In order to identify the risk factors for shunt dependency and to select the optimal treatment modality of hydrocephalus in medulloblastoma patients, fifty two patients with medulloblastoma, who were treated at Seoul National University Hospital
from
1982 to 1992, were reviewed. Retrospectively the authors analyzed the types of treatment and the results in the fifty patients of hydrocephalus associated with medulloblastoma. Two cases of perioperative death were excluded from the statistical
analysis. The initial management of hydrocephalus in 50 patients included : 1) ventriculoperitoneal shunt (VPS) before the removal of tumor in 6, 2) external ventricular drainage (EVD) before the removal of tumor in 4, 3) intraoperative
ventricular
puncture and placement of EVD or internalized reservoir (IVD, internalized ventricular drainage) for continuous or intermittent cerebrospinal fluid (CSF) drainage postoperatively in 29, 4) no procedures of CSF drainage before or duing the
operation
for
tumor removal in 11. Among the 44 patients in whom preoperative VPS was not inserted, 16 finally required permanent VPS after tumor removal and 28 remained shunt-free. Factors which might predict the need for permanent VPS were analyzed and
tested
statistically. The presence of symptoms and signs of increased intracranial pressure, the severity of hydrocephalus on computerized tomography (CT), T and M stage, and the type of initial management of hydrocephalus were not statistically
significant
factors influencing the shunt dependency. The only two statistically significant features predicung the need for a subsequent permanent shunt were : 1) incomplete tumor removal (p=0.025) and 2) postoperative hematoma in the fourth ventricle
(p=0.004).
Among the 11 patients without proeoperative or intraoperative EVD, 4 required shunt placement after tumor removal, of which 3 needed urgent placement of EVD. Preoperative or intraoperative EVD could have played a role as a safety mechanism in
those
patients. On the other hand, the preoperative or intraoperative EVD did not enhance the rate of infection or shunt dependency.
In the cases of ventricular installations for CSF drainage, simple externalization of ventricular catheter and postoperative continuous CSF drainage (EVD) seemed to be associated with a lower rate of infection compared with the method of
postoperative
intermittent puncture and drainage through the internalized reservoir (IVD). but the difference of infection rates between the two group sere statistically insignificant (p=0.557). The average duration of CSF drainage in the externalized EVD
group
was
shorter than that of internalized reservoir group (5.4 versus 10.7 days). In conclusion. 1) the preoperative or intraoperative EVD is useful as a safety mechanism while it dose not enhance the possibilites of shunt dependency and infection, 2)
for
the
patients in whom the placement of subsequent permanent shunt is highly expected, the EVD with the internalized reservoir can be a good choice. An algorithm for the management of hydrocephalus was suggested.
|