LOS ANGELES – Hang a do-not-disturb sign on brain arteriovenous malformations.
Patients who underwent invasive interventions to repair an unruptured arteriovenous malformation (AVM) in their brain faced a greater than two-fold increased rate of death or stroke during an average 4 years of follow-up, compared with patients who received medical treatment only with no active intervention, Dr. Christian Stapf reported at the International Stroke Conference.
When analyzed on an intention-to-treat basis, for every five AVM patients treated by endovascular surgery, conventional surgery, or radiotherapy, one additional patient died or had a stroke, compared with the death or stroke rate among control patients who received only medical management. When analyzed based on the treatments that patients actually received, the number-needed-to-harm fell to one excess death or stroke for every three AVM patients who underwent an invasive procedure, compared with control patients, reported Dr. Stapf, a professor in the department of neurosciences at the University of Montreal.
The results from A Randomized Trial of Unruptured Brain AVMs (ARUBA) “show us that we clearly have not been as good as we thought in helping patients against their stroke risk,” said Dr. Stapf in a video interview during the meeting. “Given that the risk of death or stroke was reduced three- to fivefold with no [invasive] treatment and leaving the AVM alone makes us think that we can’t recommend preventive intervention with currently-used techniques. Living with the AVM seems like the far better option.”
The ARUBA study, run at 39 centers in nine countries including 13 U.S. centers, randomized 226 patients with unruptured AVMs before the study’s data safety and monitoring board stopped study enrollment prematurely in April 2013. The study group included 110 patients randomized to receive medical interventions only and 116 randomized to medical intervention plus “best possible” AVM eradication. The exact type of eradication for each patient was left up to local clinicians, who tailored the intervention to address the size, location, and anatomy of each AVM. Medical management included steps such as treatment with antiepileptic drugs to treat seizures, various treatments for headaches, and physiotherapy for patients with neurologic deficits.
The study’s primary endpoint was the combined rate of death or stroke, which occurred in 41 of the 116 patients (35%) randomized to receive an invasive intervention and in 15 of the 110 (14%) randomized to medical treatment only during an average follow-up of 50 months, with many patients followed for 5 years.
When analyzed by the treatment patients actually received, 106 underwent an invasive intervention and 43 of these patients (41%) died or had a stroke, and 120 patients received medical treatments only, of whom 13 (11%) died or had a stroke.
A secondary endpoint was the rate of death or disability after 5-year follow-up, with disability defined as a modified Rankin Scale score of 2 or more. This occurred in 38% of the 45 patients who underwent AVM eradication and had this follow-up available, and in 18% of 51 patients who had medial treatment only and received this follow-up.
Interim results from the study came out 2 years ago, with an average follow-up of 33 months (Lancet. 2014 Feb 15;383[9917]:614-21), but the trial was designed to have 5-year follow-up, largely accomplished in the new data reported by Dr. Stapf.
Many clinicians had already abandoned invasive interventions to treat brain AVMs following release of the interim results, and Dr. Stapf predicted that this trend will further strengthen now that the final results are in and confirm the earlier indication of hazard. Until the ARUBA results became available, clinicians had presumed invasive interventions to resolve or minimize malformations were beneficial based on intuition. ARUBA is the first systematic comparison of procedures versus a hands-off approach for brain AVMs, he said.
“Neurologists will now be less likely to refer patients for intervention, and interventionalists will be less enthusiastic to perform procedures,” Dr. Stapf said during the meeting, sponsored by the American Heart Association. In addition, anyone now performing an intervention in routine practice would need to consider the possible legal implications if the patient were to have a bad outcome. Dr. Stapf also noted that some professional societies are now considering recommendations against routine interventions. He conceded that some invasive interventions might still occur for selected cases on an investigational basis, but the ARUBA results “set the bar very high against performing any new interventions,” he concluded.
Approximately 3,000 patients annually are newly diagnosed with an unruptured brain AVM in the United States and Canada, he estimated.
ARUBA received no commercial support. Dr. Stapf had no disclosures.