Intervention for AVM Still Too High Risk: Latest From ARUBA

June 26, 2020

Longer-term follow-up from the ARUBA study confirms earlier results showing that intervention for patients with an unruptured brain arteriovenous malformation (AVM) does more harm than good.

Enrollment into the trial, which compared medical management alone to medical management with interventional therapy (neurosurgery, embolization, or stereotactic radiotherapy, alone or in combination), was stopped prematurely in 2013 after 33 months of follow-up because of a much higher rate of death and stroke in the intervention group.

Now the investigators are reporting extended follow-up to 50 months. The results were very similar to those at 33 months.

The current 50-month follow-up results show that 15 of 110 patients in the medical group had died or had had a stroke (3.39 per 100 patient-years), vs 41 of 116 (12.32 per 100 patient-years) in the intervention group. The results reaffirm the strong benefit of not undergoing intervention (hazard ratio, 0.31; 95% confidence interval, 0.17 – 0.56).

These latest results were published in the July issue of Lancet Neurology.

"With an AVM, the natural reflex is to try and fix it, but our trial shows that the tools we have to do that seem to be more damaging than just living with the AVM. If we try to take it out, the stroke risk is three to five times higher than just leaving it alone," coauthor Christian Stapf, MD, University of Montreal, Canada, told Medscape Medical News.

Stapf explained that an AVM is a congenital abnormality in the linking of the arteries to the veins. "There are an excess number of arteries and veins. They usually sit there silently, but they can trigger seizures, as they can tickle the neurons in the vicinity."

It is estimated that one to two AVMs are found spontaneously in every 100,000 persons every year, but this is dependent on the availability of MRI, and many go undetected, he noted. In MRI studies in healthy volunteers, the rate was about one AVM in every 2000 individuals.

With AVMs, rupture and intracerebral hemorrhage occur at a rate of about 1% to 2% per year. Until the ARUBA results were published, the standard practice was to intervene to embolize or excise the malformation, Stapf said.

"The standard treatment was intervention. The experiment was not to do it. We were challenging standard practice, and the trial was not popular with interventionalists," he commented.

The initial study, which was published in 2014, received much criticism from the interventionalist community. Among the criticisms were that the selection criteria for enrollment limited its generalizability, fewer patients than expected in the intervention arm were referred for microvascular surgery, and the follow-up was too short to allow a meaningful comparison.

"Yes, the study received criticism, but this was mainly from interventionalists, who were having their income threatened," Stapf said. "This was very unhappy news for them, especially in the US, with the fee-for-service system."

But he says these longer-term results, together with additional analyses and data from other cohorts, reinforce their initial conclusions.

The current report also shows a benefit in functional outcome in the medical group. "After 5 years, patients are twice as likely to have a neurological handicap, defined as a score of 2 or higher on the modified Rankin scale in the intervention group," he noted. "We also found that more patients in the intervention group had deficits not related to stroke, such as an increase in seizures."

Results of subgroup analysis were consistent in all patient groups.

"Study was designed for 400 patients, but we only recruited about half that number. But even so, the effect of intervention on stroke is so strong there is no subgroup where it looks favorable," Stapf said. "This result was not heterogeneous. The same effect is seen regardless of age, gender, presence of symptoms, size of AVM, location, anatomy, drainage. No matter how you look, there is no benefit for intervention."

He also referred to a Scottish population-based cohort study that showed a similar risk reduction by not intervening. "This was an unselected population of every unruptured AVM patient in Scotland, which found a 65% relative reduction in death/stroke over 12 years. We found a 69% reduction. The Scottish study did not select any particular types of patients but showed the same result as us," he noted. "It is hard to argue against these findings.”

Regarding the claim of selection bias, Stapf acknowledged that the study excluded patients who were judged to be in need of intervention and those judged to be at very low risk and who would not be considered for an intervention.

"But when we compared our cohort to two other unselected cohorts, they look very similar, apart from the fact that very large AVMs were not entered in our study, as they were considered too difficult to treat," he said. "If there is a selection bias at all, it actually trends towards the intervention group, as we excluded those at the highest treatment risk, but we still showed more benefit of not intervening."

He also says the microvascular surgery rates were consistent with real-world practice, with about 25% of patients undergoing such surgery. "This is similar to the Scottish population study. Our trial also showed a similar result in patients treated with the various different interventions ― they all showed a much higher risk than not intervening," he added.

He says practice has changed since the trial was first reported. "There are far fewer interventions now for unruptured AVMs. Most interventionalists have accepted the results now, although there are some who continue to find reasons to criticize the trial and carry on with the procedures."

He says his advice to patients who have an unruptured AVM is to forget about it.

"There doesn't seem to be a trigger for rupture," he said. "It doesn't seem to be dependent on blood pressure or physical activity, and we can't tell if it's just about to go by looking at it. They are very different from an aneurysm in that regard.

"When I explain to patients that they are at an increased stroke risk and tell them about the results of the ARUBA study, they say they would prefer to get that stroke later in life than earlier. These patents can live for 30 or 40 years without a stroke.

"But, yes, there remains a major unmet need. We need to find a way to protect these patients. In future, we might find a better way of intervening, but at this point in time, the treatment we have is more dangerous than doing nothing," he said.

In an editorial that accompanies the current study, Peter M. Rothwell, MD, University of Oxford, United Kingdom, also dismisses much of the criticism of the ARUBA study. On the issue of external validity, he said: "I do not think that this is really any greater an issue for ARUBA than for most other similar trials."

But Rothwell does believe that follow-up for longer than 5 years is needed.

"To really understand the benefit/risk balance, we would need a 20- or 30-year follow-up. These patients are often in their 20s, 30s, or 40s, so we really need to know their cumulative risk over decades," he told Medscape Medical News.

Noting that the study was funded by the National Institute of Neurological Disorders and Stroke (NINDS), Rothwell says funding should have been provided for much longer follow-up. "Patients who generously agreed to be randomly assigned in ARUBA, and future similar patients, have been let down by NINDS.

"We probably now won't ever know the very-long-term impact, although the Scottish population study is following patients longer term," he added.

"After this trial was first published, the guidelines recommended not to intervene. These latest results will not change that," he said.

The ARUBA trial was funded internationally by the National Institutes of Health/NINDS. Stapf and Rothwell have disclosed no relevant financial relationships.

Lancet Neurol. 2020 Jul;19:573-581. Abstract, Editorial

For more Medscape Neurology news, join us on Facebook and Twitter.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.