Key Point

A nurse-driven insulin therapy protocol controlled blood glucose in ICU patients better than physician-administered insulin or a sliding scale insulin regimen.

SAN FRANCISCO—In the ICU, insulin therapy is a job better left to nurses, suggest study findings presented at the Society of Critical Care Medicine’s 35th Critical Care Congress. In the study, a nurse-driven protocol for giving intensive insulin therapy to medical and surgical ICU patients achieved more effective blood glucose control than did standard physician-initiated insulin therapy.¹

The protocol resulted in lower mean blood glucose levels, higher normoglycemia rates, and less chance of substantial hyperglycemia, said the authors. "The nurse-initiated program was more effective than physician-directed insulin therapy because it provided closer attention to detail and more aggressive glucose management," lead author Martha A. Naber, PharmD, told Pulmonary Reviews. Dr. Naber is a Clinical Pharmacy Specialist in the Medical ICU at Albany Medical Center Hospital in New York.

Dr. Naber and colleagues prospectively studied 20 consecutive medical and surgical ICU patients who received insulin via the nurse-driven protocol during a three-month period; following a slight protocol modification, prospective data were collected on nine more patients. The two groups were similar enough to be combined, however.

The nurse-initiated protocol included a physician order sheet with instructions for intensive insulin therapy, bedside blood glucose monitoring, the coordination of insulin therapy with carbohydrate delivery, and the management of glycemic events; an appendix to the order sheet provided specific instructions on insulin titration. The goal was to keep blood glucose levels in the normal range (70 to 120 mg/dL).

There were two comparison groups of medical/surgical ICU patients, those who received insulin therapy from physicians during a two-week period and those given insulin on a sliding scale basis during a one-week period. A sliding scale regimen usually involves short-acting insulin given four to six times daily based on regular capillary blood glucose measurements. Some believe that such regimens best regulate blood glucose and avoid glycemic episodes.

The final analysis covered 200 patient days and 3,509 blood glucose values. Blood glucose averaged 126 mg/dL among the patients assigned to the nurse-driven protocol, 153 mg/dL among those who received insulin from a physician, and 166 mg/dL in the sliding scale group.

The patients in the nurse-driven protocol were normoglycemic 51% of the time versus 32.4% and 16.8% of the time, respectively, for those who received physician-directed and sliding scale insulin therapy. The percentage of time with hypoglycemia was 1%, 0.4%, and 1.2% in the three groups, respectively; these rates were not significantly different. No adverse events were reported in the patients who developed hypoglycemia.

The authors concluded that their continued use of nurse-driven insulin therapy, first initiated at their hospital two years ago, is justified. However, "Further analysis using matched samples is required to generalize these findings to other institutions," they acknowledged.

—Timothy Begany

Reference
1. Naber MA, Timm EG, Thompson DR, et al. A comparison of an insulin infusion protocol versus standard practice in controlling blood glucose in the medical/surgical ICU. Presented at: annual meeting of the Society of Critical Care Medicine; January 8, 2006; San Francisco, Calif.

Return to table of contents