Pediatric Diabetes 2006: 7: 191–195 All rights reserved # 2006 The Authors Journal compilation # 2006 Blackwell Munksgaard Pediatric Diabetes Editorial Strategies to diminish the danger of cerebral edema in a pediatric patient presenting with diabetic ketoacidosis It is appalling that with the advances in critical care management, the incidence, morbidity, and mortality of the most dreaded complication of therapy for diabetic ketoacidosis (DKA), cerebral edema, are unacceptably high (1). Accordingly, our current approach to therapy in these patients requires revisions to minimize the risk of developing cerebral edema. In this issue of the journal, Agus and coworkers (2) emphasized the importance of measuring the PCO 2 . They used the end-tidal PCO 2 to calculate the arterial pH and bicarbonate (HCO 3 2 ) concentration (P HCO3 ) to detect a failure of insulin to act later in therapy. While of some value, this does not help us decide how to minimize the risk of developing cerebral edema. Our view of the importance of measuring the PCO 2 is different. We use the brachial venous PCO 2 to provide insights into the risk of developing cerebral dysfunction (3). To place this in context, risk factors for developing cerebral edema will be summarized. This will be followed by a discussion of guidelines for intravenous therapy. To plan this therapy, a quantitative assessment of the deficits of sodium (Na 1 ) and HCO 3 2 in the extracellular fluid (ECF) compartment is needed (4). Based on this new information, two amendments to the conventional therapy in children with DKA are pro- vided. We also emphasize that therapy must be designed for an individual patient and that recipes based on ‘one size fits all’ need to be amended (5). Risk factors for cerebral edema: physiological considerations Cerebral edema occurs because excess water – which normally accounts for 80% of the brain volume – is present in brain’s intracellular fluid (ICF) and/or ECF compartment(s) (3). It is obvious that there would be a rise in intracranial pressure if the volume of one of these compartments expands without an equivalent decrease in the volume of the other compartment. Hence, risk factors for cerebral edema should be examined in two categories, those that lead to expansion of brain cell volume and those that lead to retention of more water in the ECF within the skull. Increased brain cell volume Water will be retained in cells if these cells gain ‘effective’ osmoles and/or if the effective osmolality of plasma (P Effective osm ) declines (P Effective osm ¼ 2P Na 1 P Glucose , where values are in mmol/L terms and P Na and P Glucose represent the concentrations of Na 1 and glucose in plasma). Cells may gain effective osmoles (Na 1 ) if insulin, which activates their Na 1 /H 1 exchanger, crosses the blood–brain barrier (BBB), and achieves sufficiently high concentrations in the ECF of the brain (6). This could occur following a bolus of insulin (3) (hence, it should not be given). Similarly, a fall in the P Effective osm should be prevented; this is achieved by infusing fluids with the appropriate tonicity. Increased brain ECF volume During intravenous therapy, one must avoid an undue rise in the hydrostatic pressure and a large decline in the colloid osmotic pressure in capillaries at the BBB. To achieve these aims, one must be careful about the rate and the volume of infused saline. To guide decision making, a quantitative assessment of the deficit of Na 1 and measurement of the brachial venous PCO 2 are needed. This later measurement provides information about the effectiveness of the bicarbonate buffer system (BBS) in skeletal muscle cells and, thereby, the H 1 load that is bound to intracellular proteins in vital organs, e.g., brain cells (4). Quantitative assessment of the deficit of Na 1 In large multi-center studies on patients with DKA (7, 8), patients who developed cerebral edema could not be separated from those who did not based on the magnitude of their ECF deficits of Na 1 and 191