About 20 percent of childhood deaths from diabetes mellitus are related to diabetic ketoacidosis (DKA). The appropriate management of DKA includes lowering the serum glucose level and reversing abnormalities in body fluids and electrolytes. However, one complication of DKA treatment is cerebral edema, which can occur if fluid is administered too fast. Some studies report that a rate of more than 4 L per m² per 24 hours is more likely to cause cerebral edema. Recent recommendations have called for lower amounts of replacement fluid. Felner and White discuss a simplified protocol for the management of DKA in children with type 1 diabetes and review complication rates.

Children from before and after the protocol change were included in the study if they had a discharge diagnosis of DKA. Children in the first group (those who were managed before the protocol change) were given a bolus of 20 mL per kg of normal saline administered over 30 to 45 minutes. They then were given intravenous human insulin administered at a rate of 0.1 U per kg per hour. In the first group, fluids were administered after a fluid deficit was calculated. The clinically determined percentage of dehydration was multiplied by the patient's weight in kilograms. This figure plus 1.5 times the normal maintenance fluid amount was determined to be the amount of fluid needed in 24 hours, with one half given in the first 12 hours and one half in the next 24 hours.

The second group of children (those who were treated after the protocol change) were given the same normal saline bolus and the same amount of insulin initially. However, their fluid requirements were calculated as follows: fluid was given intravenously at 2.5 times the maintenance rate for 24 hours (or less, if acidosis resolved sooner); after 24 hours, the rate was decreased to 1 to 1.5 times the maintenance rate. Group 2 received fluid with 0.675 percent sodium chloride after the initial fluid bolus. Both groups received subcutaneous insulin and food by mouth after their venous pH reached at least 7.3. Potassium (potassium chloride or potassium phosphate) was also administered, depending on serum levels of potassium, phosphate, and calcium. In both groups, dextrose was added to the intravenous fluids when the blood glucose concentration was less than 300 mg per dL (16.65 mmol per L) or if the glucose level decreased by 350 mg per dL (19.45 mmol per L) in less than six hours (or more than 100 mg per dL [5.55 mmol per L] per hour). The fluid administration rate was decreased if patients developed signs or symptoms of cerebral edema such as headache, change in level of consciousness, or bradycardia.

The “before protocol change” group included 363 children and the “after protocol change” group included 502. On average, the new protocol was associated with a per-patient fluid administration rate of 4.1 L per m², compared with a per-patient rate of 5.3 L per m² in the old protocol group. Resolution of acidosis occurred more quickly in the new protocol group than in the old protocol group (12.6 versus 16.7 hours, respectively). Total costs per patient were $1,036 less in the new protocol group than in the old protocol group. There was no significant difference in the amount of sodium delivered to patients in either group and no difference in the frequency or degree of hypokalemia. Complication rates were very low and not significantly different between the groups.

The authors conclude that reducing the rate of fluid given to children with DKA while simultaneously increasing the amount of sodium administered can decrease the number of fluid changes needed, reduce the cost associated with treatment and, perhaps most importantly, reduce the time to resolution of acidosis.