Hyperglycemia management strategy in the pediatric intensive care setting

摘要

Critical illness requiring intensive care is often caused by acute tissue injury (such as sepsis or trauma). The metabolic response to injury (acute metabolic stress response) is, in part, characterized by the cytokine-induced elaboration and secretion of counter-regulatory hormones (including, catecholamines, glucagon, and Cortisol). These hormones are known to counteract the anabolic effects of insulin and growth hormone (causing insulin and growth hormone resistance). They also promote glycogenosis, glu-coneogenesis, and other associated metabolic pathways which, in concert, lead to hyperglycemia in relative proportion to the magnitude of the injury insult and the physiologic reserve of the host. Progressively severe immunosuppression as well as an array of toxic cellular effects are known to be associated with increasing hyperglycemia (1,2), and a number of adult and pediatric studies have demonstrated increased mortality and morbidity in association with elevated blood glucose concentrations in a variety of medical and surgical critically ill patient populations (3-10). Furthermore, controlled randomized studies evaluating the use of intravenous, insulin to keep blood glucose concentrations within either tight (80-110 mg/dL) (3, 4) or even more liberal (<=150 mg/dL) (5, 6) control ranges have demonstrated or suggested significant improvement in clinical outcome in the adult intensive care setting. These findings have generated considerable interest among pediatric subspecialists as to whether similar outcome benefits might be achieved with insulin therapy to control hyperglycemia in critically ill children. In this issue of Pediatric Critical Care Medicine, Preissig et al. (11) reportthe results of clinical observations over a 6-month period following the initiation of a protocol designed to identify and treat hyperglycemia in the pediatric intensive care unit (PICU) setting. The authors deserve commendation for their careful approach in developing a pediat-ric-specifk standardized protocol to identify and treat hyperglycemia in critically ill children. According to this protocol, patients were identified for blood glucose surveillance based on the presence of predetermined clinical triggers (mechanical ventilation, vasoactive infusions, and continuous renal replacement therapy) or physician discretion. Insulin infusion was initiated intravenously for hyperglycemia (>140 mg/dL) and titrated to target a blood glucose concentration range between 80-140 mg/dL. Point-of-care monitoring of arterial, venous, and/or capillary serum blood glucose samples was carried out hourly while infusion rates were adjusted. Early caloric administration was initiated using Schoefield and White equations. The incidence of hyperglycemia observed was 20% and was associated with the clinical triggers selected as well as with high illness severity scores and increased PICU length of stay. The incidence of hypoglycemia was 6.7% (<60 mg/dL) and 4% (<40 mg/dL), respectively. This study raises a number of important points for consideration.