Validation of an insulin infusion nomogram for intensive glucose control in critically ill patients

Abstract

Study objective: To evaluate the effectiveness, safety, and associated patient outcomes of a simplified, nurse-directed insulin nomogram designed to achieve intensive blood glucose level control (target range 90-144 mg/dl).

Design: Prospective study with a retrospective control group.

Setting: A medical-surgical intensive care unit (ICU) in a quaternary care, university-affiliated hospital in an urban center.

Patients: Eighty-six critically ill adult patients (aged>or=18 yrs) requiring blood glucose control, with 42 in the retrospective control group and 44 in the prospective nomogram group.

Intervention: Control patients received insulin subcutaneously or intravenously based on ad hoc insulin sliding scales; nomogram patients received intravenous insulin at a rate specified by the nomogram, based on capillary blood glucose levels measured at the bedside.

Measurements and main results: Insulin infusion in the prospective patient group was titrated by the bedside nurse based on a predefined nomogram to attain the target blood glucose level. The retrospective control group was used as a comparison to assess the safety and effectiveness of the nomogram. Fewer patients in the nomogram (32%) than control (67%) group had a diagnosis of diabetes mellitus on admission. Overall, blood glucose levels in the nomogram group were within the target range 52% of the time versus 20% in the control group (p<0.001). Morning blood glucose levels were significantly lower compared with the control group (mean+/-SD 128+/-32 vs 176+/-50 mg/dl, p<0.001). Nomogram patients achieved target blood glucose levels faster than control patients (median 15 vs 66 hrs, p<0.0001). This improved blood glucose control remained statistically significant after adjusting for baseline differences in diabetes status. Hyperglycemia occurred less often in the nomogram than the control group (14% vs 53%, p<0.0001), and hypoglycemia occurred more often (3.8% vs 2.2%, p=0.004). The frequency of severe hypoglycemia was similar in both groups (0.2% vs 0.4%, p=NS). Such control required slightly more blood glucose checks/day in the nomogram group (7.1+/-1.5 vs 5.8+/-1.1, p<0.001). No significant reduction was observed in duration of vasopressor or antibiotic therapy or in length of stay in the ICU.

Conclusion: This study demonstrated that intensive blood glucose control is achievable using a nurse-directed nomogram. This improved control was achieved, regardless of diabetes status of the patient, without substantially compromising safety or increasing resource use.