Staged gamma knife radiosurgery for large cerebral arteriovenous malformations

Kwame Amponsah, Thomas L. Ellis, Michael D. Chan, J. Daniel Bourland, Stephen S. Glazier, Kevin P. McMullen, Edward G. Shaw, Stephen B. Tatter

Research output: Contribution to journalArticle

6 Scopus citations


Background: We present our experience over a 10-year period of staged radiosurgery for large arteriovenous malformations (AVMs) including patient outcomes and methods. Methods: From July 2000 to December 2010, 80 patients with AVMs were treated with gamma knife radiosurgery (GKS) at our institution; of these patients, 5 were treated for large AVMs with staged GKS (volumes >20 cm 3). The mean interval between treatments was 10 months (range 7-16). The mean dose for the margin used was 18.0 Gy (range 16-20). The mean volume treated was 37.2 cm 3 (range 22-50). The mean total follow-up was 76.5 months (range 42-120). Results: Two patients had complete obliteration of the AVM nidus. One patient had 95% obliteration (31 months after radiosurgery), one had 90% obliteration (38 months after radiosurgery), and one had less than 50% obliteration at 53 months with a 16-month interval between staged treatments. Conclusions: Staged radiosurgery is an effective and safe method for the treatment of large AVMs. We report achieving higher marginal doses with staging in planned intervals of 6-9 months between staged treatments. It appears that the longer the wait between treatments, the less likely it is that complete obliteration will be achieved.

Original languageEnglish (US)
Pages (from-to)365-371
Number of pages7
JournalStereotactic and functional neurosurgery
Issue number6
StatePublished - Dec 1 2011


  • Arteriovenous malformation
  • Gamma knife
  • Staged radiosurgery
  • Stereotactic radiosurgery

ASJC Scopus subject areas

  • Surgery
  • Clinical Neurology

Fingerprint Dive into the research topics of 'Staged gamma knife radiosurgery for large cerebral arteriovenous malformations'. Together they form a unique fingerprint.

Cite this