BACKGROUND/MOTIVATION Glucose dependency renders the developing brain particularly vulnerable to the dysglyacemic insults of type 1 diabetes [1]. Whilst much of the research to date has focussed upon the impacts of diabetic dysglycaemia after diagnosis, recently there has been a focus upon the brain impacts of diabetic ketoacidosis (DKA) [2-5]. DKA can occur at any time but affects up to 15-70An understanding of the exact cellular mechanisms of cerebral oedema has remained elusive- due in part to the infrequency of the problem and the idiosyncratic nature of each clinical episode of resuscitation. In vitro and in vivo animal studies have thus been employed to identify the pathophysiology of cerebral oedema in DKA with both potential vasogenic and cytotoxic mechanisms being identified [2, 3, 11-13]. Previously we have reported data from a cohort of patients presenting with newly diagnosed type 1 diabetes during childhood with or without DKA [10]. In the primary report we identified transient regional brain volume changes that were associated with executive function performance 6 months thereafter. This was associated with lower levels of neuronal activity (as spectroscopically measured by N-acetylaspartate [NAA]) in frontal grey matter and basal ganglia. Other investigators have also shown lower levels of NAA levels in the basal ganglia in children during DKA [14]. The purpose of this study then is to expand our understanding of the sequence and nature of cellular injury in the brain in children with DKA. In order to do this, we have investigated a broader range of spectroscopically measured metabolites that are associated with neuronal function and integrity in a series of prospective and longitudinal assessments in our DKA study cohort. The data set was obtained from a cohort of newly diagnosed diabetes patients, some of whom had DKA. Measurements were made at 1, 5, 28 days and again at 6 months. Variables include volume, neuropsychological testing, MRI and single voxel MRS recorded from basal ganglia, frontal grey and white matter. The data-set resulted in a large number of variables with high ’noise’. There was little prior knowledge of relationships between variables, and so an exploratory methodology was undertaken with few modeling assumptions. Variable Name Type Abbreviation ---------------------- ------------ -------------- Frontal Grey Matter Region FGM Frontal White Matter Region FWM Left Basal Ganglia Region LBG myo-Inositol Metabolite mI N-Acetylaspartarte Metabolite NAANAAG Glutamine-Glutamate Metabolite GluGln Choline Metabolite GPCPCh : Variable Abbreviation Key
