Use of i.v. insulin in well-controlled non-insulin-dependent diabetics undergoing major surgery

Size: px

Start display at page:

Download "Use of i.v. insulin in well-controlled non-insulin-dependent diabetics undergoing major surgery"

Wilfrid Davis
6 years ago
Views:

1 British Journal of Anaesthesia 1996; 76: Use of i.v. insulin in well-controlled non-insulin-dependent diabetics undergoing major surgery M. RAUCOULES-AIMÉ, Y. LABIB, J. LEVRAUT, P. GASTAUD, C. DOLISI AND D. GRIMAUD Summary We conducted a randomized, prospective study to assess the effect of i.v. insulin on blood glucose control, development of ketone bodies and hormonal changes in 60 well-controlled, non-insulindependent diabetics (NIDDM) undergoing major surgery. In group A, patients were given only 0.9 % saline; in group B, patients were given insulin as a continuous i.v. infusion (1.25 u. h 1 ); in group C, patients were given insulin 10 u. i.v. boluses every 2 h. Patients in all three groups were given insulin 5 u. when their intraoperative blood glucose concentration increased to greater than 11.1 mmol litre 1. Blood glucose concentrations were measured every 15 min, from just before induction of anaesthesia to 2 h after surgery. Plasma lactate, pyruvate, ketone body, C-peptide and counter-regulatory hormone concentrations were also measured. Blood glucose concentrations in the three groups did not differ significantly. There was a mild-to-moderate increase in plasma ketone body concentrations in group A, but without any deleterious consequences. Plasma C-peptide concentrations decreased significantly in groups B and C, especially in patients given bolus injections of insulin. Plasma growth hormone concentrations also increased significantly in group B and C patients. This study indicated that the no insulin no glucose regimen was a simple, effective way to control blood glucose in well-controlled NIDDM patients, provided blood glucose was measured frequently and insulin used appropriately. (Br. J. Anaesth. 1996; 76: ) Key words Complications, diabetes. Metabolism, insulin. Metabolism, glucose. Diabetes. There is some controversy on the metabolic management of patients with non-insulin-dependent diabetes mellitus (NIDDM) undergoing surgery [1 4]. Clearly, poorly controlled NIDDM patients should be given insulin, while well-controlled ones undergoing minor surgery do not require any special treatment [1, 5, 6]. The case for the metabolic management of NIDDM patients undergoing major surgery with insulin is not simple [1, 4]. Insulin may induce hypoglycaemia and metabolic abnormalities [2], complicating the management of anaesthesia. Controlling blood glucose with insulin appears to provide little or no benefit over no-insulin treatment [7]. Some authors therefore propose giving insulin only when blood glucose concentrations exceed 11.1 mmol litre 1, assuming there is adequate preoperative control and perioperative blood glucose monitoring [8], while others routinely give NIDDM patients insulin [1]. They point out that, in spite of hyperinsulinaemia, frequently found in these patients, hyperglycaemia, osmotic diuresis, ketosis and acidaemia may occur under stressful conditions. We therefore conducted a randomized, prospective study to evaluate the effects of i.v. insulin on blood glucose control, ketone body development and hormonal changes in well-controlled NIDDM patients undergoing major surgery. Patients and methods We studied 60 NIDDM patients undergoing elective orthopaedic surgery (hip replacement, spinal surgery) or abdominal surgery (cholecystectomy, gastrectomy, bowel resection). The study was approved by the Ethics Committee of the University Hospital Centre and informed consent was obtained from all patients. Inclusion criteria were: fasting plasma blood glucose 7.9 mmol litre 1 and glycosylated haemoglobin (HbA 1C) less than 8 % [8, 9]. Patients with impaired renal function (creatinine 120 mol litre 1 ) or impaired hepatic function (based on routine liver function tests) were excluded from the study. None of the patients had been given insulin previously. Sulphonylurea hypoglycaemic drugs were discontinued the night before surgery, and biguanides 72 h before. All had fasted for h before surgery. Patients were allocated randomly to one of three groups. Group A patients were given only 0.9 % saline 125 ml h 1 ( no insulin no glucose ). Patients in groups B and C were given a continuous infusion of 5 % glucose 6.25 g h 1 with i.v. shortacting insulin (Actrapid HM, Novo, France), started just before induction of anaesthesia. In group B, patients were given insulin as a continuous infusion (1.25 u. h 1 ) by an electrically driven syringe (40 u./31 ml 0.9 % saline (1.25 u. 1 ml)) [10]; patients in group C were given insulin 10 u. by direct i.v. bolus injection every 2 h [11]. Patients in all three groups were anaesthetized only M. RAUCOULES-AIMÉ, MD, PHD, Y. LABIB, MD, J. LEVRAUT, MD, D. GRIMAUD, MD (Department of Anaesthesia); P. GASTAUD, MD (Department of Ophthalmology); C. DOLISI, MD (Department of Physiology); Hôpital Saint-Roch, 5 rue P. Dévoluy, Centre Hospitalier et Universitaire, Nice Cedex 1, France. Accepted for publication: September 29, Correspondence to M.R.-A.

2 Non-insulin-dependent diabetes and major surgery 199 when blood glucose concentration was mmol litre 1 ; if the concentration was 5.5 mmol litre 1, glucose was given; if it was 11.1 mmol litre 1, additional insulin (5 u.) was given as an i.v. bolus until blood glucose was 11.1 mmol litre 1. During surgery, all patients were given a bolus injection of insulin 5 u. whenever blood glucose exceeded 11.1 mmol litre 1, with at least 60 min between doses. A second venous cannula was used to infuse other medications or 0.9 % saline for volume expansion (fluids containing glucose and lactate were prohibited). All patients were premedicated with flunitrazepam 1 2 mg orally. All procedures were performed under general anaesthesia. Anaesthesia was induced with thiopentone 5 mg kg 1, and tracheal intubation was facilitated with vecuronium 0.1 mg kg 1. Fentanyl 2 g kg 1 was used at induction and during anaesthesia in response to hypertension, tachycardia, or both. Anaesthesia was maintained with % enflurane and 50 % nitrous oxide in oxygen. Neuromuscular block was maintained with vecuronium, which was not antagonized after surgery. Ventilation of the lungs was controlled to maintain end-tidal PCO 2 at kpa. The study began just before induction and ended 2 h after surgery. Capillary concentrations of glucose were measured every 15 min. Plasma ketone bodies (acetoacetate, -hydroxybutyrate), lactate, pyruvate, C-peptide and counter-regulatory hormones (catecholamines, cortisol, glucagon, growth hormone) were measured in venous samples (S) taken just before induction (S1), during surgery (S2 : 30 min after the beginning of cholecystectomy; 60 min for gastrectomy, bowel resection, hip replacement and spinal surgery), and immediately after completion of surgery and tracheal extubation (S3). Details of the assays have been published previously [10]. The following variables were recorded: capillary glucose concentrations; treatment failure (defined as blood glucose concentration 3.3 mmol litre 1 or 16.5 mmol litre 1 ); total amount of insulin and number of additional boluses given; plasma acetoacetate and -hydroxybutyrate : acetoacetate ratios at S1, S2, S3; plasma lactate, pyruvate concentrations and lactate : pyruvate ratios at S1, S2, S3; plasma C-peptide and counter-regulatory hormone concentrations at S1, S2, S3. Data were analysed by two-way analysis of variance (ANOVA) for repeated measurements. Student s unpaired two-tailed t test was used to compare groups. Comparison of two time points within a group were made using Student s two-tailed t test for paired data. The resulting P values were adjusted by Bonferroni s method. ANOVA was used to compare the amount of insulin given in both groups and the quantities of fentanyl. hydroxybutyrate : acetoacetate and lactate : pyruvate ratios were compared using non-parametric tests. Non-parametric tests were also used to compare plasma growth hormone concentrations. The chisquare test was used for non-continuous variables. Results are expressed as mean (SEM). P 0.05 was considered statistically significant. Statistical analyses were performed using a MacIntosh II computer with the Statview program. Results There were no significant differences between the groups in patient data, type or duration of surgery, duration of diabetes or preoperative control of glycaemia, determined by HbA 1c, fasting plasma glucose and capillary glucose concentrations measured just before induction of anaesthesia (table 1). The doses of fentanyl given to the three groups were similar (A: 665 (25) g; B: 602 (25) g; C: 650 (22) g). Seven patients had blood glucose concentrations controlled by diet alone (two in group A, three in group B, two in group C). Six patients in group A, six in group B and five in group C were given blood products during surgery (packed red blood cells or plasma). The distribution of all blood glucose concentrations (expressed as percentiles) measured every Figure 1 Distribution of blood glucose values measured every 15 min during the study, the displayed as percentiles (%), for groups A (no insulin no glucose), B (continuous infusion) and C (bolus). Table 1 Patient data, blood glucose concentrations (BGC) just before induction of anaesthesia, haemoglobin A1c (HbA1c) and duration of diabetes in group A (no insulin no glucose), group B (continuous infusion) and group C (bolus) (number or mean (range or SEM)). No significant differences between groups Group A Group B Group C Sex distribution (M/F) 12/8 13/7 13/7 Age (yr) 61.9 (45 81) 64.5 (48 80) 61.1 (41 77) Weight (kg) 70.2 (56 90) 69.1 (55 83) 76.1 (50 88) BGC (mmol litre 1 ) 6.7 ( ) 7.5 ( ) 7.1 ( ) HbA1c (%) 6.8 ( ) 7.0 ( ) 7.1 ( ) Duration of diabetes (yr) 7.0 (0.9) 8.1 (1.5) 8.0 (2.0) Duration of surgery (h) 3.5 (0.3) 2.6 (0.4) 3.0 (0.3)

3 200 British Journal of Anaesthesia Figure 2 Mean (SEM) blood glucose concentrations measured every 15 min in groups A (no insulin no glucose), B (continuous infusion) and C (bolus). No significant differences between groups. 15 min during the study in all groups is shown in figure 1. Blood glucose concentrations in the three groups were not significantly different (two-way repeated ANOVA) (fig. 2). Infusion of glucose was increased in three patients (one in group A, two in group B) just before induction of anaesthesia to offset low blood glucose concentrations ( 5.5 mmol litre 1 ). The number of additional insulin boluses given when intraoperative blood glucose concentrations exceeded 11.1 mmol litre 1 was similar in all three groups (table 2). No case of hypoglycaemia (blood glucose 3.3 mmol litre 1 ) or hyperglycaemia (blood glucose 16.5 mmol litre 1 ) was encountered. Plasma -hydroxybutyrate concentration increased just before anaesthesia (S1) and gradually decreased during surgery (P 0.05), except in group A, where concentrations remained increased (S2: 0.63 ( ) mmol litre 1 ; S3: 0.71 ( ) mmol litre 1 ) (table 3). Acetoacetate concentrations did not differ significantly (group A, S2: 0.39 ( ) mmol litre 1 ; S3: 0.45 ( ) mmol litre 1 ) and the -hydroxybutyrate: acetoacetate ratio did not vary significantly with time. It remained 1 in group A, but decreased in groups B and C (P 0.001, P 0.01). Plasma lactate and pyruvate concentrations increased significantly (P 0.05 and P 0.01) during surgery, with no differences between groups, although pyruvate concentration increased to a lesser extent in group A. The lactate : pyruvate ratio did not change throughout the study (table 3). There were no major metabolic problems during the postoperative period in any of the three groups. Plasma C-peptide concentrations differed significantly between the groups (P 0.01) (table 4). The time course of changes in the concentrations were also significantly different (P ). C-peptide concentrations decreased significantly during surgery in groups B and C (table 4), especially in patients given bolus injections of insulin (P 0.001), while plasma C-peptide concentrations remained stable in group A. Plasma cortisol and catecholamine concentrations increased similarly during surgery in the three groups (table 4). Plasma glucagon concentrations remained stable. Plasma growth hormone concentrations were significantly different between groups (P 0.01); they increased significantly in groups B and C (P 0.05, P 0.01) and remained constant in group A. Discussion The majority of diabetics undergoing surgery in Western countries are non-insulin-dependent. Nevertheless, there is no consensus on the best method of managing blood glucose concentrations of patients with well-controlled NIDDM undergoing major surgery, especially in the use of insulin [1, 5, 8]. The main metabolic complications in diabetics during surgery are hypoglycaemic reactions caused by exogenous insulin, hyperglycaemia, ketosis and acidaemia [8]. This study showed that the control of blood glucose was similar, whether or not patients Table 2 Insulin given (mean (SEM or range)). **P 0.01, group A vs group B; P 0.01, group A vs group C; P 0.01, group B vs group C Group A Group B Group C Total amount (u.) 3.4 (0 15) 11.5 (4 20)** 27.0 (20 37) Amount per hour (u. h 1 ) 0.6 (0 3.5) 2.4 ( )** 5.5 ( ) Additional boluses (n) 0.50 (0 3) 1.3 (0 3) 0.6 (0 3)

4 Non-insulin-dependent diabetes and major surgery 201 Table 3 Mean (range) lactate, pyruvate, ketone body concentrations and lactate: pyruvate, -hydroxybutyrate:acetoacetate concentration ratios before operation (S1), during surgery (S2) and after tracheal extubation (S3). Significant differences compared with preoperative values (S1): P 0.05; P 0.01; P 0.001; significant differences between groups: *P 0.05, **P 0.01, ***P S1 S2 S3 No insulin no glucose Lactate (mmol litre 1 ) 1.4 ( ) 1.9 ( ) 2.1 ( ) Pyruvate ( mol litre 1 ) 56 (36 81) 60 (33 107) 73 (40 123) Lactate : pyruvate ratio 25.2 ( ) 26.6 ( ) 26.2 ( ) -Hydroxybutyrate (mmol litre 1 ) 0.43 ( ) 0.63 ( ) 0.71 ( ) Acetoacetate (mmol litre 1 ) 0.37 ( ) 0.39 ( ) 0.45 ( ) -Hydroxybutyrate:acetoacetate 1.23 ( ) 1.60 ( ) 1.68 ( ) Continuous infusion Lactate (mmol litre 1 ) 1.5 ( ) 1.9 ( ) 2.1 ( ) Pyruvate ( mol litre 1 ) 55 (15 91) 65 (22 147) 79 (27 123) Lactate: pyruvate ratio 26.5 ( ) 24.3 ( ) 25.5 ( ) -Hydroxybutyrate (mmol litre 1 ) 0.70 ( ) 0.34 ( ) 0.19 ( ) *** Acetoacetate (mmol litre 1 ) 0.55 ( ) 0.45 ( ) 0.39 ( ) -Hydroxybutyrate : acetoacetate 1.31 ( ) 0.75 ( ) 0.50 ( ) *** Bolus Lactate (mmol litre 1 ) 1.6 ( ) 2.0 ( ) 2.3 ( ) Pyruvate ( mol litre 1 ) 60 (20 94) 90 (40 199)* 102 (42 130) ** Lactate : pyruvate ratio 25.5 ( ) 22.5 ( ) 23.1 ( ) -Hydroxybutyrate (mmol litre 1 ) 0.55 ( ) 0.15 ( ) * 0.14 ( ) *** Acetoacetate (mmol litre 1 ) 0.31 ( ) 0.27 ( ) 0.18 ( ) -Hydroxybutyrate : acetoacetate 1.56 ( ) 0.58 ( ) ** 0.79 ( )** Table 4 Mean (range) hormonal data before operation (S1), during surgery (S2) and after tracheal extubation (S3). Significant differences compared with preoperative values (S1): P 0.05, P 0.01, P 0.001; significant differences between groups: *P 0.05, **P 0.01, ***P S1 S2 S3 No insulin no glucose Nordrenaline (nmol litre 1 ) 2.17 ( ) 4.09 ( ) 6.41 ( ) Adrenaline (nmol litre 1 ) 0.47 ( ) 0.83 ( ) 1.58 ( ) Cortisol (nmol litre 1 ) 345 ( ) 468 ( ) 668 ( ) Growth hormone ( iu ml 1 ) 0.9 ( ) 2.6 ( ) 2.7 ( ) Glucagon (pmol litre 1 ) 134 (38 439) 126 (32 457) 150 (32 476) C-peptide (nmol litre 1 ) 2.65 ( ) 2.76 ( ) 2.27 ( ) Continuous infusion Noradrenaline (nmol litre 1 ) 2.16 ( ) 3.95 ( ) 4.74 ( ) Adrenaline (nmol litre 1 ) 0.95 ( ) 2.27 ( ) 3.63 ( ) Cortisol (nmol litre 1 ) 372 ( ) 450 ( ) 757 ( ) Growth hormone ( iu ml 1 ) 1.4 ( ) 6.6 ( ) * 5.7 ( ) Glucagon (pmol litre 1 ) 174 (68 555) 206 (52 518) 169 (55 523) C-peptide (nmol litre 1 ) 2.18 ( ) 1.82 ( )* 1.50 ( ) * Bolus Noradrenaline (nmol litre 1 ) 1.67 ( ) 2.76 ( ) 4.34 ( ) Adrenaline (nmol litre 1 ) 0.48 ( ) 1.33 ( ) 1.89 ( ) Cortisol (nmol litre 1 ) 442 ( ) 859 ( ) 1024 ( ) Growth hormone ( iu ml 1 ) 1.1 ( ) 6.5 ( ) ** 7.7 ( ) ** Glucagon (pmol litre 1 ) 104 (45 247) 96 (33 264) 106 (42 259) C-peptide (nmol litre 1 ) 2.39 ( ) 1.29 ( ) *** 1.02 ( ) *** were given insulin. The no insulin no glucose patients developed mild-to-moderate concentrations of ketone bodies. Some authors believe that well-controlled NIDDM patients do not require special treatment before and during surgery [5, 6, 8], and that insulin therapy should be considered only when intraoperative blood glucose concentrations exceed 11.1 mmol litre 1. This value was proposed because the renal threshold for glucose is mmol litre 1 in most patients with normal renal function [12]. Osmotic diuresis results in water and electrolyte losses, hyperosmolarity and impaired CNS function if this level is exceeded. The literature also suggests that impaired wound healing and strength, and impaired phagocytic function may occur when plasma glucose concentrations exceed 11.1 mmol litre 1 [3, 13, 14]. Intraoperative blood glucose control was good in most patients on the no insulin no glucose regimen in our study. These finding are consistent with those of Fletcher, Langman and Kellock for minor and moderate surgery [15]. Fasting was probably the cause of these stable blood glucose concentrations [16], because glucose given without insulin can result in marked hyperglycaemia [11, 17]. Fasting and a lower insulin : glucagon ratio were also responsible for the accelerated ketogenesis in the no insulin no glucose group. But, because of residual endogenous insulin secretion in our patients, the development of

5 202 British Journal of Anaesthesia ketone bodies was limited, and without any deleterious consequences. Ketogenesis was similar to that of normal subjects after short-term starvation [16]. Concentrations of lactate and pyruvate in the insulin-treated patients were increased, presumably as a result of insulin inhibiting gluconeogenesis and increased extrahepatic lactate production from glucose [16]. The lactate : pyruvate ratios were within normal limits for venous blood samples and were not affected by insulin administration. Thus, our results suggest that the no insulin no glucose regimen provided sufficient metabolic control during the operative period and might even be preferable as it avoids the risk of hypoglycaemia. The incidence of hypoglycaemia was 5 10 % whatever insulin regimen was used [10]. While our study was relatively short, it is always possible that some metabolic disturbance may occur later, but no major metabolic problem was identified during the postoperative period. The decrease in plasma C-peptide concentrations in groups B and C probably resulted from negative feedback by exogenous insulin on pancreatic insulin secretion. This decrease was even more significant in the group treated with large bolus doses of insulin. This inhibition develops over a few hours after insulin therapy is initiated in non-surgical situations [18]. The lack of increase in plasma GH in the no insulin no glucose group appears to be surprising, because there is usually a gradual increase in GH during stress, which parallels that of cortisol and takes place a few minutes after induction of anaesthesia [19]. It has been shown that elevated plasma ketone body concentrations decrease the release of GH [20, 21]. The increase in GH concentrations in groups B and C may also result from stimulation of GH release by exogenous insulin [20]. However, this difference in growth hormone concentration did not alter blood glucose. The deleterious effect of increased growth hormone on blood glucose in diabetics is probably not as pronounced as the effect of catecholamines and cortisol [22]. This lack of difference in blood glucose concentration may also be caused by administration of exogenous insulin, which counteracts the effect of growth hormone on glucose metabolism. Thus this study indicated that the no insulin no glucose regimen was a simple, effective way of controlling blood glucose in well-controlled NIDDM patients, provided that blood glucose was measured frequently and insulin used as appropriate. References 1. Alberti KGMM. Diabetes and surgery. Anesthesiology 1991; 74: Thompson J, Husband DJ, Thai AC, Alberti KGMM. Metabolic changes in the non-insulin-dependent diabetic undergoing minor surgery: Effect of glucose insulin potassium infusion. British Journal of Surgery 1986; 73: Milaskiewicz RM, Hall GM. Diabetes and anaesthesia: The past decade. British Journal of Anaesthesia 1992; 68: Hirsch IB, McGill JB. Role of insulin in management of surgical patients with diabetes mellitus. Diabetes Care 1990; 13: Podolsky S. Management of diabetes in the surgical patient. Medical Clinics of North America 1982; 66: Schade DS. Surgery and diabetes. Medical Clinics of North America 1988; 72: Thomas DJ, Platt HS, Alberti KGMM. Insulin infusion (GIK) in the treatment of type 2 (non-insulin-dependent) diabetes during the perioperative period. British Journal of Surgery 1986; 73: Hirsch IB, McGill JB, Cryer PE, White PF. Perioperative management of surgical patients with diabetes mellitus. Anesthesiology 1991; 74: Rosenstock J, Raskin P. Surgery. Practical guidelines for diabetes management. Clinical Diabetes 1987; 5: Raucoules-Aimé M, Ichai C, Roussel LJ, Romagnan MJ, Gastaud P, Dolisi C, Grimaud D. Comparison of two methods of i.v. insulin administration in the diabetic patient during the perioperative period. British Journal of Anaesthesia 1994; 72: Walts LF, Miller J, Davidson MB, Brown J. Perioperative management of diabetes mellitus. Anesthesiology 1981; 55: Elsas L, Rosenberg, L. Renal glycosuria. In: Earley L, Gottschalk C, eds. Strauss and Welt s Diseases of the Kidney, 3rd Edn. Boston: Little Brown, 1979; McMurray JF Jr. Wound healing with diabetes mellitus. Surgical Clinics of North America 1984; 61: Yue DK, McLennan S, Marsh M, Mai YW, Spaliviero J, Delbridge L, Reeve J, Turtle JR. Effects of experimental diabetes, uremia, and malnutrition on wound healing. Diabetes 1987; 36: Fletcher J, Langman MJS, Kellock TD. Effect of surgery on blood-sugar levels in diabetes mellitus. Lancet 1965; 2: Alberti KGMM, Thomas DJ. The management of diabetes during surgery. British Journal of Anaesthesia 1979; 51: Brenner WI, Lansky Z, Engleman RM, Stohl WM. Hyperosmolar coma in surgical patients: An iatrogenic disease of increasing incidence. Annals of Surgery 1973; 178: Liljenquist JE, Horwitz DL, Jennings AS, Chiasson JL, Keller U, Rubenstein AH. Inhibition of insulin secretion by exogenous insulin in normal man as demonstrated by C- peptide assay. Diabetes 1978; 27: Weissman C. The metabolic response to stress: An overview and update. Anesthesiology 1990; 73: Johnston DG, Davies RR, Prescott PW. Regulation of growth hormone secretion in man: A review. Journal of the Royal Society of Medicine 1985; 78: Quabbe HJ. Hypothalamic control of GH secretion: Pathophysiology and clinical implications. Acta Neurochirurgica 1985; 75: Raucoules-Aimé M, Lugrin D, Boussofara M, Gastaud P, Dolisi C, Grimaud D. Intraoperative glycaemic control in non-insulin-dependent and insulin-dependent diabetes. British Journal of Anaesthesia 1994; 73:

Comparison of two methods of i.v. insulin administration in the diabetic patient during the perioperative period

British Journal of Anaesthesia 1994; 72: 5-10 CLINICAL INVESTIGATIONS Comparison of two methods of i.v. insulin administration in the diabetic patient during the perioperative period. RAUCOULES-AIE, C.