Abstract

A dose-defining insulin algorithm for attainment and maintenance of glycemic targets during therapy of hyperglycemic crises

Author(s): Radha Devi, Geetha Selvakumar, Lisa Clark, Carol Downer & Susan S Braithwaite

According to current standards of care, insulin therapy of the hyperglycemic crises diabetic ketoacidosis and hyperglycemic hyperosmolar state may be ordered using a combination of weight-based and qualitative rules for the initiation and adjustment of the rate of intravenous insulin infusion. The early hours of treatment are often managed with a fixed‑dose insulin regimen, such as 0.1–0.14 units/kg/h of insulin. Higher-dose insulin protocols, once used routinely, were replaced safely and effectively in the 1970s by low-dose regimens, possibly with a reduction in late hypoglycemia and hypokalemia. In a pediatric study comparing 43 cases of cerebral edema to 169 matched control subjects, the dose of insulin in the first 2 h was significantly associated with the risk of cerebral edema (p < 0.02 for trend over categories of insulin dose). Even with the use of low-dose insulin therapy, hypokalemia and hypoglycemia continue to occur. In one series, 13% (18 out of 144) of diabetic ketoacidosis patients had blood glucose <60 mg/dl at some time during insulin infusion. In hyperinsulinemic-euglycemic clamp studies of nondiabetic subjects, after interruption of infusion of intravenous regular insulin infusions, the time required for deactivation from maximum to half maximum effect upon the glucose disposal rate was 63 ± 5 min. Insulin action, although declining in effect, may persist for at least 90 min after interruption of insulin infusion. We hypothesize that late glycemic excursions and possibly hypokalemia after the first 4 h of treatment might be reduced by replacing a fixed-dose rule with a standardized, but dynamic, insulin titration protocol in the early hours of treatment, to be used in conjunction with attention to other aspects of care.