Clinical and pathological analysis of spinal cord astrocytomas in children

E. Rossitch, S. M. Zeidman, P. C. Burger, J. T. Curnes, C. Harsh, M. Anscher, W. J. Oakes

Research output: Contribution to journalArticlepeer-review

57 Scopus citations


Twelve children with pathologically confirmed, well-differentiated spinal cord astrocytomas were studied, and correlations among the degree of resection, pathological characteristics, and time of recurrence were examined. Eight tumors were sampled for biopsy or subtotally resected, and 4 were thought to have been totally removed. Clinical recurrence was seen in 4 of 12 patients, 2 of whom died of their disease. The time to recurrence was 1, 2, 2, and 35 years, respectively. The other 8 children remain free of symptoms, with follow-up ranging from 6 months to 35 years (mean, 8.8 years). In 3 of 8 patients who underwent biopsy or subtotal resection, the tumor recurred, and 2 patients died, whereas there was one recurrence in the patients in whom a 'total' resection had been obtained. The histologically well-differentiated nature of the lesions correlated well with the relatively prolonged clinical course seen in this series during the period of observation. The relatively long clinical courses seen in our limited series should be considered before high-risk therapy for spinal cord astrocytomas in children is implemented. The presence of four pilocytic astrocytomas in this group was of special interest, and it seems likely that these discrete neoplasms can be distinguished from the more infiltrating fibrillary astrocytic neoplasms by magnetic resonance imaging with enhancement with gadopentetate dimeglumine.

Original languageEnglish (US)
Pages (from-to)193-196
Number of pages4
Issue number2
StatePublished - 1990
Externally publishedYes


  • Astrocytoma
  • Spinal cord tumor

ASJC Scopus subject areas

  • Surgery
  • Clinical Neurology


Dive into the research topics of 'Clinical and pathological analysis of spinal cord astrocytomas in children'. Together they form a unique fingerprint.

Cite this