Diabetic ketoacidosis (DKA) occurs in 25-40% of children with new onset of type 1 diabetes, and occurs in children with established diabetes at a rate of approximately 8% per year. Cerebral edema (CE) is the most common serious complication of DKA in children, occurring in 1-5% of DKA episodes. CE has a high mortality rate (20-90%) and is the most common diabetes-related cause of death in children. The physiological mechanisms responsible for DKA-related CE are not understood. Previous studies by our group demonstrated that children at greatest risk for CE are those who present with greater dehydration and greater hypocapnia. The identification of these risk factors as those most strongly associated with CE suggests that central nervous system hypoperfusion may be an important factor in the pathogenesis of CE. Preliminary studies by our group in children with DKA and in an animal model of DKA support the hypothesis that cerebral hypoperfusion during untreated DKA may play a central role in causing CE and that hypersecretion of NaCI and water by the blood-brain barrier (BBB) may occur in this setting. We hypothesize that; (1) DKA results in stimulation of transport of NaCI and water by the BBB, (2) pharmacological interventions which block NaCI transport will decrease edema formation during DKA, (3) pharmacological interventions which rapidly restore cerebral perfusion will lead to more rapid normalization of cerebral metabolic abnormalities associated with DKA. In the current proposal, we aim to determine whether two interventions, bumetanide (a specific inhibitor of NaCl transport by the BBB), and mannitol (an osmotic agent which enhances cerebral perfusion) prevent or ameliorate CE in DKA and whether these agents improve cerebral metabolic abnormalities associated with DKA. We plan to conduct parallel studies in children with DKA and in an animal model of DKA to compare the effects of pharmacological interventions (bumetanide, mannitol and dexamethasone) on edema formation, cerebral perfusion and cerebral metabolism. In both the human and animal studies, we will use magnetic resonance (MR) imaging methods, (proton and phosphate MR spectroscopy, diffusion weighted imaging and perfusion weighted imaging) to study these interventions. Initial animal studies will provide information regarding the optimal treatment protocol to prevent or ameliorate CE in children with DKA. This initial animal data will form the basis for a randomized clinical trial of interventions for the prevention of CE in children with DKA which will be the main focus of the study.
