Worse Outcomes After Repeat vs Initial Stereotactic Radiosurgery for Cerebral Arteriovenous Malformations: A Retrospective Matched-Cohort Study

BACKGROUND: Incompletely obliterated cerebral arteriovenous malformations (AVMs) after initial treatment with stereotactic radiosurgery (SRS) can be treated with a repeat session of SRS. However, the relative efficacy of repeat vs initial SRS is not well specified. OBJECTIVE: To retrospectively compare in matched cohorts the outcomes of repeat vs initial SRS for the treatment of matched cohorts with angioarchitecturally similar AVMs. METHODS: We studied a data set of patients with AVM treated with radiosurgery during the period spanning 1989 to 2013. Patients with AVM who underwent repeat SRS with radiologic follow-up of >= 2 years or nidus obliteration were identified for the study and matched, in a 1: 1 fashion that was blinded to outcome, to patients with previously untreated AVMs who underwent initial SRS. Statistical analyses were performed to compare the outcomes after repeat vs initial SRS. RESULTS: The matching approach resulted in 84 patients for the repeat and the initial SRS cohort (mean margin doses, 20.7 and 20.9 Gy, respectively; P =.74). In the repeat SRS cohort, obliteration was achieved in 67%; the radiologic, symptomatic, and permanent radiation-induced change rates were 35%, 10%, and 4%, respectively; and the post-SRS hemorrhage rate was 3.1%/y. Compared with the initial SRS cohort, the repeat SRS cohort had significantly lower obliteration rates (P = .04) and higher post-SRS hemorrhage rates (P = .04). The radiation-induced change rates of the 2 cohorts were not significantly different. CONCLUSION: Repeat SRS yields considerably poorer outcomes than initial SRS for angioarchitecturally comparable AVMs. Further studies in AVM radiobiology and vascular structure are necessary to elucidate this potentially differential response.