Cerebral edema is the most frequent and most feared serious complication of diabetic ketoacidosis (DKA) in children. Furthermore, it is the most important cause of death in this vulnerable pediatric population with a common chronic illness. Overt, symptomatic cerebral edema occurs in approximately 1% of pediatric DKA episodes, but recent data suggest that mild, or even apparently asymptomatic, cerebral edema occurs more frequently, likely in most children with DKA. Furthermore, recent pilot data suggest that even apparently uncomplicated DKA may result in subtle cerebral injury and long-term neurocognitive alterations. The relationship of fluid therapy for DKA and risk of cerebral injury and cerebral edema is not well understood and has been debated for decades. Some investigators have suggested that cerebral edema might be caused by rapid administration of intravenous fluids resulting in a rapid decline in serum osmolality and osmotically-mediated cell swelling. Recent data from our group, however, suggest that DKA may be associated with substantial reductions in cerebral blood flow and that DKA-related cerebral edema may be more similar to cerebral edema resulting from stroke or other hypoxic/ischemic cerebral injuries than to osmotically-mediated edema. Slower infusion of intravenous fluids has been advocated by some investigators, with the goal of avoiding rapid osmotic change. Conversely, more rapid infusion of fluids may be beneficial if cerebral edema is related to cerebral hypo-perfusion, resulting in more rapid restoration of normal cerebral blood flow. Because of the lack of prospective, appropriately controlled studies investigating the relationship between fluid treatment protocols and risk of cerebral edema, substantial variation exists in currently-used pediatric DKA treatment protocols across the United States. In the proposed project, we aim to study the effects of variations in fluid infusion protocols for DKA on several neurological outcomes including the risk of mental status declines during DKA treatment (which have been associated with cerebral edema in previous studies), the risk overt, symptomatic cerebral edema, and long- term neurocognitive function. We plan to utilize the resources of the Pediatric Emergency Care Applied Research Network (PECARN) to conduct a large, randomized controlled trial comparing four fluid protocols (differing in rate and sodium content) for DKA treatment using a factorial design. Children presenting with DKA will be randomly assigned to one of the four protocols. Mental status during DKA treatment will be assessed hourly and the frequency of declines in mental status, frequency of overt, symptomatic cerebral edema and frequency of other DKA complications will be recorded. Children will return for further evaluation three months after the DKA episode to assess memory function and IQ. The effects of both fluid infusion rate and sodium content of infused fluids on risks of both short term and long term neurological injury will be determined. The unique opportunity to conduct this study through the PECARN network will allow us to enroll the large sample size (>1500 children with DKA) necessary to answer this important clinical question using rigorous, prospective standardized methods.
Children with diabetes may develop a life-threatening condition, diabetic ketoacidosis (DKA). Most deaths caused by DKA are related to swelling of the brain (cerebral edema) during DKA treatment, and recent data indicates that even children who survive DKA apparently without complications may have subtle brain injuries which may result in learning and memory problems. Whether DKA treatment causes or contributes to cerebral edema is not known, nor do we know whether one treatment regimen is safer than others. In the proposed study, we plan to compare four different DKA treatment regimens to determine which regimen results in the least risk of brain injury.