Patients with an unruptured arteriovenous malformation (AVM) in the brain are four times more likely to have a stroke or die if they have a procedure to eradicate the AVM than if they receive medical management alone, according to the results of a clinical trial led by researchers at Columbia University Medical Center, Icahn School of Medicine at Mount Sinai, and APHP-Hôpital Lariboisière in Paris, France.
The results of the trial, which followed participants for an average of 33 months, were published on Nov. 20 in the British journal The Lancet. The trial was led by J.P. Mohr, MD, Daniel Sciarra Professor of Neurology at CUMC, and Christian Stapf, MD, adjunct professor of neurology at CUMC and a tenured neurologist at APHP-Hôpital Lariboisière.
“Based on these findings, it is difficult to recommend intervention at this time,” Mohr says. “In the first one to three years of follow-up, we haven’t seen the beneficial effect of eradication we had hoped for.”
AVMs are abnormal collections of blood vessels in an area of the brain where high pressure can enlarge the AVM or cause hemorrhaging. Until the rise of MRI and CT, most AVMs were discovered because they had hemorrhaged. Now unruptured AVMs are being detected when brain imaging is done for headache, seizures, or symptoms unrelated to stroke. Researchers estimate that every year AVMs are discovered in one to two people in every 100,000, with two unruptured AVMs discovered for every ruptured AVM.
The idea for a trial to compare eradication with medical management began, Mohr says, when physicians noticed that many patients with unruptured AVMs—who either had refused invasive treatment or had AVMs that were too complex for eradication—were not hemorrhaging at the rate feared. And many patients who had undergone eradication were experiencing adverse events. “It was possible that AVMs were more benign, and treatment more dangerous, than we originally thought,” Mohr says.
The randomized, NIH-funded trial started in 2007. Only patients with unruptured AVMs eligible for eradication were enrolled. Patients were randomized to receive medical management only or to undergo AVM eradication along with medical management.
Thirty-nine sites around the world, including 15 in the U.S., participated; each site was experienced in AVM eradication, performing at least 10 procedures a year. Each treatment team chose the intervention method—surgical removal, embolization via a catheter inserted through an artery, radiation, or any combination thereof.
In April 2013, six years after the trial began, the data and safety monitoring board appointed by the National Institute of Neurological Diseases and Stroke stopped enrollment into the trial, after an interim data analysis showed that medical management was superior to intervention, exceeding the statistical model for stopping the randomization phase.
At that point, 223 patients had been enrolled and followed for an average of 33 months. Of the 114 assigned to the interventional arm of the trial, three died and 34 had a stroke; of the 109 patients assigned to medical management, two died and nine had a stroke.
“One of the study’s major findings overall is that the risk of hemorrhage and death from unruptured AVMs is lower than widely thought,” Mohr says. “When we started the trial, the hemorrhage risk was thought to be 4 percent each year. Our data show that it’s closer to 1 percent.”
Mohr says that it is possible that the number of strokes and deaths in the medically managed group will increase with further follow-up. The investigators also hope that any patients who have been harmed by eradication treatments may show clinical improvement. They have applied to NIH for support to follow the patients for five more years.
“This is not the definitive study for what to do with all AVMs,” Mohr says. “All these interventions seem to make sense and are in regular use for those who have bled; but right now, if you compare the outcomes, intervention presents four times the risk, compared with medical management, for unruptured AVMs.”
Further research may be able to identify a group of AVM patients with a high risk of hemorrhage who would benefit from intervention. Forms of medical therapy may be developed to reduce the risk of hemorrhage.
“We want people who need treatment to get treatment, but we don’t yet know who those patients are,” says Mohr. “Our trial is a major step toward that goal.”