A dose-response analysis of arteriovenous malformation obliteration after radiosurgery

Purpose: Although radiosurgery is effective in obliterating the pathologic vessels of intracranial arteriovenous malformations (AVM), the relationships of both dose and volume to obliteration have not been well defined. Methods and Materials: The results of radiosurgery in 197 AVM patients with 3-year angiographic follow-up were analyzed. Volume varied from 0.06-18 cc (median: 4.1 cc), and minimum target dose (D-min) varied from 12.0-25.6 Gy (median: 20.0 Gy). Results: Follow-up angiography revealed complete AVM obliteration in 142 out of 197 patients (72%), The targeted AVM nidus failed to obliterate in 20 patients (10%), but in-field obliteration was complete in the remaining 35 patients (18%) discovered to have residual untargeted AVM nidus, Multivariate logistic regression a analysis of in-field obliteration revealed a significant independent correlation with D-min (p = 0.04), but not with volume or maximum dose. A sigmoid dose-response curve for in-field obliteration was constructed that significantly differed from the dose-volume-response relationships that would have been expected from overall obliteration data. Conclusions: The success rate for in-field obliteration of AVM after radiosurgery depends on D-min but does not appear to change appreciably with volume or maximum dose, Success rates for complete obliteration additionally are limited by problems defining the complete AVM nidus. Copyright (C) 1996 Elsevier Science Inc.