Treatment of Unruptured Brain Arteriovenous Malformations: A Single-Center Experience of 86 Patients and a Critique of the A Randomized Trial of Unruptured Brain Arteriovenous Malformations (ARUBA) Trial.

OBJECTIVE: The A Randomized Trial of Unruptured Brain Arteriovenous Malformations (ARUBA) trial has received fierce criticism, including considerable selection bias, poor generalizability, questionable clinical practices (only 15.8% underwent surgical resection, the gold standard for arteriovenous malformation [AVM] treatment), and short follow-up (33 months) for a disease process that carries a life-long risk. In this study, we sought to present our own experience treating unruptured brain AVMs to provide supporting evidence of the ARUBA trial criticism.

METHODS: All cases of treated brain AVMs from 2004 to 2017 at our institution were retrospectively reviewed and included in the analysis if they met ARUBA trial inclusion criteria. The primary outcome was symptomatic stroke or death. Secondary outcomes included AVM obliteration, long-term clinical impairment (modified Rankin Scale score >1), and new major or minor postoperative deficit.

RESULTS: Of the 245 reviewed cases, 86 met the ARUBA trial criteria. Treatment included microsurgical resection alone (2.3%), preoperative embolization followed by microsurgical resection (62.8%), stereotactic radiosurgery alone (10.5%), embolization followed by stereotactic radiosurgery (15.1%), and embolization alone (9.3%). The primary outcome was met in 8.3%, new perioperative major and minor complications occurred in 5.8% and 12.8%, and long-term clinical impairment in 4.5%. AVM obliteration was observed in 92.4% overall and in 100% of patients who underwent surgical resection.

CONCLUSIONS: The criticism of the ARUBA trial is warranted, as our study found that treatment of unruptured brain AVMs has an acceptable safety profile when approached in a multidisciplinary manner at an experienced institution, using surgical resection as the primary treatment modality when applicable.