A Comparison of the Cerebral and Hemodynamic Effects of Mannitol and Hypertonic Saline in a Rabbit Model of Acute Cryogenic Brain Injury

There has recently been an increased interest in the use of hypertonic saline solutions in the fluid resuscitation of trauma victims and patients with uncontrollable intracranial hypertension. In this study, the cerebral and hemodynamic effects of 3.2% hypertonic saline solution were compared with those of an equiosmolar (20%) mannitol solution or 0.9% saline in a rabbit model of acute cryogenic brain injury. Forty-five minutes following the creation of a left hemispheric cryogenic brain lesion, equal volumes (10 ml/kg) of hypertonic saline, 0.9% saline, or mannitol were infused over a 5-min period. Monitored variables over the ensuing 120 min included mean arterial pressure, central venous pressure, intracranial pressure (ICP), hematocrit, and serum osmolality. At the conclusion of the 2-h study period, hemispheric water contents were determined by gravimetric analysis and the wet/dry weight method. There were no significant differences in mean arterial pressure between the three groups at any time during the experiment. Plasma osmolality was significantly increased by +/- 10 mOsm/kg following infusions in both the mannitol and hypertonic groups compared to the saline group. The infusion of either mannitol or hypertonic saline produced a transient and significant decrease in ICP during the first 60-90 min but not at 120 min after cryogenic brain lesion, whereas animals in the saline group demonstrated a continual increase in ICP. However, there appeared to be no significant differences in ICP between animals receiving mannitol or hypertonic saline at any time point following infusion of solutions. We conclude that following acute cryogenic brain injury, infusions of equal volumes of equiosmolar solutions of hypertonic saline or mannitol will transiently reduce ICP as compared to equal volumes of normal saline. However, hypertonic saline is not superior to mannitol in its ability to reduce ICP in this model of intracranial hypertension.