Kinetics of crystalloid fluid in hyperglycemia; an open-label exploratory clinical trial.

ABSTRACT

BACKGROUND: Infusion with 0.9% saline is a mainstay in the treatment of severe hyperglycemia, but the kinetics of the saline volume in this setting has not been studied. METHODS: An intravenous infusion of 1 L of 0.9% saline over 30 minutes was given on 31 occasions to 17 patients with hyperglycemia due to poorly controlled diabetes (mean age 51 years). A two-volume kinetic model was fitted to serial data on the hemodilution and urinary excretion, using mixed-effects modeling software. RESULTS: Plasma glucose was 36 ± 9 mmol/L on arrival to the hospital. The central volume of distribution (the plasma) was only 2.38 L (mean; 95% confidence interval 1.73-3.04) on the day of admission. Uptake into a remote compartment, believed to be the cells, amounted to 300 mL of the first liter of saline, although only small amounts of insulin were given. Plasma glucose, plasma bicarbonate, urine glucose, and plasma creatinine served as covariates in the kinetic model and mathematically affected the urinary excretion. For example, elimination of the infused fluid tripled from an increase in plasma glucose from 5 to 35 mmol/L and doubled from a reduction in plasma bicarbonate from 24 to 5 mmol/L. CONCLUSIONS: The excretion of 0.9% saline was increased depending on the degree of hyperglycemia. The kinetics was characterized by glucose-accelerated diuresis, and an intracellular uptake that occurred at two thirds the urine flow rate. These data could help to determine appropriate volumes and rates of infusion of crystalloids in hyperglycemia.