Effects of ventriculoperitoneal shunt removal on cerebral oxygenation and brain compliance in chronic obstructive hydrocephalus.

OBJECT The pathophysiology of shunt malfunction has not been fully examined, probably because of the paucity of appropriate animal models. Using a canine model of chronic obstructive hydrocephalus, the effects of shunt placement and removal on physiological parameters were evaluated. METHODS Fifteen dogs, nine in which chronic hydrocephalus was induced and six controls, were used in the experiment. Thirteen weeks after the induction of hydrocephalus, intracranial pressure (ICP), tissue and cerebrospinal fluid O2 saturation, response to hyperventilation, and brain compliance at low (5-15 mm Hg) and high (15-25 mm Hg) pressures were measured (untreated stage). Following this procedure, ventriculoperitoneal shunts were implanted in the dogs suffering from hydrocephalus. Two weeks later, the same series of measurements were repeated (shunted stage), following which the shunt systems were removed. One week after shunt removal, the last measurements were obtained (shunt-removed stage). All dogs underwent magnetic resonance imaging four times: before induction of hydrocephalus and before each measurement. All dogs with hydrocephalus also had ventriculomegaly (1.42 +/- 0.89 ml before induction of hydrocephalus compared with 3.4 +/- 1.64 ml 13 weeks after induction, p = 0.0064). In dogs in the untreated hydrocephalus stage, ICP remained within the normal range (8.33 +/- 2.60 mm Hg)--although it was significantly higher than that in the control group (5 +/- 1.41 mm Hg, p = 0.014). Tissue O2 saturation in the dogs in the hydrocephalus group (26.1 +/- 5.33 mm Hg) was lower than that in the dogs in the control group (48.7 +/- 4.27 mm Hg, p < 0.0001). After the dogs underwent shunt placement, significant improvement was observed in their ICP (5.22 +/- 2.17 mm Hg, p = 0.012) and tissue O2 saturation (35.2 +/- 6.80 mm Hg, p = 0.0084). However, removal of the shunt reversed these improvements back to the preshunt status. Hyperventilation induced significant decreases in ICP and O2 saturation at every measurement time and induced a significant decrease in tissue O2 saturation during the shunted stage, but not during the untreated and shunt-removed stages. Brain compliance measured at high pressure demonstrated a significant gradual decrease at every measurement. CONCLUSIONS In chronic obstructive hydrocephalus, shunt placement improves ICP and cerebral oxygenation as well as the response to hyperventilation in the tissue. Shunt removal reverses these improvements back to levels present during the untreated stage. The decrease in brain compliance may be one of the factors responsible for symptoms in shunt malfunction.