During a 10-year period shunt infections occurred in 27% of the 289 hydrocephalic patients who had cerebrospinal fluid shunts inserted at Children's Hospital Medical Center. The rate of infection did not vary with the type of shunt. Staphylococcus epidermidis and Staphylococcus aureus were responsible for one-half and one-quarter of the infections, respectively. Removal of the infected shunt in conjunction with administration of systemic antibiotics was effective therapy. Use of systemic antibiotics alone was generally ineffective and was associated with an increased mortality rate. Infection itself was a significant risk factor, raising the mortality rate from 17% to 40%. Clustering of infection within two months of surgery and similar rates of infection for ventriculo-atrial and ventriculo-peritoneal shunts indicate that the infecting organisms are usually introduced during the perioperative period. The possibility that prophylactic antibiotics or different shunt materials might reduce the infection rate requires further study.

Caudal herniation of the hindbrain, indistinguishable from the Chiari I deformity, may occur after the establishment of spinal subarachnoid shunts and become symptomatic years after the procedure. Examples are presented and others are cited from the literature. It is proposed that the force responsible for the displacement is the difference in pressure between the cranial and spinal compartments. On the basis of these observations and other considerations as well, a similar process, disproportionate absorption of cerebrospinal fluid from the spinal region, might account for the spontaneous form of the Chiari I deformity.

No abstract

In order to assess the efficacy of high-dose irradiation to the posterior fossa and low-dose irradiation to the spinal axis, we reviewed the results of 60 patients with biopsy-proven medulloblastoma treated at the Joint Center for Radiation Therapy (JCRT) between 1968 and 1984. The 5- and 10-year actuarial survival rates for all patients were 68% and 44%, respectively. The median time to recurrence was 19 months. Extent of surgery, age, and radiation dose to the posterior fossa all were of prognostic value. Complete or subtotal gross resection appeared to be a favorable prognostic indicator compared with biopsy only (P less than 0.05), with a 69% versus 40% actuarial survival rate at 5 years, respectively. Infants 2 years of age or less had a diminished 5-year actuarial survival rate of 48% (P less than 0.05) compared with older age groups. The posterior fossa was the predominant site of recurrence and accounted for 78% of all failures. Local control in the posterior fossa was dose dependent. Seventy-nine percent of the tumors that received 5000 cGy or greater were controlled versus only 33% of the tumors that received less than 5000 cGy (P less than 0.02). There were no supratentorial failures, and there was only one isolated spinal cord failure. There were no solitary spinal failures in 24 patients who received a median dose of only 2400 cGy to the spinal axis. We concluded that low-dose irradiation to the spine and whole brain may be indicated with maintenance of a posterior fossa dose of greater than 5000 cGy.