Effects of head elevation on intracranial hemodynamics in patients with ventriculoperitoneal shunts.

ABSTRACT

The present study was performed to investigate the effects of head elevation on intracranial hemodynamics in patients with ventriculoperitoneal (VP) shunts. The series included 35 hydrocephalic patients and five individuals without hydrocephalus who were used as controls. The hydrocephalic patients were divided into three groups 15 patients who received VP shunts with a differential-pressure valve (DP group); 11 who received VP shunts with a variable-resistance valve (VR group), and 13 hydrocephalic patients (Hyd group) who had not received shunts (four underwent VP shunts later). The cerebral blood flow (CBF) of patients in the supine and upright positions was measured by technetium-99m hexamethylpropylenamine oxide (HMPAO) single-photon emission computerized tomography in each patient, using the subtraction technique. Cerebral perfusion pressure (CPP) was taken as the difference between the mean arterial blood pressure and ventricular fluid pressure, both referenced to the level of the foramen of Mono. The patients' heads were elevated stepwise from supine to upright. Percent changes of the mean CBF in the upright position (% delta mCBFupr) were 24.9% +/- 4.3% (mean +/- standard error of the mean) in the DP group, 6.2% +/- 2.7% in the VR group, 3.5% +/- 2.6% in the Hyd group, and 4.5% +/- 2.2% in the control group. Patients in the DP group showed a pathological increase in CPP with head elevation, whereas those in the Hyd and VR groups showed a physiological decrease in CPP. Three patients with differential-pressure valves, whose % delta mCBFupr was markedly high, developed low-intracranial pressure syndrome. In conclusion, shunted patients with a DP valve showed pathological intracranial hemodynamics in the upright position. This pathological hemodynamic stress in patients with long-standing differential-pressure valve implantation may induce pathological changes in the brain such as subependymal gliosis.