Gamma Knife Radiosurgery for Brain Arteriovenous Malformations: A 15-Year Single-Center Experience in Southern Vietnam.

ABSTRACT

BACKGROUND: In the present study, we aimed to identify the obliteration outcomes, complications, and predictors in gamma knife radiosurgery (GKRS) treatment of brain arteriovenous malformations (AVMs) at a tertiary center in a developing country in a 15-year experience. METHODS: We retrospectively reviewed the clinical data and GKRS procedures of patients who had undergone GKRS from 2006 to 2011 (cohort 1) and from 2011 to 2020 (cohort 2) at Cho Ray Hospital, Vietnam. The exclusion criteria included patients with <24 months of follow-up without obliteration or AVM-related hemorrhage during the study period. RESULTS: A total of 870 patients were included in the final analysis. The patients in cohort 1 had had significantly smaller AVMs (8.4 ± 11.6 cm3 vs. 11.2 ± 12.8 cm3; P < 0.001), and the AVMs were less frequently located in eloquent locations (46.6% vs. 65.5%; P < 0.001) than in cohort 2. The mean follow-up time was 49.6 ± 22.6 months (range, 5.9-102.6). The overall AVM obliteration rate was 66.6%. Cohort 1 had a significantly higher rate of complete obliteration compared with cohort 2 (81.0% vs. 55.1%; P < 0.001). The post-GKRS annual hemorrhage risk was 1.0%. Significant radiosurgery-induced brain edema and radiosurgery-induced cyst formation was reported in 24 (2.6%) and 4 (0.5%) patients in cohorts 1 and 2, respectively. Using multivariate analysis, we identified prior AVM hemorrhage (hazard ratio [HR], 1.430; 95% confidence interval [CI], 1.182-1.729), a higher margin dose (HR, 1.136; 95% CI, 1.086-1.188), a noneloquent location (HR, 0.765; 95% CI, 0.647-0.905), and smaller AVM volume (HR, 0.982; 95% CI, 0.968-0.997) as predictive factors for obliteration. CONCLUSIONS: GKRS is a safe and effective treatment of brain AVMs. The lack of prior AVM hemorrhage, an eloquent location, and higher AVM were unfavorable predictors for post-GKRS obliteration.