.02 35.59 52.47 153.46 56.23
14.16 137.86 13.94 23.58 32 60.01 162.21
14.3 47.64 9.22 24.82 33.46 73.42 80.74
2.93 2.35 2.34 1.27 1.09 0.96 0.46
1.88 1.69 1.2 0.93 1.05 0.97 0.57
3.04 2.81 3.31 4.24 4.84 3.84 1.54
1.66 1.56 1.21 2.16 3.59 2.28 1.53
679
effect on insulin sensitivity.27,28 In our case we chose open laparotomy, since it is the most common approach in our hospital and allows including patients with larger hepatic resection and more complications that could be related to insulin resistance. Insulin resistance can also be reduced by adequate pain regime perioperatively,29 in our case with epidural anaesthesia. Lattermann and Thorell demonstrated that a thoracic epidural lowered insulin resistance by reducing levels of glucagon and cortisol and thereby lower the endogenous glucose production.30,31 All patients in our study received a thoracic epidural. It was activated during operation and assessed postoperatively. Also, insulin and glucose infusion given prior and during surgery may have an effect on insulin resistance by inducing an enhanced insulin action, maintained during surgery.25,32 These findings have been reproduced by giving a preoperative oral carbohydrate loading 2e3 h before surgery.33,34 Patients included in the present study were all fasted over night, which is the standard routine in the hospital. We chose to keep that routine even if longer fasting time is known to alter glucose metabolism. Perioperative glucose infusions are routine in our hospital, this may have contributed to the perioperative hyperglycemia reported in this study and may have aggravated the postoperative insulin resistance. The concept of tight glucose control has been widely adapted in the ICU setting since the Leuven studies,6 where a remarkable impact on both mortality and morbidity was demonstrated. In the following study in 2006, the effect on mortality was not so clear and a slightly higher risk for hypoglycemia was seen.5 That latter study was conducted in a medical ICU, while the first one was set in a surgical ICU with primarily cardiothoracic patients. The limitation of intensive insulin treatment has been heavily debated. Especially the risk of hypoglycemia has been in focus since the NICE-SUGAR study was presented.9 Hyperglycemia induces a range of metabolic changes in several defense system, for example a decrease in immunological function, enhanced immune factors, endothelial dysfunction, and disturbances in the coagulation system.17 The result is an increased risk for infections, cardiac, neurological and renal damage. In a retrospective study on patients undergoing carotid endarterectomy by McGirt, all glucose measurements over a ten-year period were reviewed. Hyperglycaemic patients, with a preoperative glucose value of >11 mmol/l, had a 4.3 and 2.8 times higher risk of having a myocardial infarction or stroke event respectively.4 Diabetics are generally more prone to postoperative complications. However, studies have demonstrated a higher mortality and morbidity among non-diabetic patients exposed to hyperglycemia.35,36 It seems that acute hyperglycemia is one factor in the development of postoperative complications. Hyperglycemia in neurosurgical patients is known to cause increased infarct size, vasospasm and may compromise recovery.11,37 Ouattara reported an association to malignant arrhythmia, need for longer vasopressor support and ventilation as well as neurological and renal impairment in patients with poor glycaemic control subjected to cardiovascular surgery.15 These findings have been reproduced in a prospective study by Subramaniam in 2009 where patients reaching a glucose target of 5e8 mmol/l had less risk of postoperative myocardial infarction.38 The liver is closely involved in glucose metabolism. In postoperative insulin resistance there is a change in both peripheral glucose uptake and hepatic glycogenolysis and gluconeogenesis. Theoretically there is a risk for hypoglycemia if a larger part of the liver is removed during surgery. Liver failure has been demonstrated to be a factor in developing hypoglycemia.39 However, recent studies performed on liver surgery, where insulin treatment protocols have been used, have had few recordings of severe hypoglycemia.40
680
C. Blixt et al. / Clinical Nutrition 31 (2012) 676e681
Where severe hypoglycemia has been reported a lower glucose target has been evaluated, in one study 3.6e6.1 mmol/l.41 In our study all patients had rising glucose values during surgery. In our study we chose a glucose target of 6e8 mmol/l, a more liberal target range than discussed in several studies. During surgery the effect of insulin might be impaired due a neuroendocrine stress response. The counter-acting hormones such as glucagon, cortisol and catecholamines induce changes in the lipid-, protein and glucose metabolism as well as the immune system.17,42 The hormones are secreted to supply substrates to the organism. The liver is the primary target, resulting in increased gluconeogenesis and glycogenolysis.1 But also peripheral tissues are affected, where the stress hormones inhibit glucose uptake. Due to the hepatic insulin resistance during stress, circulating cortisol aggravates hyperglycemia despite hyperinsulinemia.1 Cortisol has been demonstrated to be a stronger predictor of stress induced insulin resistance compared to some cytokines.43 But also a direct effect of insulin on cortisol has been demonstrated during surgery and critical illness.32,44 Our results do not explain the underlying mechanisms of insulin sensitivity. The trial was designed as a study to elucidate the effect of perioperative glucose control. Obviously a more thorough protocol, using labelled glucose to measure glucose turnover rates, is required to further elucidate the metabolic mechanisms. In our study we see a trend of rising cortisol levels during and after surgery in both groups. However, the mean cortisol level in the treatment group is somewhat lower, though this difference is not statistical significant. These findings are similar to those of van Wesel,13 where cortisol levels were clearly affected during and after clamping in CABG-patients. One reason for not finding a statistical difference in our study is probably due to the small groups. 5. Conclusions We showed an impact of intraoperative glucose control on postoperative insulin resistance. The treatment group retained 46.8 15.5% of their preoperative insulin sensitivity, and the control group 21.9% 16.2%, (p < 0.005). We found a trend to a mean lower cortisol level in the treatment group, though not statistical significant. The treatment group was kept between 6 and 8 mmol/l during surgery using insulin infusion. Statement of authorship CB, CA carried out the design of study, collected and analyzed data and drafted the manuscript. OL contributed by interpretation of data, and drafting the manuscript. BI and SK participated in the design of the study and coordination and helped to draft the manuscript. OR contributed by the design of the study, data interpretation and analyzing, statistical analyses and drafting the manuscript. All authors have read and approved the final manuscript. Conflict of interest OL owns a patent and receives royalties for a carbohydrate rich drink used in studies referred to in the paper. The patent is licensed to Nutricia who makes and sells a product based on it. None of the other authors have any conflict of interest to declare. Acknowledgements Grants/funding: The study was sponsored by the Swedish Medical Research Council (no: 14244), Karolinska Institute Funds and the Stockholm County Council.
A special thanks to the skilled nursing assistance from Gunilla Herman, Viveka Gustafsson, Christina Hebert and Eva Skog Nejman, Dept of Anesthesia and Intensive Care, KarolinskaUniversity Hospital, Huddinge, Sweden. The staff at Endocrinology Laboratory, Dept of Medicine, KarolinskaUniversity Hospital, Huddinge, Sweden is gratefully acknowledged. References 1. Lipshutz AK, Gropper MA. Perioperative glycemic control: an evidence-based review. Anesthesiology 2009;110:408e21. 2. Finney SJ, Zekveld C, Elia A, Evans TW. Glucose control and mortality in critically ill patients. JAMA 2003;290:2041e7. 3. Krinsley JS. Effect of an intensive glucose management protocol on the mortality of critically ill adult patients. Mayo Clin Proc 2004;79:992e1000. 4. McGirt MJ, Woodworth GF, Brooke BS, Coon AL, Jain S, Buck D, et al. Hyperglycemia independently increases the risk of perioperative stroke, myocardial infarction, and death after carotid endarterectomy. Neurosurgery 2006;58:1066e73. discussion e 73. 5. Van den Berghe G, Wilmer A, Hermans G, Meersseman W, Wouters PJ, Milants I, et al. Intensive insulin therapy in the medical ICU. N Engl J Med 2006;354:449e61. 6. van den Berghe G, Wouters P, Weekers F, Verwaest C, Bruyninckx F, Schetz M, et al. Intensive insulin therapy in the critically ill patients. N Engl J Med 2001;345:1359e67. 7. Van den Berghe G, Wouters PJ, Bouillon R, Weekers F, Verwaest C, Schetz M, et al. Outcome benefit of intensive insulin therapy in the critically ill: insulin dose versus glycemic control. Crit Care Med 2003;31:359e66. 8. Gandhi GY, Nuttall GA, Abel MD, Mullany CJ, Schaff HV, O’Brien PC, et al. Intensive intraoperative insulin therapy versus conventional glucose management during cardiac surgery: a randomized trial. Ann Intern Med 2007;146:233e43. 9. Finfer S, Chittock DR, Su SY, Blair D, Foster D, Dhingra V, et al. Intensive versus conventional glucose control in critically ill patients. N Engl J Med 2009;360:1283e97. 10. Malmstedt J, Wahlberg E, Jorneskog G, Swedenborg J. Influence of perioperative blood glucose levels on outcome after infrainguinal bypass surgery in patients with diabetes. Br J Surg 2006;93:1360e7. 11. Bilotta F, Spinelli A, Giovannini F, Doronzio A, Delfini R, Rosa G. The effect of intensive insulin therapy on infection rate, vasospasm, neurologic outcome, and mortality in neurointensive care unit after intracranial aneurysm clipping in patients with acute subarachnoid hemorrhage: a randomized prospective pilot trial. J Neurosurg Anesthesiol 2007;19:156e60. 12. Furnary AP, Gao G, Grunkemeier GL, Wu Y, Zerr KJ, Bookin SO, et al. Continuous insulin infusion reduces mortality in patients with diabetes undergoing coronary artery bypass grafting. J Thorac Cardiovasc Surg 2003;125:1007e21. 13. van Wezel HB, Zuurbier CJ, de Jonge E, van Dam EW, van Dijk J, Endert E, et al. Differential effects of a perioperative hyperinsulinemic normoglycemic clamp on the neurohumoral stress response during coronary artery surgery. J Clin Endocrinol Metab 2006;91:4144e53. 14. Visser L, Zuurbier CJ, Hoek FJ, Opmeer BC, de Jonge E, de Mol BA, et al. Glucose, insulin and potassium applied as perioperative hyperinsulinaemic normoglycaemic clamp: effects on inflammatory response during coronary artery surgery. Br J Anaesth 2005;95:448e57. 15. Ouattara A, Lecomte P, Le Manach Y, Landi M, Jacqueminet S, Platonov I, et al. Poor intraoperative blood glucose control is associated with a worsened hospital outcome after cardiac surgery in diabetic patients. Anesthesiology 2005;103:687e94. 16. Gandhi GY, Nuttall GA, Abel MD, Mullany CJ, Schaff HV, Williams BA, et al. Intraoperative hyperglycemia and perioperative outcomes in cardiac surgery patients. Mayo Clin Proc 2005;80:862e6. 17. Bagry HS, Raghavendran S, Carli F. Metabolic syndrome and insulin resistance: perioperative considerations. Anesthesiology 2008;108:506e23. 18. Ramos M, Khalpey Z, Lipsitz S, Steinberg J, Panizales MT, Zinner M, et al. Relationship of perioperative hyperglycemia and postoperative infections in patients who undergo general and vascular surgery. Ann Surg 2008;248:585e91. 19. Krinsley J. Perioperative glucose control. Curr Opin Anaesthesiol 2006;19:111e6. 20. DeFronzo RA, Tobin JD, Andres R. Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol 1979;237:E214e23. 21. Thorell A, Alston-Smith J, Ljungqvist O. The effect of preoperative carbohydrate loading on hormonal changes, hepatic glycogen, and glucoregulatory enzymes during abdominal surgery. Nutrition 1996;12:690e5. 22. Thorell A, Efendic S, Gutniak M, Haggmark T, Ljungqvist O. Insulin resistance after abdominal surgery. Br J Surg 1994;81:59e63. 23. Thorell A, Loftenius A, Andersson B, Ljungqvist O. Postoperative insulin resistance and circulating concentrations of stress hormones and cytokines. Clin Nutr 1996;15:75e9. 24. Thorell A, Nygren J, Essen P, Gutniak M, Loftenius A, Andersson B, et al. The metabolic response to cholecystectomy: insulin resistance after open compared with laparoscopic operation. Eur J Surg 1996;162:187e91.
C. Blixt et al. / Clinical Nutrition 31 (2012) 676e681 25. Ljungqvist O, Thorell A, Gutniak M, Haggmark T, Efendic S. Glucose infusion instead of preoperative fasting reduces postoperative insulin resistance. J Am Coll Surg 1994;178:329e36. 26. Sato H, Carvalho G, Sato T, Lattermann R, Matsukawa T, Schricker T. The association of preoperative glycemic control, intraoperative insulin sensitivity, and outcomes after cardiac surgery. J Clin Endocrinol Metab 2010;95:4338e44. 27. Carli F, Galeone M, Gzodzic B, Hong X, Fried GM, Wykes L, et al. Effect of laparoscopic colon resection on postoperative glucose utilization and protein sparing: an integrated analysis of glucose and protein metabolism during the fasted and fed states using stable isotopes. Arch Surg 2005;140:593e7. 28. Kanno H, Kiyama T, Fujita I, Tani A, Kato S, Tajiri T, et al. Laparoscopic surgery improves blood glucose homeostasis and insulin resistance following distal gastrectomy for cancer. J Parenter Enteral Nutr 2009;33:686e90. 29. Uchida I, Asoh T, Shirasaka C, Tsuji H. Effect of epidural analgesia on postoperative insulin resistance as evaluated by insulin clamp technique. Br J Surg 1988;75:557e62. 30. Lattermann R, Carli F, Wykes L, Schricker T. Perioperative glucose infusion and the catabolic response to surgery: the effect of epidural block. Anesth Analg 2003;96:555e62 [table of contents]. 31. Thorell A, Rooyackers O, Myrenfors P, Soop M, Nygren J, Ljungqvist OH. Intensive insulin treatment in critically ill trauma patients normalizes glucose by reducing endogenous glucose production. J Clin Endocrinol Metab 2004;89:5382e6. 32. Nygren JO, Thorell A, Soop M, Efendic S, Brismar K, Karpe F, et al. Perioperative insulin and glucose infusion maintains normal insulin sensitivity after surgery. Am J Physiol 1998;275:E140e8. 33. Svanfeldt M, Thorell A, Hausel J, Soop M, Rooyackers O, Nygren J, et al. Randomized clinical trial of the effect of preoperative oral carbohydrate treatment on postoperative whole-body protein and glucose kinetics. Br J Surg 2007;94:1342e50. 34. Nygren J, Soop M, Thorell A, Efendic S, Nair KS, Ljungqvist O. Preoperative oral carbohydrate administration reduces postoperative insulin resistance. Clin Nutr 1998;17:65e71.
681
35. Umpierrez GE, Isaacs SD, Bazargan N, You X, Thaler LM, Kitabchi AE. Hyperglycemia: an independent marker of in-hospital mortality in patients with undiagnosed diabetes. J Clin Endocrinol Metab 2002;87:978e82. 36. Egi M, Bellomo R, Stachowski E, French CJ, Hart GK, Hegarty C, et al. Blood glucose concentration and outcome of critical illness: the impact of diabetes. Crit Care Med 2008;36:2249e55. 37. Badjatia N, Topcuoglu MA, Buonanno FS, Smith EE, Nogueira RG, Rordorf GA, et al. Relationship between hyperglycemia and symptomatic vasospasm after subarachnoid hemorrhage. Crit Care Med 2005;33:1603e9. quiz 23. 38. Subramaniam B, Panzica PJ, Novack V, Mahmood F, Matyal R, Mitchell JD, et al. Continuous perioperative insulin infusion decreases major cardiovascular events in patients undergoing vascular surgery: a prospective, randomized trial. Anesthesiology 2009;110:970e7. 39. Vriesendorp TM, DeVries JH, Hoekstra JB. Hypoglycemia and strict glycemic control in critically ill patients. Curr Opin Crit Care 2008;14:397e402. 40. Cammu G, Vermeiren K, Lecomte P, De Gendt S, Deloof T, Foubert L. Perioperative blood glucose management in patients undergoing tumor hepatectomy. J Clin Anesth 2009;21:329e35. 41. Sato H, Lattermann R, Carvalho G, Sato T, Metrakos P, Hassanain M, et al. Perioperative glucose and insulin administration while maintaining normoglycemia (GIN therapy) in patients undergoing major liver resection. Anesth Analg 2010;110:1711e8. 42. Ljungqvist O, Soop M, Hedstrom M. Why metabolism matters in elective orthopedic surgery: a review. Acta Orthop 2007;78:610e5. 43. Lehrke M, Broedl UC, Biller-Friedmann IM, Vogeser M, Henschel V, Nassau K, et al. Serum concentrations of cortisol, interleukin 6, leptin and adiponectin predict stress induced insulin resistance in acute inflammatory reactions. Crit Care 2008;12:R157. 44. Vanhorebeek I, Peeters RP, Vander Perre S, Jans I, Wouters PJ, Skogstrand K, et al. Cortisol response to critical illness: effect of intensive insulin therapy. J Clin Endocrinol Metab 2006;91:3803e13.
Contents lists available at SciVerse ScienceDirect
Clinical Nutrition journal homepage: http://www.elsevier.com/locate/clnu
Original article
The effect of perioperative glucose control on postoperative insulin resistanceq Christina Blixt a, b, *, Christian Ahlstedt a, b, Olle Ljungqvist c, Bengt Isaksson b, d, Sigridur Kalman a, b, Olav Rooyackers a, b a
Dept of Anaesthesia and Intensive Care, Karolinska University Hospital, Huddinge, Sweden Dept of Clinical Science, Intervention and Technology (CLINTEC), Karolinska Institutet, Huddinge, Sweden c Dept of Surgery, Örebro University Hospital, Örebro, Sweden d Division of Surgery, Karolinska University Hospital, Huddinge, Sweden b
a r t i c l e i n f o
s u m m a r y
Article history: Received 22 November 2011 Accepted 22 February 2012
Background & aims: Postoperative insulin resistance and the consequent hyperglycemia affects clinical outcome. Insulin sensitivity may be modulated by preoperative nutrition, adequate pain management and minimal invasive surgery. This study aims to disclose the impact of perioperative glucose control on postoperative insulin resistance. Methods: Twenty patients scheduled for elective open hepatectomy were enrolled in this prospective, randomized study. In the treatment group (n ¼ 9) insulin was administered intravenously to keep blood glucose between 6 and 8 mmol/l during surgery. The control group (n ¼ 8) received insulin if blood glucose >14 mmol/l. Insulin sensitivity was measured by a hyperinsulinemic normoglycemic clamp (0.8 mU/kg/min), performed on all patients both on the day before surgery and immediately postoperatively. Plasma cortisol, insulin and C-peptide were measured. Results: There was a significant difference in mean glucose value during surgery. In the control group 8.8 mmol/l (SD 1.5) vs. 6.9 mmol/l (SD 0.4) in the treated group, p ¼ 0.003. In the control group insulin sensitivity decreased to 21.9% 16.2% of the preoperative value and in the insulin treated group to 46.8 15.5%, p < 0.005. Insulin levels were significantly higher in the treatment group as well as consequently lower C-peptide levels. Conclusions: This trial revealed a significant difference in postoperative insulin resistance in the group treated with insulin during surgery. Ó 2012 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
Keywords: Liver resection Hyperinsulinemic normoglycemic clamp Abdominal surgery Cortisol Stress hormones
1. Introduction Hyperglycemia is often observed in surgical and critically ill patients. The changes that are seen are similar to that of a patient with type 2 diabetes and are the result of a transient insulin resistance, characterized by enhanced hepatic gluconeogenesis and glycogenolysis and a compromised peripheral insulin-dependent glucose uptake.1 Hyperglycemia is associated with an increased morbidity and mortality both in surgical and critically ill patients. Perioperative hyperglycemia has been associated with a higher risk for morbidity such as infections, myocardial infarction, acute
Abbreviations: M, total glucose disposal, mg/kg/min; M%, difference between pre-and postoperative M-value expressed as percentage. q Poster presentation, ESPEN conference, Nice, 2010. * Corresponding author. Dept of Anaesthesia and Intensive Care, Karolinska University Hospital, Sweden. Tel.: þ46 8 58580000; fax: þ46 8 7795424. E-mail addresses: [email protected], [email protected] (C. Blixt).
renal failure, and neurological damage.2e4 The significance of hyperglycemia during critical illness and especially the relevance of normalizing glucose levels has been a hot topic since the studies from van den Berghe and colleagues,5,6 where a significantly reduced risks of both morbidity and mortality in ICU-patients was shown when glucose was normalized with intensive insulin treatment. Post-hoc analysis of this latter study also showed that normoglycemia rather than the insulin dose given was responsible for these positive effects.7 However, several later studies have not been able to reproduce these results3,8,9 and the issue is still heavily debated. Studies in the perioperative period are less abundant.10e12 Whether controlling perioperative glucose levels have an effect on postoperative outcome is not entirely clear. Studies investigating this have mostly been focused on cardiac surgery with diabetic subjects.8,13e16 From these studies it was suggested that poor perioperative metabolic control is associated with increased risk of postoperative complication such as wound infection and cardiac events. Even effects on mortality have been reported.17
0261-5614/$ e see front matter Ó 2012 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved. doi:10.1016/j.clnu.2012.02.011
C. Blixt et al. / Clinical Nutrition 31 (2012) 676e681
Our thesis in this prospective study was that perioperative metabolic management in the form of normalizing glucose levels can be advantageous also in abdominal surgery. Postoperative insulin resistance was used as the outcome measure since we know that it is related to clinical outcome from other studies.18,19 We hypothesized that controlling glucose level between 6 and 8 mmol/ l during surgery may lower the degree of postoperative insulin resistance in comparison to a level >10 mmol/l. We investigated the effect in major open abdominal surgery. 2. Methods 2.1. Patients Patients scheduled for elective partial hepatectomy due to neoplastic condition were enrolled in this randomized prospective study. Twenty patients studied per protocol were planned. Inclusion criteria were patients scheduled for open laparotomy, over 18 years of age and with no contraindications for epidural anaesthesia. Patients with known diabetes mellitus or medication with corticosteroids were excluded. The study design was reviewed and approved by the regional ethics committee in Stockholm. The subjects were informed about the purpose and the nature of the study and written informed consent was given before inclusion. 2.2. Study design Insulin sensitivity was measured by a normoglycemic hyperinsulinemic clamp, as described below, the day before surgery and immediately after surgery. Patients were randomized to perioperative insulin treatment (with a glucose target of 6e8 mmol/l) or to a control group. The control group was treated accordingly to current clinical practice. The day before scheduled operation, the patient arrived to the hospital at 1 pm, fasted for a minimum of 4 h. A normoglycemic hyperinsulinemic clamp was performed as described in detail below. At the day of surgery the patient arrived to the operating theatre at 8 am after fasting since midnight. The patients received oxycodone (OxycontinÒ) as premedication, intra- and postoperative analgesia was managed with a thoracic epidural catheter that was inserted at Th 6e9. After a test dose with Bupivacain epinephrine 5 mg/ml (3e5 ml) (Marcain adrenalinÒ, Astra Zeneca) and an epidural bolus dose of fentanyl (50 mg) (Fentanyl, BraunÒ), an epidural infusion containing Bupivacain 1 mg/ml, epinephrine 2 mg/ml and fentanyl 2 mg/ml (12e15 ml/h) was started. General anaesthesia was induced with Propofol-Lipuro (BraunÒ) and fentanyl and maintained with Sevoflurane (Sevoran, AbbottÒ). Atracurium (HamelnÒ) was used for muscle relaxation. All antibiotics given were diluted in sodium chloride or sterile water. All patients received a central venous catheter and an arterial catheter which were inserted after induction of anaesthesia. The blood samples for the study were taken from the arterial line. Continuous infusion of glucose 25 mg/ml, 1 ml/kg/h, started as soon as the central venous line was inserted, following the clinical routine at the hospital. Glucose was measured every 10 min using a bedside glucose monitor (Hemocue Glucose 201þÒ, HemocueAB, Ängelholm, Sweden). In addition, hourly plasma samples were obtained and stored at 80 C for later plasma glucose analyses in the laboratory. Before induction the patient was randomized, with a sealed, opaque envelope, either to the insulin treatment group or to the control group. When insulin treatment was required, the infusion (50 IU ActrapidÒ in 50 ml saline ¼ 1 IU/ml) was started. Insulin was given either as a peripheral infusion or a central infusion as soon as the central catheter was inserted. An initial glucose level higher
677
than 8 mmol/ml, or gradually increasing values higher than 7 mmol/l in two following samples, was considered an indication for insulin treatment. The infusion was started at 2 IU/h and then adjusted according to the current glucose level. In the control group glucose was targeted at 10e14 mmol/l. Insulin was only administered (by infusion or by intermittent boluses) when glucose levels exceeded 14 mmol/l. After surgery the patient was transferred to the postoperative ward for observation over the next 12e24 h. The infusions stayed unchanged until the postoperative hyperinsulinemic normoglycemic clamp was performed. The postoperative clamp was started as soon as the epidural anaesthetic effect could be assessed and the patient appeared circulatory stable, i.e. 60e120 min after extubation. Samples for plasma cortisol, insulin and C-peptide analyses were obtained at the start and the end of the clamping procedures and at the start of anaesthesia, at the start and at the end of the liver resection phase. These samples taken in pre-chilled EDTA tubes, centrifuged within 30 min (2700 rpm/min, 10 min in 4 C, Universal 32 R Hettich ZentrifugenÒ, Andreas Hettich GmbH & Co. KG, Tuttlingen, Germany) and stored at 80 C until analyses. Data for circulation, saturation, blood losses, blood transfusion, extension and time of surgery were collected. Also the rate of glucose-, norepinephrine- and insulin infusions was recorded. Crystalloids (Ringer-AcetateÒ 2 ml/kg) and colloids (Hydroxyethyl-starch, VoluvenÒ, 2 ml/kg) were given as intraoperative fluid replacement. Blood loss during surgery was replaced with crystalloids, colloids and/or blood products. 2.3. Hyperinsulinemic normoglycemic clamp Two peripheral vein catheters were inserted for infusions and blood sampling. Two plasma glucose values 10 and 5 min before the start of the clamp were used as a baseline mean value. During the clamp, plasma glucose was measured every 5 min. The clamp was initiated by a bolus of insulin (1.1 U/m2) and then maintained by a constant insulin infusion (0.8 mU/kg/min) and a simultaneous variable infusion of glucose (200 mg/ml) to maintain normoglycemia. Albumin (400 mg) and potassium (32 mmol) were added to the insulin infusion. The goal level for steady state was set to 0.5 mmol/ml of the baseline mean value. Total glucose disposal during hyperinsulinemic clamp is expressed as M (mg/ kg/min). A high M-value represents a high insulin sensitivity. For safety reason, potassium was measured before and after the clamp using a venous blood sample analyzed on a blood-gas analyzer (ABL 800 Flex, Radiometer, Denmark). The clamp was repeated at the postoperative ward approximately 1 h after arrival. The preoperative glucose level was used as a target. For practical reasons arterial samples were drawn in the postoperative clamp. At the end of the 2 h clamp the insulin infusion was interrupted and thereafter the glucose infusion continued. After 30 min the infusion of glucose was stopped when a control sample showed normoglycemia. The clamp was continued for 2 h. Steady state condition was assumed to be obtained after 60 min. Steady states for both the glucose infusion and glucose levels during the second hour were evaluated blinded by a person not involved in performing the clamps and unaware of the randomization. A period of at least 30 min with steady state for both glucose infusion and glucose levels was identified by this person and used for calculations. Clamps with no steady state during at least 30 min were excluded. This validation was done before any group calculation was performed. The amount of glucose given during the steady state period was used to calculate a mean M-value for whole body sensitivity to insulin (mg glucose/kg/min).
678
C. Blixt et al. / Clinical Nutrition 31 (2012) 676e681
2.4. Analysis Plasma glucose was measured with a bedside glucose analyzer (Hemocue Glucose 201þÒ, HemocueAB, Ängelholm, Sweden) both during surgery and clamping. Plasma glucose was analyzed on an automatic analyzer (Konelab 20, Thermo Scientific, Jönköping, Sweden) using a GOD-POD analysis (Thermo Fisher Scientific, Vantaa, Finland). Plasma cortisol, insulin and C-peptide were analyzed using an ELISA-based standard analyzing kit (IMMULITEÒ 1000 Immunoassay System, SiemensÒ, Illinois, USA). 2.5. Statistics Ten patients in each group were analyzed per protocol. Two patients in the treatment group and one patient in the control group were excluded due to suboptimal insulin clamping, assessed blinded. Both Student’s T-test and ANOVA were used to analyze the glucose and insulin sensitivity data. For the hormone analyses several values were out of range and therefore non-parametric KruskaleWallis ANOVA was used for comparison between the two groups.
4. Discussion
Between May 2008 and January 2010, 22 patients scheduled for elective open hepatectomy were included and 10 þ 10 patients were studied per protocol. Two patients were excluded in the intraoperative period due to unexpected corticosteroid treatment. Ten patients were planned in each group. Due to poor quality of the hyperinsulinemic normoglycemic clamping (i.e. unable to reach steady state) another three patients were excluded (assess blinded), one patient in the treatment group and two patients in the control group. The two cohorts were statistically similar, there were no difference in BMI, age, operation time, blood loss or blood products given (see Table 1). The indication for surgery was metastasized liver cancer for all patients in the treatment group (9/9). In the control group 6/8 had metastasized liver cancer, one patient was proven not to have cancer after the microscopic evaluation of the biopsy and one patient had primary liver cancer. 4 out of 9 patients in the treatment group had insulin infusion started after induction of surgery. No patient received insulin treatment prior to induction of anaesthesia. During surgery 9 U of insulin was given (mean value, range 4e25 U). Table 1 Distribution of baseline characteristics. Treated (n ¼ 9)
Control (n ¼ 8)
6/3
8/0
Mean SD
Mean SD
p-value
Age (yrs) BMI (kg/m2) Comorbidity IHD Hypertonia Lung Kidney Blood loss (ml) Blood products RPB (units) Op time (min) Hepatectomy > 2 segments Hepatectomy < 2 segments
64 4.3 25.7 4.8
68 9.9 27.5 2.8
0.35 0.37
1 1 0 0 1219 1103 1.9 214 67 5 4
0 4 0 0 1303 961 2.1 242 70 4 4
0.87
M preop (mg/kg/min) M postop (mg/kg/min) M% difference pre-post
6.94 2.9 3.32 1.7 46.8 15.5
7.01 2.5 1.63 1.7 21.9 16.2
0.94 0.06 <0.005
0.4
This randomized trial showed that continuous glucose control during open liver resection maintains higher insulin sensitivity immediately after surgery. The majority of studies on perioperative glucose control have been focused on cardiovascular surgery and predominantly diabetic subjects. The research has shown a positive effect on the risk of developing myocardial infarction in the postoperative period. The present study, on major upper abdominal surgery, was not powered to show any impact on postoperative morbidity or mortality. Nevertheless, it revealed an effect of glucose control on postoperative insulin resistance. We used a hyperinsulinemic normoglycemic clamp to evaluate insulin sensitivity, which is regarded as the golden standard.20 In the group where glucose levels were kept at a level between 6 and 8 mmol/l the postoperative insulin resistance was significantly less. Small but significant differences in plasma levels of insulin, but primarily C-peptide were seen. This latter finding is expected, since the endogenous insulin production is affected during exogenous insulin infusion. Perioperative glucose
20
Blood glucose (mmol/l)
3. Results
Gender (m/f)
All patients received norepinephrine during surgery at different rates (approximately 0.02e0.1 mg/kg/min). After surgery 5/8 in the control group received norepinephrine (0.02e0.1 mg/kg/min) and in the treatment group 5/9 received norepinephrine (0.02e0.05 mg/ kg/min). Glucose levels during surgery were significantly different between the two groups (see Fig. 1), with mean glucose values in the control group of 8.8 mmol/l 1.5 (range 6.4e11.6) and in the treatment group of 6.9 0.4 (range 6.3e7.5) (p ¼ 0.003; t-test). In the control group the preoperative M decreased from 7.0 2.3 to 1.7 1.7 whereas in the treated group the preoperative M decreased from 7.2 2.0 to 3.4 1.6 (p ¼ 0.056; ANOVA). When the postoperative insulin sensitivity was expressed as percentage of preoperative value, the control group retained 21.9 16.2% and the treated group 46.8 15.5% (p < 0.005; t-test) of the preoperative value (Fig. 2). The circulating cortisol levels increased during surgery, but this increase seemed to be less severe in the treated group (p ¼ 0.06; KruskaleWallis, Table 2). As expected, the treatment group had a higher mean insulin level during surgery as well as postoperatively while the mean C-peptide level diminished under the same period (p ¼ 0.02; KruskaleWallis, Table 2).
control treatment
15
10
5
0
Operating time (min) Fig. 1. Perioperative glucose measurement with individual values presented. The control group had significantly higher glucose levels during surgery. The control group had a mean glucose value of 8.8 mmol/l 1.5 (range 6.4e11.6) compared to the treatment group which had a mean glucose value of 6.9 0.4 (range 6.3e7.5) (p ¼ 0.003; t-test). (D ¼ control group, - ¼ treatment group).
C. Blixt et al. / Clinical Nutrition 31 (2012) 676e681
100 80
M%
60
40 20
0
treatment group
control group
Fig. 2. Percentage of retained insulin sensitivity post-surgery (M%), individual and mean values are presented. The treatment group kept 46.8 15.5% (SD) compared to the control group which retained 21.9 16.2% (SD) of the preoperative value, (p < 0.005; t-test).
control also affected stress hormones during surgery. Our findings did not show a statistical difference in cortisol levels in the treatment group, but merely a trend to a lower mean cortisol level during surgery was observed. This pilot study aimed to reveal any difference in insulin resistance due to glucose control during major upper abdominal surgery. The patients have been subjected to liver surgery and were non-diabetics. To our knowledge, there are a few studies on major upper abdominal surgery with the focus on glucose control and insulin resistance, but they are mostly from open cholecystectomies in the 1990’s.21e25 Insulin resistance has been shown to have a negative effect on patients’ outcome. A large descriptive study on the impact of insulin resistance showed that postoperative complications were reduced for every 1 mg/kg/min in decrease in insulin sensitivity,26 making the small difference observed in our study clinically relevant. Reduced insulin resistance can be achieved by minimizing the surgical trauma, giving adequate pain relief and by shortening the fasting time. All these limit the sympathetic stress response. Laparoscopic colonic surgery has been shown to have a positive
Table 2 Stress hormone levels in the two groups. Treated Mean SD Cortisol (mg/Dl) Preoperative clamp-before Preoperative clamp-after Start operation Pre resection Post resection Postoperative clamp-before Postoperative clamp-after Insulin (mIU/mL) Preoperative clamp-before Preoperative clamp-after Start operation Pre resection Post resection Postoperative clamp-before Postoperative clamp-after C-peptide (ng/mL) Preoperative clamp-before Preoperative clamp-after Start operation Pre resection Post resection Postoperative clamp-before Postoperative clamp-after
Control Mean SD
12.8 8.59 11.42 24.29 24.15 27.15 20.79
6.64 1.9 2.95 12.36 13.76 13.94 8.23
11.9 9.16 14.67 28.85 32.88 36.01 33.19
4.95 4.59 10.99 13.15 11.2 11.68 14.83
10.65 106.05 9.98 50.44 47.06 115.69 166.31
11.62 33.67 6.02 35.59 52.47 153.46 56.23
14.16 137.86 13.94 23.58 32 60.01 162.21
14.3 47.64 9.22 24.82 33.46 73.42 80.74
2.93 2.35 2.34 1.27 1.09 0.96 0.46
1.88 1.69 1.2 0.93 1.05 0.97 0.57
3.04 2.81 3.31 4.24 4.84 3.84 1.54
1.66 1.56 1.21 2.16 3.59 2.28 1.53
679
effect on insulin sensitivity.27,28 In our case we chose open laparotomy, since it is the most common approach in our hospital and allows including patients with larger hepatic resection and more complications that could be related to insulin resistance. Insulin resistance can also be reduced by adequate pain regime perioperatively,29 in our case with epidural anaesthesia. Lattermann and Thorell demonstrated that a thoracic epidural lowered insulin resistance by reducing levels of glucagon and cortisol and thereby lower the endogenous glucose production.30,31 All patients in our study received a thoracic epidural. It was activated during operation and assessed postoperatively. Also, insulin and glucose infusion given prior and during surgery may have an effect on insulin resistance by inducing an enhanced insulin action, maintained during surgery.25,32 These findings have been reproduced by giving a preoperative oral carbohydrate loading 2e3 h before surgery.33,34 Patients included in the present study were all fasted over night, which is the standard routine in the hospital. We chose to keep that routine even if longer fasting time is known to alter glucose metabolism. Perioperative glucose infusions are routine in our hospital, this may have contributed to the perioperative hyperglycemia reported in this study and may have aggravated the postoperative insulin resistance. The concept of tight glucose control has been widely adapted in the ICU setting since the Leuven studies,6 where a remarkable impact on both mortality and morbidity was demonstrated. In the following study in 2006, the effect on mortality was not so clear and a slightly higher risk for hypoglycemia was seen.5 That latter study was conducted in a medical ICU, while the first one was set in a surgical ICU with primarily cardiothoracic patients. The limitation of intensive insulin treatment has been heavily debated. Especially the risk of hypoglycemia has been in focus since the NICE-SUGAR study was presented.9 Hyperglycemia induces a range of metabolic changes in several defense system, for example a decrease in immunological function, enhanced immune factors, endothelial dysfunction, and disturbances in the coagulation system.17 The result is an increased risk for infections, cardiac, neurological and renal damage. In a retrospective study on patients undergoing carotid endarterectomy by McGirt, all glucose measurements over a ten-year period were reviewed. Hyperglycaemic patients, with a preoperative glucose value of >11 mmol/l, had a 4.3 and 2.8 times higher risk of having a myocardial infarction or stroke event respectively.4 Diabetics are generally more prone to postoperative complications. However, studies have demonstrated a higher mortality and morbidity among non-diabetic patients exposed to hyperglycemia.35,36 It seems that acute hyperglycemia is one factor in the development of postoperative complications. Hyperglycemia in neurosurgical patients is known to cause increased infarct size, vasospasm and may compromise recovery.11,37 Ouattara reported an association to malignant arrhythmia, need for longer vasopressor support and ventilation as well as neurological and renal impairment in patients with poor glycaemic control subjected to cardiovascular surgery.15 These findings have been reproduced in a prospective study by Subramaniam in 2009 where patients reaching a glucose target of 5e8 mmol/l had less risk of postoperative myocardial infarction.38 The liver is closely involved in glucose metabolism. In postoperative insulin resistance there is a change in both peripheral glucose uptake and hepatic glycogenolysis and gluconeogenesis. Theoretically there is a risk for hypoglycemia if a larger part of the liver is removed during surgery. Liver failure has been demonstrated to be a factor in developing hypoglycemia.39 However, recent studies performed on liver surgery, where insulin treatment protocols have been used, have had few recordings of severe hypoglycemia.40
680
C. Blixt et al. / Clinical Nutrition 31 (2012) 676e681
Where severe hypoglycemia has been reported a lower glucose target has been evaluated, in one study 3.6e6.1 mmol/l.41 In our study all patients had rising glucose values during surgery. In our study we chose a glucose target of 6e8 mmol/l, a more liberal target range than discussed in several studies. During surgery the effect of insulin might be impaired due a neuroendocrine stress response. The counter-acting hormones such as glucagon, cortisol and catecholamines induce changes in the lipid-, protein and glucose metabolism as well as the immune system.17,42 The hormones are secreted to supply substrates to the organism. The liver is the primary target, resulting in increased gluconeogenesis and glycogenolysis.1 But also peripheral tissues are affected, where the stress hormones inhibit glucose uptake. Due to the hepatic insulin resistance during stress, circulating cortisol aggravates hyperglycemia despite hyperinsulinemia.1 Cortisol has been demonstrated to be a stronger predictor of stress induced insulin resistance compared to some cytokines.43 But also a direct effect of insulin on cortisol has been demonstrated during surgery and critical illness.32,44 Our results do not explain the underlying mechanisms of insulin sensitivity. The trial was designed as a study to elucidate the effect of perioperative glucose control. Obviously a more thorough protocol, using labelled glucose to measure glucose turnover rates, is required to further elucidate the metabolic mechanisms. In our study we see a trend of rising cortisol levels during and after surgery in both groups. However, the mean cortisol level in the treatment group is somewhat lower, though this difference is not statistical significant. These findings are similar to those of van Wesel,13 where cortisol levels were clearly affected during and after clamping in CABG-patients. One reason for not finding a statistical difference in our study is probably due to the small groups. 5. Conclusions We showed an impact of intraoperative glucose control on postoperative insulin resistance. The treatment group retained 46.8 15.5% of their preoperative insulin sensitivity, and the control group 21.9% 16.2%, (p < 0.005). We found a trend to a mean lower cortisol level in the treatment group, though not statistical significant. The treatment group was kept between 6 and 8 mmol/l during surgery using insulin infusion. Statement of authorship CB, CA carried out the design of study, collected and analyzed data and drafted the manuscript. OL contributed by interpretation of data, and drafting the manuscript. BI and SK participated in the design of the study and coordination and helped to draft the manuscript. OR contributed by the design of the study, data interpretation and analyzing, statistical analyses and drafting the manuscript. All authors have read and approved the final manuscript. Conflict of interest OL owns a patent and receives royalties for a carbohydrate rich drink used in studies referred to in the paper. The patent is licensed to Nutricia who makes and sells a product based on it. None of the other authors have any conflict of interest to declare. Acknowledgements Grants/funding: The study was sponsored by the Swedish Medical Research Council (no: 14244), Karolinska Institute Funds and the Stockholm County Council.
A special thanks to the skilled nursing assistance from Gunilla Herman, Viveka Gustafsson, Christina Hebert and Eva Skog Nejman, Dept of Anesthesia and Intensive Care, KarolinskaUniversity Hospital, Huddinge, Sweden. The staff at Endocrinology Laboratory, Dept of Medicine, KarolinskaUniversity Hospital, Huddinge, Sweden is gratefully acknowledged. References 1. Lipshutz AK, Gropper MA. Perioperative glycemic control: an evidence-based review. Anesthesiology 2009;110:408e21. 2. Finney SJ, Zekveld C, Elia A, Evans TW. Glucose control and mortality in critically ill patients. JAMA 2003;290:2041e7. 3. Krinsley JS. Effect of an intensive glucose management protocol on the mortality of critically ill adult patients. Mayo Clin Proc 2004;79:992e1000. 4. McGirt MJ, Woodworth GF, Brooke BS, Coon AL, Jain S, Buck D, et al. Hyperglycemia independently increases the risk of perioperative stroke, myocardial infarction, and death after carotid endarterectomy. Neurosurgery 2006;58:1066e73. discussion e 73. 5. Van den Berghe G, Wilmer A, Hermans G, Meersseman W, Wouters PJ, Milants I, et al. Intensive insulin therapy in the medical ICU. N Engl J Med 2006;354:449e61. 6. van den Berghe G, Wouters P, Weekers F, Verwaest C, Bruyninckx F, Schetz M, et al. Intensive insulin therapy in the critically ill patients. N Engl J Med 2001;345:1359e67. 7. Van den Berghe G, Wouters PJ, Bouillon R, Weekers F, Verwaest C, Schetz M, et al. Outcome benefit of intensive insulin therapy in the critically ill: insulin dose versus glycemic control. Crit Care Med 2003;31:359e66. 8. Gandhi GY, Nuttall GA, Abel MD, Mullany CJ, Schaff HV, O’Brien PC, et al. Intensive intraoperative insulin therapy versus conventional glucose management during cardiac surgery: a randomized trial. Ann Intern Med 2007;146:233e43. 9. Finfer S, Chittock DR, Su SY, Blair D, Foster D, Dhingra V, et al. Intensive versus conventional glucose control in critically ill patients. N Engl J Med 2009;360:1283e97. 10. Malmstedt J, Wahlberg E, Jorneskog G, Swedenborg J. Influence of perioperative blood glucose levels on outcome after infrainguinal bypass surgery in patients with diabetes. Br J Surg 2006;93:1360e7. 11. Bilotta F, Spinelli A, Giovannini F, Doronzio A, Delfini R, Rosa G. The effect of intensive insulin therapy on infection rate, vasospasm, neurologic outcome, and mortality in neurointensive care unit after intracranial aneurysm clipping in patients with acute subarachnoid hemorrhage: a randomized prospective pilot trial. J Neurosurg Anesthesiol 2007;19:156e60. 12. Furnary AP, Gao G, Grunkemeier GL, Wu Y, Zerr KJ, Bookin SO, et al. Continuous insulin infusion reduces mortality in patients with diabetes undergoing coronary artery bypass grafting. J Thorac Cardiovasc Surg 2003;125:1007e21. 13. van Wezel HB, Zuurbier CJ, de Jonge E, van Dam EW, van Dijk J, Endert E, et al. Differential effects of a perioperative hyperinsulinemic normoglycemic clamp on the neurohumoral stress response during coronary artery surgery. J Clin Endocrinol Metab 2006;91:4144e53. 14. Visser L, Zuurbier CJ, Hoek FJ, Opmeer BC, de Jonge E, de Mol BA, et al. Glucose, insulin and potassium applied as perioperative hyperinsulinaemic normoglycaemic clamp: effects on inflammatory response during coronary artery surgery. Br J Anaesth 2005;95:448e57. 15. Ouattara A, Lecomte P, Le Manach Y, Landi M, Jacqueminet S, Platonov I, et al. Poor intraoperative blood glucose control is associated with a worsened hospital outcome after cardiac surgery in diabetic patients. Anesthesiology 2005;103:687e94. 16. Gandhi GY, Nuttall GA, Abel MD, Mullany CJ, Schaff HV, Williams BA, et al. Intraoperative hyperglycemia and perioperative outcomes in cardiac surgery patients. Mayo Clin Proc 2005;80:862e6. 17. Bagry HS, Raghavendran S, Carli F. Metabolic syndrome and insulin resistance: perioperative considerations. Anesthesiology 2008;108:506e23. 18. Ramos M, Khalpey Z, Lipsitz S, Steinberg J, Panizales MT, Zinner M, et al. Relationship of perioperative hyperglycemia and postoperative infections in patients who undergo general and vascular surgery. Ann Surg 2008;248:585e91. 19. Krinsley J. Perioperative glucose control. Curr Opin Anaesthesiol 2006;19:111e6. 20. DeFronzo RA, Tobin JD, Andres R. Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol 1979;237:E214e23. 21. Thorell A, Alston-Smith J, Ljungqvist O. The effect of preoperative carbohydrate loading on hormonal changes, hepatic glycogen, and glucoregulatory enzymes during abdominal surgery. Nutrition 1996;12:690e5. 22. Thorell A, Efendic S, Gutniak M, Haggmark T, Ljungqvist O. Insulin resistance after abdominal surgery. Br J Surg 1994;81:59e63. 23. Thorell A, Loftenius A, Andersson B, Ljungqvist O. Postoperative insulin resistance and circulating concentrations of stress hormones and cytokines. Clin Nutr 1996;15:75e9. 24. Thorell A, Nygren J, Essen P, Gutniak M, Loftenius A, Andersson B, et al. The metabolic response to cholecystectomy: insulin resistance after open compared with laparoscopic operation. Eur J Surg 1996;162:187e91.
C. Blixt et al. / Clinical Nutrition 31 (2012) 676e681 25. Ljungqvist O, Thorell A, Gutniak M, Haggmark T, Efendic S. Glucose infusion instead of preoperative fasting reduces postoperative insulin resistance. J Am Coll Surg 1994;178:329e36. 26. Sato H, Carvalho G, Sato T, Lattermann R, Matsukawa T, Schricker T. The association of preoperative glycemic control, intraoperative insulin sensitivity, and outcomes after cardiac surgery. J Clin Endocrinol Metab 2010;95:4338e44. 27. Carli F, Galeone M, Gzodzic B, Hong X, Fried GM, Wykes L, et al. Effect of laparoscopic colon resection on postoperative glucose utilization and protein sparing: an integrated analysis of glucose and protein metabolism during the fasted and fed states using stable isotopes. Arch Surg 2005;140:593e7. 28. Kanno H, Kiyama T, Fujita I, Tani A, Kato S, Tajiri T, et al. Laparoscopic surgery improves blood glucose homeostasis and insulin resistance following distal gastrectomy for cancer. J Parenter Enteral Nutr 2009;33:686e90. 29. Uchida I, Asoh T, Shirasaka C, Tsuji H. Effect of epidural analgesia on postoperative insulin resistance as evaluated by insulin clamp technique. Br J Surg 1988;75:557e62. 30. Lattermann R, Carli F, Wykes L, Schricker T. Perioperative glucose infusion and the catabolic response to surgery: the effect of epidural block. Anesth Analg 2003;96:555e62 [table of contents]. 31. Thorell A, Rooyackers O, Myrenfors P, Soop M, Nygren J, Ljungqvist OH. Intensive insulin treatment in critically ill trauma patients normalizes glucose by reducing endogenous glucose production. J Clin Endocrinol Metab 2004;89:5382e6. 32. Nygren JO, Thorell A, Soop M, Efendic S, Brismar K, Karpe F, et al. Perioperative insulin and glucose infusion maintains normal insulin sensitivity after surgery. Am J Physiol 1998;275:E140e8. 33. Svanfeldt M, Thorell A, Hausel J, Soop M, Rooyackers O, Nygren J, et al. Randomized clinical trial of the effect of preoperative oral carbohydrate treatment on postoperative whole-body protein and glucose kinetics. Br J Surg 2007;94:1342e50. 34. Nygren J, Soop M, Thorell A, Efendic S, Nair KS, Ljungqvist O. Preoperative oral carbohydrate administration reduces postoperative insulin resistance. Clin Nutr 1998;17:65e71.
681
35. Umpierrez GE, Isaacs SD, Bazargan N, You X, Thaler LM, Kitabchi AE. Hyperglycemia: an independent marker of in-hospital mortality in patients with undiagnosed diabetes. J Clin Endocrinol Metab 2002;87:978e82. 36. Egi M, Bellomo R, Stachowski E, French CJ, Hart GK, Hegarty C, et al. Blood glucose concentration and outcome of critical illness: the impact of diabetes. Crit Care Med 2008;36:2249e55. 37. Badjatia N, Topcuoglu MA, Buonanno FS, Smith EE, Nogueira RG, Rordorf GA, et al. Relationship between hyperglycemia and symptomatic vasospasm after subarachnoid hemorrhage. Crit Care Med 2005;33:1603e9. quiz 23. 38. Subramaniam B, Panzica PJ, Novack V, Mahmood F, Matyal R, Mitchell JD, et al. Continuous perioperative insulin infusion decreases major cardiovascular events in patients undergoing vascular surgery: a prospective, randomized trial. Anesthesiology 2009;110:970e7. 39. Vriesendorp TM, DeVries JH, Hoekstra JB. Hypoglycemia and strict glycemic control in critically ill patients. Curr Opin Crit Care 2008;14:397e402. 40. Cammu G, Vermeiren K, Lecomte P, De Gendt S, Deloof T, Foubert L. Perioperative blood glucose management in patients undergoing tumor hepatectomy. J Clin Anesth 2009;21:329e35. 41. Sato H, Lattermann R, Carvalho G, Sato T, Metrakos P, Hassanain M, et al. Perioperative glucose and insulin administration while maintaining normoglycemia (GIN therapy) in patients undergoing major liver resection. Anesth Analg 2010;110:1711e8. 42. Ljungqvist O, Soop M, Hedstrom M. Why metabolism matters in elective orthopedic surgery: a review. Acta Orthop 2007;78:610e5. 43. Lehrke M, Broedl UC, Biller-Friedmann IM, Vogeser M, Henschel V, Nassau K, et al. Serum concentrations of cortisol, interleukin 6, leptin and adiponectin predict stress induced insulin resistance in acute inflammatory reactions. Crit Care 2008;12:R157. 44. Vanhorebeek I, Peeters RP, Vander Perre S, Jans I, Wouters PJ, Skogstrand K, et al. Cortisol response to critical illness: effect of intensive insulin therapy. J Clin Endocrinol Metab 2006;91:3803e13.