David Story^1,2,3

¹Austin Health, Melbourne, Australia
²University of Melbourne, Melbourne, Australia
³Trials Group, Australian and New Zealand College of Anaesthetists

The single most important aspect of managing diabetes during the perioperative period is to measure the blood sugar level frequently. Regular insulin (eg Actrapid) has been the mainstay of perioperative diabetes care as an infusion or intermittent boluses. Recently, patients with diabetes have started using new types of insulin and delivery devices 1. Two rapid acting insulins (lispro and aspart) are modified human insulins that have onset and offset after subcutaneous injection that is half that of regular insulin. Used as boluses before meals, rapid acting insulins are more physiological and are associated with less hyperglycaemia and less hypoglycaemia. As an alternative to intermittent injections computerized pumps with subcutaneous needles provide constant background infusion with manual boluses for meals. Another new insulin is glargine, which lasts for 24 hours and has a very flat profile. Glargine provides good basal levels of insulin as a daily bolus injection.

There are limited data on perioperative diabetes management. Perioperative insulin management has three components2: (1) basal; (2) nutritional; and (3) corrective. Glargine insulin has such a flat profile that patients can take their usual dose before surgery with minimal risk of hypoglycaemia while also avoiding ketosis. This is easier than running a glucose-insulin or glucose-insulin-potassium (GIK) infusion. Basal insulin can also be provided with continuous insulin pumps using rapid acting insulin such as aspart or lispro. Corrective insulin in the operating room can be with regular insulin by the intravenous route that hastens the onset. Subcutaneous rapid acting insulin is an alternative. Insulin pumps can also be used to provide corrective insulin doses but one risk with a pump is failure to recognize a malfunction by anaesthetists unfamiliar with these pumps (that is most of us).
There has recently been conflicting evidence in the critical care literature about very tight control of diabetes in ICU 3. A recent study in cardiac surgery found worse outcome in a tightly controlled group4 leading to some to conclude that very tight intraoperative control is unwarranted in the intraoperative period 3. An important risk with tight glucose control is hypoglycaemia. A reasonable trade off between avoiding hypoglycaemia and possible complications of hyperglycaemia is to aim for an intraoperative blood glucose level of 5 to 10 mmol/L. Using the same principles used in managing blood pressure the important components are first to frequently (at least hourly) measure the blood glucose and not to over treat hyperglycaemia and hypoglycaemia (avoid Alpine BSLs). 5% dextrose can be used to treat hypoglycaemia (2 ml/Kg) intravenously, but can also be given orally. Hyperglycaemia is treated with corrective doses of insulin. In the postoperative period the rapid acting insulins are likely to be much better for sliding scales than regular insulin.

1. Ahmed Z, Lockhart CH, Weiner M, Klingensmith G. Advances in diabetic management: implications for anesthesia. Anesth Analg 2005;100:666-9.

2. Asudani D, Calles-Escandon J. Inpatient hyperglycemia: slide through the scale but cover the bases first. J Hosp Med 2007;2 Suppl 1:23-32.

3. Van den Berghe G. Does tight blood glucose control during cardiac surgery improve patient outcome? Ann Intern Med 2007;146:307-8.

4. 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;146:233-43.