Search Results

You are looking at 1 - 2 of 2 items for :

  • Author or Editor: Mitchel S. Berger x
  • Journal of Neurosurgery: Pediatrics x
  • Refine by Access: all x
Clear All Modify Search
Free access

Michael Safaee, Michael C. Oh, Praveen V. Mummaneni, Philip R. Weinstein, Christopher P. Ames, Dean Chou, Mitchel S. Berger, Andrew T. Parsa, and Nalin Gupta

Object

Ependymomas are a common type of CNS tumor in children, although only 13% originate from the spinal cord. Aside from location and extent of resection, the factors that affect outcome are not well understood.

Methods

The authors performed a search of an institutional neuropathology database to identify all patients with spinal cord ependymomas treated over the past 20 years. Data on patient age, sex, clinical presentation, symptom duration, tumor location, extent of resection, use of radiation therapy, surgical complications, presence of tumor recurrence, duration of follow-up, and residual symptoms were collected. Pediatric patients were defined as those 21 years of age or younger at diagnosis. The extent of resection was defined by the findings of the postoperative MR images.

Results

A total of 24 pediatric patients with spinal cord ependymomas were identified with the following pathological subtypes: 14 classic (Grade II), 8 myxopapillary (Grade I), and 2 anaplastic (Grade III) ependymomas. Both anaplastic ependymomas originated in the intracranial compartment and spread to the spinal cord at recurrence. The mean follow-up duration for patients with classic and myxopapillary ependymomas was 63 and 45 months, respectively. Seven patients with classic ependymomas underwent gross-total resection (GTR), while 4 received subtotal resection (STR), 2 received STR as well as radiation therapy, and 1 received radiation therapy alone. All but 1 patient with myxopapillary ependymomas underwent GTR. Three recurrences were identified in the Grade II group at 45, 48, and 228 months. A single recurrence was identified in the Grade I group at 71 months. The mean progression-free survival (PFS) was 58 months in the Grade II group and 45 months in the Grade I group.

Conclusions

Extent of resection is an important prognostic factor in all pediatric spinal cord ependymomas, particularly Grade II ependymomas. These data suggest that achieving GTR is more difficult in the upper spinal cord, making tumor location another important factor. Although classified as Grade I lesions, myxopapillary ependymomas had similar outcomes when compared with classic (Grade II) ependymomas, particularly with respect to PFS. Long-term complications or new neurological deficits were rare. Among patients with long-term follow-up, those who underwent GTR had a recurrence rate of 20% compared with 40% among those with STR or biopsy only, suggesting that extent of resection is perhaps a more important prognostic factor than histological grade in predicting PFS, which has been suggested by other data in the literature. Given the relative paucity of these lesions, collaborative multiinstitutional studies are needed, and such efforts should also focus on molecular and genetic analysis to refine the current classification system.

Restricted access

Edward F. Chang, Catherine Christie, Joseph E. Sullivan, Paul A. Garcia, Tarik Tihan, Nalin Gupta, Mitchel S. Berger, and Nicholas M. Barbaro

Object

Dysembryoplastic neuroepithelial tumors (DNETs) are a subset of relatively rare glioneuronal tumors that typically present with epilepsy during childhood. The authors' aim was to identify factors that predict seizure control following excision.

Methods

The authors reviewed the cases of 50 patients who underwent resection of DNETs at the University of California, San Francisco, between 1990 and 2006. Demographic, seizure history, radiographic, and histopathological data were collected and analyzed for statistical association with postoperative seizure control.

Results

Of the 50 patients, 86% presented with intractable epilepsy. The median age at surgery was 21 years (range 4–46 years; 40% were < 18 years old at time of surgery), with a median duration of 8 years from onset of seizures (24% were adult-onset seizures). Fifty-two percent of the cases were associated with adjacent focal cortical dysplasia. Complete resection was achieved in 78% of cases. Intraoperative electrocorticography in 23 patients identified extralesional interictal activity in 16 cases, which led to extended lesionectomy or lobectomy. The remaining patients underwent lesionectomy alone. The median follow-up was 5.6 years, during which time tumor progression occurred after subtotal resection. The proportional estimates of seizure freedom (Engel Class I outcome) were 0.86 at 1 year and 0.85 at 5 years. Seizure freedom was predicted by complete or extended resection (OR 1.68, 95% CI 1.39–2.03; p < 0.0001) and extratemporal location (OR 1.20, 95% CI 1.02–1.42; p = 0.03) on multivariate analysis. Secondary analysis for intraoperative electrocorticography cases demonstrated that seizure outcome was better when extralesional spiking foci were detected (94% seizure free) compared with when they were absent (43% seizure free).

Conclusions

Excision of DNETs and, when present, adjacent dysplastic cortex was highly effective for seizure control. Excellent seizure-free outcomes and tumor control were seen with lesionectomy alone in most cases. Electrocorticography with extended resection was useful for patients with pharmacoresistant epilepsy.