Insulin subcutaneous application vs. continuous infusion for postoperative blood glucose control in patients with non-insulin-dependent diabetes mellitus

Frequently, the use of insulin is considered for metabolic control in postoperative patients with non-insulin-dependent diabetes mellitus (NIDDM). We sought to determine the best method for control of glucose in NIDDM non-insulin patients during postoperative care.

Two algorithms were developed: subcutaneous administration of insulin (SC), and continuous intravenous infusion (IV). A randomized, controlled clinical trial was designed. In addition, both experimental groups were compared with a non-concurrent routinely managed group (RM) with insulin administration under no predetermined algorithm. Eligible patients were those subjected to major surgery under general anesthesia or spinal blockade. They were followed for 48 h after surgery. Target variables were capillary and central blood glucose, insulin dose administered, urine glucose and ketone strip determination, and development of hyper- or hypoglycemia.

A total of 62 patients were studied (RM = 25, SC = 19, IV = 18). Results for both experimental algorithms were similar except for the IV group that required less insulin per hour compared to SC (0.64 vs. 0.34 U/h; p = 0.0003). The RM control group showed poor control in all capillary glucose measurements (194.9 +/- 26.8 mg/dL) compared with the two experimental algorithms (SC = 129.9 +/- 21; IV = 131.6 +/- 20.4) (p <0.05). More hyperglycemia events appeared in the RM group (p = 0.016). Only one hypoglycemia event occurred in the IV group.

Postoperative control of NIDDM is similar with both tested methods. The use of any of the algorithms studied improves metabolic control substantially because it standardizes postoperative management of the diabetic patient with timely determination of capillary blood glucose and insulin administration. However, IV administration has the advantage of accomplishing adequate control with a smaller insulin dose.