Management of diabetes during surgery with glucose-insulin-potassium infusion

One hundred and twenty-eight surgical operations in diabetic patients have been studied to assess the effectiveness, under routine clinical conditions, of a management regimen based on the use of glucose-insulin-potassium infusion (GIK). Forty-four non-insulin-dependent diabetic (NIDDM) and 41 insulin-dependent diabetic (IDDM) patients received GIK. Mean blood glucose on the day of operation was 9.3 +/- S.D. 2.2 mmol/l in NIDDM and 8.9 +/- 2.3 mmol/l in IDDM patients. Acceptable control on the day of operation (defined as mean blood glucose 5-12 mmol/l without hypoglycaemia) was achieved in 70 (82%) patients. Eleven of 15 failures were attributable to incorrect implementation of the protocol. Though 10 units Soluble insulin/500 ml 10% glucose (0.32 units/g glucose) was needed in 61% of patients, 26% required a higher and 13% a lower dose. Plasma potassium concentration did not change after 24 h of GIK infusion, but sodium concentration fell (136 +/- 5 to 132 +/- 5 mmol/l; p less than 0.01), with 12 of 32 patients having post-operative values less than 130 mmol/l. Forty-three NIDDM patients undergoing minor surgery were managed without insulin, and acceptable control was achieved in 40 (93%). We conclude that the regimen described is a satisfactory routine means of managing diabetes during surgery, but that optimal results depend on careful monitoring with appropriate alteration of therapy.