LITTLE IS KNOWN about how intravenous fluids influence peritumoral edema formation. This experiment was designed to determine, in a rat glioma model, whether changes in plasma osmolality alter water content, as assessed by specific gravity (SpGr), in normal and neoplastic cerebral tissue. Cells cultured from an ethylnitrosourea-induced rat glioma were stereotactically implanted into the right striatum of Fischer 344 rats. A tumor growth interval of 21 days was allowed. In a second experiment, rats underwent a 60-second cortical freeze injury followed by 24 hours' recovery. In both experiments, rats were assigned to one of three groups: hypotonic (100 ml/kg of 0.2 mol/L NaCl in H2O, intraperitoneally; resultant plasma osmolality ≈ 268 mOsm/kg); isotonic (no treatment; plasma osmolality ≈ 298 mOsm/kg); or hypertonic (10 ml/kg of 1.0 mol/L NaCl in H2O, intraperitoneally; plasma osmolality ≈ 342 mOsm/kg). Thirty minutes after fluid injection, regional SpGr was determined using a kerosene-bromobenzene gradient. In subsets of rats, the tissue morphology and blood-brain barrier permeability of Evans blue dye were assessed. Tissue within the freeze lesion was stained by Evans blue dye with sharp demarcation. Evans blue dye did not stain gliomatous tissue, and central necrosis was not histologically evident. In isotonic rats, glioma SpGr was reduced (1.0411 ± 0.0012 g/ml) relative to the contralateral striatum (1.0437 ± 0.0008 g/ml;P< 0.001). Despite this, a strong linear relation was observed for SpGr and plasma osmolality in both neoplastic and normal tissue. Within the freeze lesion in isotonic rats, SpGr was severely reduced (1.0335 ± 0.0008 g/ml;P< 0.0001) compared with contralateral frontal cortex. SpGr within the freeze lesion did not change in response to changing plasma osmolality. In tissue impermeable to Evans blue dye, SpGr varied directly with plasma osmolality. Results from this model indicate that plasma osmolality remains a critical determinant of peritumoral edema formation when the blood-brain barrier is impermeable to macromolecules. Knowledge of plasma osmotic effects of available intravenous solutions would seem essential in defining fluid therapy for patients with intracranial neoplastic processes.