INTRODUCTION
Diabetes mellitus is a relatively common condition, affecting approximately 4 to 7 per cent of the Western population. Patients with diabetes have an increased incidence of cardiovascular diseases that coupled with the microvascular complications of the disease which is translated into more surgical interventions. It is estimated that a patient with diabetes has a 50 per cent life time chance of requiring a surgery1 and the proportion of surgical patients with diabetes is around 20 per cent.2 Common insulin types and their pharmacokinetics are summarised in Table 1.1.
Careful assessment of diabetic patients prior to surgery is required because of their complexity as well as high risk of coronary heart disease which may be relatively asymptomatic compared to the non-diabetic population. Diabetes mellitus is also associated with increased risk of perioperative infections as well as postoperative cardiovascular morbidity and mortality.
Perioperative management of glycaemic control involves a complex interplay of the operative procedure, anaesthesia and additional postoperative factors such as sepsis, disrupted meal schedules, altered nutritional intake, hyperalimentation and emesis that can lead to unstable blood glucose levels. This review will discuss the perioperative management of glycaemia in patients with diabetes as well as management of glycaemia in critically ill surgical patients with diabetes.
PREOPERATIVE EVALUATION
Coronary heart disease is more common in individuals with diabetes. Diabetic patients have an increased risk of silent myocardial ischaemia. Therefore, prior assessment of cardiac risk is essential in diabetic patients before surgery. Other associated conditions, such as hypertension, obesity, chronic kidney disease, cerebro-vascular disease and autonomic neuropathy need to be assessed prior to surgery as these conditions may complicate anaesthesia as well as postoperative care.
All patients are required to have a disease history and careful physical examination, and further evaluation required in selected individuals. Key elements of the initial assessment should include the following:
- Determination of the type of diabetes, since type 1 diabetes patients are at a higher risk of diabetic ketoacidosis
- Long-term complications of diabetes mellitus, both macrovascular and microvascular complications
- Assessment of baseline glycaemic control, including frequency of monitoring, haemoglobin A1C levels (A1C) as well as fasting and mean blood glucose levels
- Assessment of hypoglycaemia including frequency, timing, awareness and severity
- Detailed history of diabetes therapy, including insulin type, dose and timing
- Other pharmacologic therapy, including type of medication, dosing and timing
- Characteristics of surgery, including when a patient should stop eating prior to surgery, type of surgery (major or minor, inpatient or outpatient), timing of the operative procedure and duration of the procedure
- Types of anaesthesia including epidural versus general anaesthesia: stress of general anaesthesia impairs glucose control by an increase in counter regulatory hormones in blood including catecholamine, cortisol and glucagon. Volatile anaesthetic agents also inhibit insulin secretion and increase hepatic glucose production. Epidural anaesthesia has minimal effects on glucose homeostasis and insulin resistance. On the other hand, regional anaesthesia carries a risk of 5hypotension to patients with autonomic neuropathy and a consequent risk of precipitating a cardiac event in those with ischemic heart disease. Patients with diabetes should be carefully assessed for intubation difficulties since there is an increased incidence of difficult intubation in these patients with a higher incidence of limited joint mobility syndrome or stiff joint syndrome.
PREOPERATIVE LABORATORY INVESTIGATIONS
Basic investigations should include a baseline electrocardiogram (ECG) and assessment of renal function using either serum creatinine (that underestimates renal function) or measurement of creatinine clearance using 24-hour urinary collection. ECG abnormalities are suggestive of previous myocardial infarction and chronic kidney disease which are risk factors for major postoperative cardiac events. Patients with abnormal ECG should be carefully considered for an exercise tolerance test and coronary angiography as the ECG has poor predictive value and 15 to 60 per cent of diabetic patients have asymptomatic coronary artery disease. A resting tachycardia and a lack of variability of RR interval are signs of autonomic neuropathy.
If not previously assessed within the last three months, A1C levels will permit the determination of long term glycaemic control and this is an important element in determining adequacy of current glycaemic management, especially insulin dose in patients. There is also some suggestion that elevated baseline glucose and A1C levels may predict a higher rate of postoperative infections.3 Further investigations including noninvasive cardiac testing should be considered on an individual basis. Poorly controlled insulin dependent diabetes with high HbA1C levels and impaired pulmonary function tests have been shown to co-exist.4 Clinically, poor cough decreased response to hypoxia and hypercapnia have been observed, consistent with respiratory function test results in such patients.5 This indicates the need for pulmonary function assessment in patients with pre-existent respiratory illness coupled with poorly controlled diabetes mellitus.
EFFECT OF SURGERY ON GLYCAEMIC CONTROL
Surgery and general anaesthesia cause a neuro-endocrine stress response with release of counter-regulatory hormones such as cortisol, epinephrine, glucagon and growth hormone as well as inflammatory cytokines such as interleukin-6 and tumour necrosis factor-alpha. These neuro-hormonal changes result in metabolic abnormalities including insulin resistance, decreased peripheral glucose utilisation, impaired insulin secretion, increased lipolysis and protein catabolism, leading to hyperglycaemia and even ketosis in some cases.6
The magnitude of counter-regulatory hormone release varies per individual and is influenced by the type of anaesthesia (general anaesthesia is associated with larger metabolic abnormalities as compared to epidural anaesthesia), the extent of the surgery (cardiovascular bypass surgery resulting in significantly higher degree of insulin resistance) and additional postoperative factors such as sepsis, hyperalimentation and steroid use. The hyperglycaemic response to these factors may be attenuated by the lack of caloric intake during and immediately after surgery, making the final glycaemic balance difficult to predict.
The general goals of perioperative management of patients with diabetes include:
- Avoidance of marked hyperglycaemia
- Avoidance of hypoglycaemia
- Maintenance of fluid and electrolyte balance
- Prevention of ketoacidosis.
Uncontrolled diabetes can lead to volume depletion from osmotic diuresis and life-threatening conditions such as diabetic ketoacidosis (DKA) or hyperosmolar hyperglycaemic state (HHS).
Patients with type 1 diabetes mellitus are insulin deficient and are prone to developing diabetic ketoacidosis and insulin should not be withheld at any time. Type 2 diabetes patients are susceptible to developing HHS that may lead to severe volume depletion and neurologic complications and they may develop ketoacidosis in the setting of extreme stress.
Hypoglycaemia is another potentially life-threatening complication of poor perioperative metabolic control. A short time (minutes) of severe hypoglycaemia (serum glucose concentration < 2.2 mmol/l) may induce arrhythmias, other cardiac events or transient cognitive deficits. Hypoglycaemia and subsequent neuroglycopaenia can be difficult to detect intraoperatively in patients having general anesthaetic or sedated patients postoperatively.
GLYCAEMIC TARGETS
Beyond avoidance of marked hyperglycaemia and hypoglycaemia, it is unclear how tight glucose control needs to be perioperatively. There is paucity of controlled trials on the benefits and risks of loose or tight glycaemic control in these patients, with the exception of patients in the intensive care unit or those who have had an acute myocardial infarction. Some studies show that achieving normoglycaemia (4.4–6.1 mmol/l) in cardiac surgery patients or those requiring postoperative surgical intensive care units (ICU) settings may reduce mortality. However, subsequent trials in mixed surgical and medical ICU patients have failed to show a benefit of such intensive control.6,7 In a multicentre randomised 7controlled trial (NICE-SUGAR), tight glycaemic control target (4.5–6.0 mmol/l) was associated with a significant increase in severe hypoglycaemia (blood glucose ≱ 2.2 mmol/l) and increased risk of death in the intensively treated group.8 The vast majority of the patients in these studies were not previously known to have diabetes but developed postoperative hyperglycaemia during the course of their ICU care. A subgroup analysis of two randomised trials assessing tight control in the ICU setting raised the possibility that the apparent benefits of tight control may not extend to patients with known diabetes.9 Despite some variability in proposed targets, these published guidelines collectively propose the achievement of reasonable normoglycaemia of glucose readings below 10 to 11 mmol/l in these patients. The American Diabetes Association (ADA) has endorsed pre-meal glucose goals of 7.8 mmol/l for general hospitalised patients with random glucose readings less than 10 mmol/l.10
EARLY PERIOPERATIVE PHASE
Several strategies exist to maintain target range glucose levels perioperatively but there is no consensus on the optimal strategy. Ideally, all patients with diabetes mellitus should have their surgery in morning as early as possible to minimise the disruption of their management routine whilst being fasted.
Type 2 Diabetes Treated with Diet Alone
Generally, patients with type 2 diabetes managed by diet alone do not require any therapy perioperatively. Stress of surgery may impair glycaemic control requiring pharmacological intervention. Supplemental short-acting insulin may be given as subcutaneous insulin boluses in patients whose glucose levels rise over the desired target. Blood glucose levels should be checked preoperatively and soon after the surgery. Dextrose-free intravenous solutions should be used for hydration if insulin is not given.8
Type 2 Diabetes Treated with Oral Hypoglycaemic Agents
Patients with type 2 diabetes who take oral hypoglycaemic drugs should continue their usual oral hypoglycaemic medications until the morning of surgery. On the morning of surgery, they should be advised to leave off their oral hypoglycaemic drugs, although, long-acting sulfonylurea should be discontinued 48 to 72 hours prior to surgery since they increase the risk of hypoglycaemia. However, inadvertent sulphonylurea treatment on the day of surgery should not postpone the procedure if careful blood glucose monitoring with continuous intravenous dextrose is undertaken. Metformin is contraindicated in conditions that increase the risk of renal hypoperfusion, hepatic insufficiency, lactate accumulation and tissue hypoxia. Metformin should be cautiously used in congestive heart failure as patients with heart failure have often co-exiting nephropathy. Thiazolidinediones should be avoided in patients with hepatic insufficiency. Thiazolidinediones may worsen fluid retention leading to peripheral oedema and could precipitate congestive heart failure. Newer agents like DPP-4 inhibitors and GLP-1 analogues could potentially alter GI motility and impact on the postoperative state.
Most patients with good metabolic control on oral agents will not need insulin for short surgical procedures. For patients who develop hyperglycaemia, supplemental short-acting insulin may be administered as subcutaneous boluses, based on frequently measured blood glucose levels that are often obtained on capillary “finger prick” samples. Most oral hypoglycaemic agents can be restarted after surgery when patients resume eating with the exception of metformin, which should be delayed in patients with suspected renal hypoperfusion until adequate renal function is confirmed. Patients who are discharged shortly after an outpatient surgical procedure need to be reminded about potential hyperglycaemia and the importance of seeking medical attention if this develops.
Type 1 or Insulin Treated Type 2 Diabetes
Generally, patients who use insulin can continue with subcutaneous insulin perioperatively (rather than a glucose, potassium, insulin infusion (GKI) or in some instances an insulin infusion alone) for procedures that are not long and complex. If the basal (long-acting) insulin is correctly titrated, it is reasonable to continue the long-acting insulin while the patient is fasting and having intravenous dextrose.
It may be prudent to reduce the night time intermediate-acting insulin on the night prior to surgery to prevent hypoglycaemia, otherwise, the usual basal insulin dose can be administered the night prior to surgery. If the basal insulin is usually given in the morning then the full morning dose is given in those with type 1 diabetes and about 50 to 100 per cent of the morning dose is administered in those with type 2 diabetes.2 Basal metabolic needs utilise approximately one half of an individual's insulin even in the absence of oral intake, highlighting the need to continue with basal insulin when not eating;11 it is mandatory in type 1 diabetes to prevent ketoacidosis.
TIMING OF PROCEDURE
Minor and Short Morning Operations
Procedures where breakfast is only likely to be postponed, patients may delay taking their usual morning insulin until after the surgery and before eating.
When short morning procedures are likely to be followed by omitting breakfast and lunch, the insulin dose is based on prior insulin therapy:
- If the patient is taking long-acting insulin glargine at night, then the usual basal dose is given the night prior to surgery
- If the patient is taking a morning insulin dose (short/rapid-acting or short-acting and intermediate), one third to half the total dose is given (total dose = intermediate plus short-acting) in the morning as intermediate insulin alone and short-acting insulin is omitted on the morning of the procedure. Re-start the usual short-acting insulin prior to the first meal. If there is a delay in the first meal or hyperglycaemia ensues, corrective short-acting insulin doses every 4 to 6 hours are given. It is estimated that in insulin sensitive patients, each unit of insulin lowers blood glucose between 2.85 to 5.6 mmol.6
- Patients on continuous subcutaneous insulin infusion (insulin pumps) may continue with their usual basal infusion rate
Long and Complex Procedures for Type 1 or Insulin Treated Type 2 Diabetes
Intravenous insulin is usually required for long and complex procedures (e.g. coronary artery bypass graft, renal transplant or prolonged neurosurgical operations). Studies comparing subcutaneous insulin administration versus intravenous infusion have found a marked increase in variability of the glucose concentration when using the subcutaneous route. This variability in plasma insulin has been attributed to the varying degrees of tissue perfusion associated with long and complex procedures. The safety of intravenous insulin infusion in highly monitored settings is established11,13,14 and their positive effect on non-glycaemic outcomes is well-documented.15 Insulin infusions are readily titrated because the half-life of intravenous insulin is short (i.e. 4–5 minutes), allowing for more precise glucose control and minimising hypoglycaemia.
Intravenous insulin regimens require close monitoring of blood glucose and electrolytes as well as appropriate interpretation by well-trained staff. Generally, insulin infusions should be started early in the morning prior to surgery to allow time to achieve glycaemic control. There are numerous intravenous insulin infusion algorithms published in the literature with insulin and glucose solutions being infused separately or as a combined glucose insulin potassium (GKI) solution.11,13 In general, the rate of insulin infusion is between 0.5 to 5 units/hr corresponding to the level of glycaemia.16 This usually equates to 0.3 units of insulin infused for each gram of glucose with up titration in insulin resistance.17
Glucose Insulin Potassium Infusion
In the normal clinical setting, particularly when the patient is not being intensively monitored in a high dependency unit, then the preferred method for providing intravenous insulin is the GKI18 infusion that is a single solution that includes 500 ml of 10 per cent dextrose, 10 mmol of potassium chloride and 16 units of short-acting insulin, which is usually infused at an initial rate of 100 ml/hr.17 The solution can be altered depending on the blood glucose measured every two hours by adding or subtracting four units of insulin. Potassium is added to prevent hypokalaemia and is monitored at six hourly intervals and serum potassium should be closely monitored in patients with pre-existing renal disease, especially those who are taking hyperkalaemia-inducing medications such as ACE inhibitors and Spironolactone.
This regimen is fundamentally safe because the insulin and glucose are given together and there is no risk of hypoglycaemia. The disadvantage is that frequent changes of the intravenous solution to increase or decrease the insulin dose may be required according to the two hourly blood glucose results. In patients with type 1 diabetes if the infusion is stopped then these patients can quickly become ketotic.11
Separate Insulin and Glucose Intravenous Solutions
With this regimen, dextrose is administered at approximately 5 to 10 gm of glucose/hr, and a separate insulin infusion is given using short-acting insulin. Most type 1 diabetes patients require an infusion at a rate of 1 to 2 units/hr with higher rates in insulin resistant type 2 diabetes. A rough guide to calculate the initial insulin rate from weight can be used (initial rate = 0.02 × kg−1 × h−1).2 The rate can be increased for hyperglycaemia, if there is evidence of insulin resistance or for pre-admission high insulin requirements. On the other hand, the rate will need to be reduced in those with hepatic or renal failure. The initial insulin infusion can be decreased or increased by 0.5 unit/hr to maintain the glucose targets. The short half-life of intravenous insulin means that if the infusion is stopped then rapid hyperglycaemia ensues with potential ketosis. Conversely, too high an infusion rate (compared to the independent dextrose infusion) may cause rapid and profound hypoglycaemia. Therefore, whilst it is more flexible this practice should be restricted to high dependency units.
LATE POSTOPERATIVE PHASE
Generally, the preoperative diabetes treatment regimen (oral agents or oral agents plus insulin) may be reinstated once the patient is eating well. However, there are a few caveats for certain oral hypoglycaemic agents.
- Metformin therapy should not be re-started in patients with renal insufficiency, significant hepatic impairment or congestive heart failure
- Sulfonylurea therapy stimulates insulin secretion and may cause hypoglycaemia; they should be started only after eating has been well established. A step-up approach can be used for patients who were taking high dose sulfonylureas prior to their admission, starting at low doses and adjusting them until the optimal dose is reached
- Thiazolidinediones should not be used if congestive heart failure, problematic fluid retention or liver function abnormalities develop postoperatively.
Insulin infusions should be continued in patients who do not resume eating postoperatively. Once solid food is tolerated the patient can be switched to subcutaneous insulin and then insulin infusion can be discontinued. Patients, who were taking subcutaneous insulin in the early postoperative 12phase, before alimentation is restarted, should continue this treatment along with intravenous 10 per cent dextrose to prevent hypoglycaemia.
SUBCUTANEOUS BOLUS INSULIN DOSES
Frequently, small doses of short-acting insulin are often used to correct elevated glucose levels. Subcutaneous insulin boluses, especially when used as the sole method of insulin delivery can be problematic as they result in wide serum glucose fluctuations; small doses of short-acting insulin are often given when hyperglycaemia has resulted. Subcutaneous insulin boluses should never be the sole insulin regimen in type 1 diabetes, because ketosis can occur before significant hyperglycaemia is present. Small doses of short-acting insulin can be given before meals and at bedtime or alternatively, in patients who ate nothing by mouth, every six hours, usually to supplement basal and prandial insulin to maintain near-normoglycaemia. This corrective insulin should be given when glucose levels are more than 8.3 mmol/l and the doses depend upon the degree of insulin sensitivity of the patient. In lean elderly patients with type 1 diabetes or individuals with renal or liver failure are usually considered to be “insulin sensitive”, while obesity or treatment with glucocorticoids are usually associated with an insulin resistant state.
SPECIAL CONSIDERATIONS
Glucocorticoid Therapy
Glucocorticoids are used for the treatment of many disorders and are often given in stress, its doses are given perioperatively to prevent adrenal insufficiency. Glucocorticoids can worsen pre-existing diabetes mellitus and may precipitate steroid-induced hyperglycaemia in others. Glucocorticoids increase hepatic glucose production, decrease peripheral glucose utilisation by muscles and influence B cell function. Therefore, glucocorticoids have a post-prandial hyperglycaemic effect on those with diabetes and on those who are predisposed to developing diabetes. Treatment with glucorticoids rarely leads to ketoacidosis or hyperosmolar hyperglycaemic syndrome.2 To circumvent the delayed efficacy of oral hypoglycaemic agents, insulin is often preferred to control hyperglycaemia in those patients. Patients on a single daily morning dose of steroids develop peak hyperglycaemia 4 to 12 hours later, so the impact on fasting and post-breakfast hyperglycaemia tends to be less. Insulin requirements are often initially underestimated in steroid therapy but glucose control is usually best established with prandial insulin.2 A variable rate insulin and glucose infusion may be appropriate in patients receiving high dose or variable dose steroids. Oral hypoglycaemic medications can be used when the steroids dose is low and constant; however, insulin is often necessary for those whose glucose levels are 13elevated (>11 mmol/l).19 Twice daily intermediate-acting insulin with short-acting insulin are given as an occasional bolus may be needed to achieve glucose control. A two to three fold increase in the total daily insulin dose is frequently needed in patients on high dose steroid therapy.
Hyperalimentation
Total parenteral nutrition (TPN) and nasogastric/enteral feeds are commonly used in patients who are malnourished or severely ill. TPN, especially in those with type 2 diabetes mellitus, will often increase the blood glucose and about 77 per cent will need insulin with an average daily dose of 100 units/day.20
A variable rate intravenous insulin infusion can be used when the patient is initiated on TPN.19 Once a stable infusion rate of TPN is established then the calculated daily requirement of insulin is added directly to the TPN. Subcutaneous insulin boluses of short-acting insulin may be used if insulin infusion is not feasible.
For nasogastric feeds administered continuously over 24 hours, a variable intravenous insulin infusion can be used initially. The total dose administered over 24 hours can be used to calculate the dose for subcutaneous twice daily intermediate-acting insulin, once or twice daily insulin glargine.2 Subcutaneous corrective dose of short-acting insulin for every four to six hours may be required whilst awaiting for insulin glargine to become effective.2 Changes in the insulin regimen must precede any changes in nasogastric feeding regimens (i.e. changes from 24 hour infusion to three times daily bolus feeds). Thus, good communication between the surgeon, dietician and the person managing diabetes care is important. For intermittent enteral feeding, one daily isophane insulin can be used in combination with short-acting insulin doses prior to each bolus and calculated according to a pre-meal and two-hours post-prandial capillary blood glucose measurement.2
Preoperative Carbohydrate Loading
Postoperative hyperglycaemia due to metabolic stress and insulin resistance is associated with increased morbidity and mortality.21 In non-diabetic patients, avoiding preoperative fasting substantially reduces postoperative stress and insulin resistance. A preoperative oral carbohydrate load has shown to improve postoperative glycaemic control, most likely by inducing endogenous insulin release before the onset of surgery. This sets the metabolic state of the patient in a fed state rather than in a fasted state at the time of surgery.22 Metabolic reactions to surgical stress are consequently markedly reduced resulting in a reduced risk for hyperglycaemia during postoperative nutrition, retained lean body mass, improved muscle strength and nitrogen economy. Although, a considerable number of patients going 14through surgery suffer from diabetes, this patient group has been denied the preoperative carbohydrate loading drink because of fear of slow gastric emptying and impaired glycaemic control. However, in a small cohort of patients with well controlled type 2 diabetes an oral carbohydrate load three hours before anaesthesia induction was shown to be safe without any risk for postoperative hyperglycaemia or aspiration pneumonia.23 Further studies are needed to determine whether intake of a preoperative carbohydrate drink has similar effects on metabolism and surgical outcomes among patients with diabetes as demonstrated for non-diabetic patients.23
Patients Undergoing Coronary Artery Bypass Graft Surgery (CABG)
For coronary artery bypass graft surgery procedures, the insulin requirements may increase up to ten fold, especially after recovery from the hypothermia, necessitating an increase in the initial insulin rate by three to five times.24 This increase in insulin requirement is related to the technique of CABG which requires large volumes of dextrose solutions to prime and profuse bypass pump, reversal of hypothermia and frequent use of vasopressor agents postoperatively for stabilisation of haemodynamics. It has also been shown that diabetic patients undergoing CABG who had GKI infusion just before the induction of anaesthesia and continued for 12 hours after surgery had a significantly lower incidence of atrial fibrillation, shorter length of stay and at two years, had less recurrent myocardial ischaemia, fewer wound infections and lower mortality compared to intermittent subcutaneous insulin therapy.25 A reasonable goal of blood glucose below 10 mmol/l is achievable using an intravenous insulin infusion. The rate of this infusion is dependant on the frequently measured blood glucose and the degree of insulin resistance. One suggested formula is:
Rate of insulin infusion = (blood glucose in mmol/l-3.3) × multiplier.26
This multiplier could be 0.02 and increased when steroid therapy is used (e.g. 0.04). An increase of blood glucose above target necessitates an increase of the multiplier by 0.01 while a reduction in the multiplier by the same amount is needed when blood glucose falls below 5.5 mmol/l. A separate intravenous dextrose infusion will need to be used when blood glucose is below 7.7 mmol/l.27
Intensive Insulin Therapy in Critically Ill Surgical Patients
Hyperglycaemia associated with critical illness (also called stress hyperglycaemia or stress diabetes) is a consequence of many factors 15including increased cortisol, catecholamines, glucagon, growth hormone, gluconeogenesis and glycogenolysis.28 Insulin resistance may also be a contributing factor, since it has been demonstrated to be increased in more than 80 per cent of critically ill patients.29 Hyperglycaemia was previously considered an adaptive response essential for survival and was not routinely controlled in ICU.30 However, more recent evidence from observational studies indicates that uncontrolled hyperglycaemia was associated with poor outcomes and this has prompted efforts to routinely correct and prevent hyperglycaemia in critically ill patients. However, the evidence to date do not prove that hyperglycaemia is the cause of the poor clinical outcomes as hyperglycaemia may simply be marker of severe illness.
Critically, ill medical and surgical patients who are hyperglycaemic have a higher mortality rate than patients who are normoglycaemic.31 Patients who are hyperglycaemic following trauma have an increased mortality rate, hospital length of stay, ICU length of stay, and incidence of nosocomial infection.32 Hyperglycaemia is also associated with worse neurologic outcomes and increased intracranial pressure in patients with traumatic brain injury.33
A single-centre trial (the Leuven surgical trial) randomly assigned surgical ICU patients to receive intensive insulin therapy (IIT) or conventional blood glucose management.34 IIT was defined as an insulin infusion targeting blood glucose of 4.4 to 6.1 mmol/l; whereas, the conventional blood glucose management targeted blood glucose of 10 to 11.1 mmol/l. It was found that ICU as well as hospital mortality was significantly lowered in IIT group. IIT also decreased critical illness polymyoneuropathy, acute renal failure, transfusion requirement and blood stream infections, although hypoglycaemia was more frequent in this group.
In contrast, in larger multicentred studies Normoglycaemia in Intensive Care Evaluation Survival Using Glucose Algorithm Regulation (NICE-SUGAR) trial, Volume Substitution and Insulin Therapy in Severe Sepsis (VISEP) trial and the Glucontrol trial showed that there was a significantly increased rate of severe hypoglycaemic as well as increased mortality and morbidity in patients who received IIT compared to those who received conventional glucose control.8,35,36 Meta-analyses have been performed in an effort to consolidate the data from numerous randomised trials37,38 comparing ITT to less stringent glycaemic control in mixed medical and surgical ICU patients have shown that those who received ITT had a similar mortality.37 In summary, mixed populations of critically ill medical and surgical patients increased the incidence of severe hypoglycaemia with either an increase or no effect on mortality when compared to the more permissive blood glucose ranges.
Hypoglycaemia is the most common adverse effect of IIT. Hypoglycaemia can lead to seizures, brain damage, depression and cardiac arrhythmias, and it is an independent risk factor for death38,39 Based on the above observations, a blood glucose target of 7.7 to 10 mmol/l in critically ill patients, rather than 16more stringent target (e.g. 4.4–6.1 mmol/l) or a more liberal target (e.g. 10–11.1 mmol/l) is recommended. This range avoids marked hyperglycaemia, while minimizing the risk of both iatrogenic hypoglycaemia and other harms associated with a lower blood glucose target.
CONCLUSIONS
Perioperative management of patients with diabetes is dynamic, being influenced by predictable and sometimes unpredictable events. There are many specific issues that require further investigation, since there is only sparse data in the literature to guide decision-making. Decisions of which regimens to follow and when will depend upon individual patients and the clinician's own judgment and experience.
- Root HF. Preoperative medical care of the diabetic patient. Postgrad Med 1966; 40 (4): 439–44.
- Clement S, Braithwaite SS, Magee MF, et al. Management of diabetes and hyperglycemia in hospitals. Diabetes Care 2004; 27 (2): 553–91.
- Dronge AS, Perkal MF, Kancir S, et al. Long-term glycemic control and postoperative infectious complications. Arch Surg 2006; 141 (1): 375–80.
- Ramirez LC, Dal Nogare A, Hsia C, et al. Relationship between diabetes control and pulmonary function in insulin-dependent diabetes mellitus. Am J Med 1991; 91: 371–6.
- Sandler M, Bunn AE, Stewart RI. Cross-section study of pulmonary function in patients with insulin-dependent diabetes mellitus. Am Rev Respir Dis 1987; 135 (1): 223–9.
- American Diabetes Association. Intensive Diabetes Management. 2nd edition. ADA; Alexandria VA: 1998.
- Smiley DD, Umpierrez GE. Perioperative glucose control in the diabetic or nondiabetic patient. South Med J 2006; 99: 580–9.
- The NICE-SUGAR Study Investigators, Finfer S, Chittock DR, et al. Intensive versus conventional glucose control in critically ill patients. N Engl J Med 2009; 360 (13): 1283–97.
- Van den Berghe G, Wilmer A, Milants I, et al. Intensive insulin therapy in mixed medical/surgical intensive care units: Benefit versus harm. Diabetes 2006; 55 (11): 3151–9.
- Moghissi ES, Korytkowski MT, DiNardo M, et al. American Association of Clinical Endocrinologists and American Diabetes Association consensus statement on inpatient glycemic control. Diabetes Care 2009; 32 (6): 1119–31.
- Hoogwerf BJ. Perioperative management of diabetes mellitus: How should we act on the limited evidence? Cleve Clin J Med 2006; 73 Suppl 1: S95–9.
- Marks JB. Perioperative management of diabetes. Am Fam Physician 2003; 67 (1): 93–100.
- Van den Berghe G, Wilmer A, Hermans G, et al. Intensive insulin therapy in the medical ICU. N Engl J Med 2006; 354 (4): 449–61.
- Alberti KG, Gill GV, Elliott MJ. Insulin delivery during surgery in the diabetic patient. Diabetes Care 1982; 5 Suppl 1: 65–77.
- Furnary AP, Zerr KJ, Grunkemeier GL, et al. Continuous intravenous insulin infusion reduces the incidence of deep sternal wound infection in diabetic patients after cardiac surgical procedures. Ann Thorac Surg 1999; 67 (2): 352–60.
- Peters A, Kerner W. Perioperative management of the diabetic patient. Exp Clin Endocrinol Diabetes 1995; 103 (4): 213–8.
- Jacober SJ, Sowers JR. An update on perioperative management of diabetes. Arch Intern Med 1999; 159 (20): 2405–11.
- Thomas DJ, Platt HS, Alberti KG. Insulin-dependent diabetes during the perioperative period. An assessment of continuous glucose-insulin-potassium infusion and traditional treatment. Anaesthesia 1984; 39 (7): 629–37.
- Park RH, Hansell DT, Davidson LE, et al. Management of diabetic patients requiring nutritional support. Nutrition 1992; 8 (5): 316–20.
- Thorell A, Nygren J, Ljungqvist O. Insulin resistance: a marker of surgical stress. Curr Opin Clin Nutr Metab Care 1999; 2 (1): 69–78.
- Svanfeldt M, Thorell A, Nygren J, et al. Postoperative parenteral nutrition while proactively minimizing insulin resistance. Nutrition 2006; 22 (5): 457–64.
- Gustafsson UO, Nygren J, Thorell A, et al. Preoperative carbohydrate loading may be used in type 2 diabetes patients. Acta Anaesthesiol Scand 2008; 52 (7): 946–51.
- Hoogwerf BJ. Perioperative management of diabetes mellitus: striving for metabolic balance. Cleve Clin J Med 1992; 59 (5): 447–9.
- Lazar HL, Chipkin SR, Fitzgerald CA, et al. Tight glycemic control in diabetic coronary artery bypass graft patients improves perioperative outcomes and decreases recurrent ischemic events. Circulation 2004; 109 (12): 1497–502.
- Davidson PC, Steed RD, Bode BW. Glucommander: a computer-directed intravenous insulin system shown to be safe, simple, and effective in 120,618 h of operation. Diabetes Care 2005; 28 (10): 2418–23.
- Shine TS, Uchikado M, Crawford CC, et al. Importance of perioperative blood glucose management in cardiac surgical patients. Asian Cardiovasc Thorac Ann 2007; 15 (6): 534–8.
- McCowen KC, Malhotra A, Bistrian BR. Stress-induced hyperglycaemia. Crit Care Clin 2001; 17 (1): 107–24.
- Saberi F, Heyland D, Lam M, et al. Prevalence, incidence, and clinical resolution of insulin resistance in critically ill patients: an observational study. JPEN J Parenter Enteral Nutr 2008; 32 (3): 227–35.
- Robinson LE, van Soeren MH. Insulin resistance and hyperglycemia in critical illness: role of insulin in glycaemic control. AACN Clin Issues 2004; 15 (1): 45–62.
- Krinsley JS. Association between hyperglycaemia and increased hospital mortality in a heterogeneous population of critically ill patients. Mayo Clin Proc 2003; 78 (12): 1471–8.
- Sung J, Bochicchio GV, Joshi M, et al. Admission hyperglycaemia is predictive of outcome in critically ill trauma patients. J Trauma 2005; 59 (1): 80–3.
- Jeremitsky E, Omert LA, Dunham CM, et al. The impact of hyperglycaemia on patients with severe brain injury. J Trauma 2005; 58 (1): 47–50.
- van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in the critically ill patients. N Engl J Med 2001; 345 (19): 1359–67.
- Brunkhorst FM, Engel C, Bloos F, et al. Intensive insulin therapy and pentastarch resuscitation in severe sepsis. N Engl J Med 2008; 358 (2): 125–39.
- Preiser JC, Devos P, Ruiz-Santana S, et al. A prospective randomised multi-centre controlled trial on tight glucose control by intensive insulin therapy in adult intensive care units: the Glucontrol study. Intensive Care Med 2009; 35 (10): 1738–48.
- Wiener RS, Wiener DC, Larson RJ. Benefits and risks of tight glucose control in critically ill adults: a meta-analysis. JAMA 2008; 300 (8): 933–44.
- Krinsley JS, Grover A. Severe hypoglycemia in critically ill patients: risk factors and outcomes. Crit Care Med 2007; 35 (10): 2262–7.