Treatment Strategies and Related Outcomes for Brain Arteriovenous Malformations in Children: A Systematic Review and Meta-Analysis.

The objective of this study was to assess the available evidence in the literature regarding treatment outcomes for pediatric patients with brain arteriovenous malformation (bAVM) with the aim of providing practice guidelines for treatment decisions and highlighting research areas that need attention. Keyword searches for studies published from January 1, 1981, to April 16, 2018, were performed in MEDLINE, Embase, and Web of Science. Predefined inclusion criteria were used to identify studies. Poisson regression analysis for associations between patient and bAVM characteristics and treatment outcomes. We identified 34 articles comprising 2158 children with bAVM who underwent treatment or observation. The mean age of the study cohort was 12.0 ± 1.6 (SD) years, and 48.1% of the patients were female; 64.3% of bAVMs were hemorrhagic at presentation. The mean follow-up was 50.6 ± 32.3 months. Overall, the meta-analysis of pooled data showed an obliteration rate of 69.8% (95% CI, 62.9-75.9%), recurrence rate of 2.2% (95% CI, 1.1-4.3%), and mortality rate of 2.4%. The pooled complication rate was 22.5% (95% CI, 15.7-31.1%) after surgery, 26.4% (95% CI, 15.2-41.9%) after embolization, and 27.1% (95% CI, 18.1-38.4%) after radiosurgery. Mortality was not associated with age, sex, or hemorrhage; however, recurrence after treatment was inversely associated with age. Complication and mortality rates were reduced for multimodal treatments. For patients with bAVM treated with observation only, complication and mortality rates were 35.9% and 23.5%, respectively. Multimodality treatments for pediatric bAVM had lower mortality and complication rates than individual treatments. However, there is a lack of evidence for long-term outcomes. The mortality rate was highest in conservatively managed patients (i.e., observation only). Further research directly comparing different treatment modalities for recurrence and complications is warranted. Gathering data prospectively through multiinstitutional registries will be key to provide strong evidence.