A meta-analysis comparing outcomes of microsurgery and gamma knife radiosurgery.

Journal Article (Journal Article)

OBJECTIVES/HYPOTHESIS: Surgery has been the most common treatment for acoustic neuromas, but gamma knife radiosurgery has emerged as a safe and efficacious alternative to microsurgery. This meta-analysis compares the outcomes of the two modalities. STUDY DESIGN: A retrospective MEDLINE search was used to find all surgical and gamma knife studies published from 1990 to 1998 and strict inclusion criteria were applied. RESULTS: For tumors less than 4 cm in diameter, there is no difference in hearing preservation (P = .82) or facial nerve outcome (P = .2). Surgery on all sized tumors has a significantly lower complication rate than radiosurgery performed on tumors smaller than 4 cm (P = 3.2 x 10(-14)). Surgery also has a lower major morbidity rate than gamma knife radiosurgery (P = 2.4 x 10(-14)). Tumor control was defined as no tumor recurrence or no tumor regrowth. Surgery has superior tumor control when tumors are totally resected (P = 9.02 x 10(-11)). Assuming that all partially resected tumors will recur, surgery still retains a significant advantage over radiosurgery for tumor control (P = .028). CONCLUSION: Data from these studies date back to the late 1960s and do not completely reflect outcomes using current imaging and procedures. A major difficulty encountered in this study is inconsistent data reporting. Future surgical and radiation reports should use standardized outcomes scales to allow valid statistical comparisons. In addition, long-term results from gamma knife radiosurgery using lower dosimetry have not been reported. Surgery should remain the therapy of choice for acoustic neuromas until tumor control rates can be established.

Full Text

Duke Authors

Cited Authors

  • Kaylie, DM; Horgan, MJ; Delashaw, JB; McMenomey, SO

Published Date

  • November 2000

Published In

Volume / Issue

  • 110 / 11

Start / End Page

  • 1850 - 1856

PubMed ID

  • 11081598

International Standard Serial Number (ISSN)

  • 0023-852X

Digital Object Identifier (DOI)

  • 10.1097/00005537-200011000-00016


  • eng

Conference Location

  • United States