Measured degree of dehydration in children and adolescents with type 1 diabetic ketoacidosis.

Objective: Successful management of diabetic ketoacidosis depends on adequate rehydration while avoiding cerebral edema. Our objectives are to 1) measure the degree of dehydration in children with type 1 diabetes mellitus and diabetic ketoacidosis based on change in body weight; and 2) investigate the relationships between measured degree of dehydration and clinically assessed degree of dehydration, severity of diabetic ketoacidosis, and routine serum laboratory values.

Design: Prospective observational study.

Setting: University-affiliated tertiary care children's hospital.

Patients: Sixty-six patients <18 yrs of age with type 1 diabetic ketoacidosis.

Interventions: Patients were weighed using a portable scale at admission; 8, 16, and 24 hrs; and daily until discharge. Measured degree of dehydration was based on the difference between admission and plateau weights. Clinical degree of dehydration was assessed by physical examination and severity of diabetic ketoacidosis was assessed by blood gas values as defined by international guidelines. Laboratory values obtained on admission included serum glucose, urea nitrogen, sodium, and osmolality.

Measurements And Main Results: Median measured degree of dehydration was 5.2% (interquartile range, 3.1% to 7.8%). Fourteen (21%) patients were clinically assessed as mild dehydration, 49 (74%) as moderate, and three (5%) as severe. Patients clinically assessed as moderately dehydrated had a greater measured degree of dehydration (5.8%; interquartile range, 3.6% to 9.6%) than those assessed as mildly dehydrated (3.7%; interquartile range, 2.3% to 6.4%) or severely dehydrated (2.5%; interquartile range, 2.3% to 2.6%). Nine (14%) patients were assessed as mild diabetic ketoacidosis, 18 (27%) as moderate, and 39 (59%) as severe. Diabetic ketoacidosis severity groups did not differ in measured degree of dehydration. Variables independently associated with measured degree of dehydration included serum urea nitrogen and sodium concentration on admission.

Conclusion: Hydration status in children with diabetic ketoacidosis cannot be accurately assessed by physical examination or blood gas values. Fluid therapy based on maintenance plus 6% deficit replacement is reasonable for most patients.