Blogging live from the European Stroke Conference:
The results of A Randomized trial of Unruptured Brain AVMs (ARUBA) were just presented at the European Stroke Conference in London by Profs. Stapf and Mohr. 
ARUBA was a randomized trial of a multidisciplinary treatment approach to AVMs. It was prospective, planned to enroll 400 patients, and was funded by NINHS. They compared best possible AVM eradication (which is considered the standard of care) vs. conservative therapy (the experimental study arm).
Enrolled subjects had unruptured AVM diagnosed by MRI, and were excluded if there was prior hemorrhage or prior partial AVM treatment or if the AVM was deemed untreatable. Primary endpoint was death or symptomatic stroke; secondary endpoint was a neurological deficit. All enrolled patients were disability-free at enrollment.
The trial was stopped on April 15th by the DSMB when 223 patients were randomized due to overwhelming efficacy of conservative therapy.  Baseline data on 223 were balanced with randomization of 109 to medical and 114 intervention. Eventually a total of 126 ended up getting conservative therapy vs. 97 got intervention. Duration of follow-up was 3 years. The primary outcome of death or stroke was seen in 11 medical group (10%) and 33 in intervention group (29%). The intention to treat for the primary endpoint had a Hazard Ratio of 0.29 with a 95% confidence interval of 0.15 to 0.58. Mortality was comparable
The results of ARUBA are striking because of the assumption that interventional therapy was far superior to doing nothing. The results should be taken with caution because all of the risk of intervention is front-leaded during the period of treatment and benefits of intervention are manifest in the decades after treatment. We do not know how the medical therapy arm will fare in the coming decades and subjects were young (mean age 40s).  The most important result of ARUBA will be how these groups look in 10 years and potentially in 20 years.