Cerebrospinal fluid drainage as influenced by ventricular pressure in the rabbit.

Artificial cerebrospinal fluid (CSF) containing radioisotope iodinated (125I) serum albumin (RISA) and either blue dextran or indigo carmine was given to white New Zealand rabbits over 4 hours. In one group it was given by ventriculocisternal perfusion, in one by ventricular infusion, and in one by cisterna magna infusion. Blood was sampled continuously from the superior sagittal sinus (SSS) or intermittently from the systemic arterial circulation. Removal of CSF from the cisterna magna during the ventriculocisternal perfusion kept the intracranial pressure (ICP) at 0 to 5 torr, whereas ventricular or cisterna magna infusion raised the ICP to 20 to 30 torr and 15 to 20 torr, respectively. In the two groups with raised ICP, an increased concentration of RISA was present in the optic nerves, olfactory bulbs, episcleral tissue, and deep cervical lymph nodes; but this was not found in the group with normal ICP. In all three groups, the concentration of RISA in the SSS blood was the same as in the systemic arterial blood. The concentration gradient of RISA across the cerebral cortex was similar in both the ventriculocisternal perfusion and the ventricular infusion groups. With cisterna magna infusion, the concentration of RISA was the same on the cortical surface and less in the ventricles compared with the ventricular infusion. It is concluded that, with elevated ICP, CSF drained via pathways that are less evident under normal pressure. Drainage of CSF was similar irrespective of whether the infusion site was the ventricles or cisterna magna. It did not appear that acute dilatation of the ventricles during ventricular infusion compromised the subarachnoid space over the surface of the hemisphere, as the concentration of RISA on the convexities and in the SSS blood did not significantly differ between the groups. Transcortical bulk transfer of CSF was not evident with raised ICP.