Letter to the Editor: Improving arteriovenous malformation research and care

Free access

If the inline PDF is not rendering correctly, you can download the PDF file here.

TO THE EDITOR: In the November issue of the Journal of Neurosurgery, Elhammady and Heros2 focused on their current management of unruptured cerebral arteriovenous malformations (AVMs) in the era after ARUBA (A Randomized Trial of Unruptured Brain Arteriovenous Malformations) had ended (Elhammady MS, Heros RC: Editorial. Management of incidental cerebral AVMs in the post-ARUBA era. J Neurosurg 121:1011–1014, November 2014). While recognizing their substantial expertise and major contributions to the field of vascular neurosurgery, we would like to go one step further and use this period of uncertainty created by ARUBA as an opportunity to improve evidence in the management of unruptured AVMs rather than continue on opinion and experience.

We are writing today with an urgent appeal to enter the post-ARUBA era. Now, while we are going through a questioning period, is the opportune time to dust ourselves off and embark on a new program combining research and practice to offer optimal care to our patients with AVM.6 The ARUBA study,4 which concluded that the medical management of unruptured AVMs is superior to any interventional treatment, has elicited from our community a multitude of editorials, critiques, baffled responses, and frustration.1–3,5,7 Some of ARUBA's shortcomings are the large number of nonrandomized but eligible patients, the definition of poor outcomes, the nonuniformity of treatment, and the short follow-up period. Given these weaknesses, many of us are reluctant to accept ARUBA's conclusions. The risk now is that we may come away from it not having learned any lesson and pursue some course of action without fully grasping the implications. Some may disregard the conclusions of the ARUBA study altogether and continue treating AVMs as they see fit. In contrast, others may embrace ARUBA's conclusions and stop treating unruptured AVMs entirely, period. As professionals concerned about the well-being of our patients, none of us can blindly ignore the fact that some patients will suffer in either case. This would be irresponsible. As controversial as this study may be, the next step seems clear: a new randomized study to justify our action or inaction before we establish a new course of action. Such a new study would allow the inclusion of all AVM patients, followed over a longer period of time, and would allow minor or predictable postoperative deficits to be considered as good outcomes given the possible devastating consequences of AVM rupture. This is the best way to minimize error and to have a chance to answer the crucial questions with which all AVM patients are confronted. We have been treating AVMs for decades, and we still do not know whether we should even treat them at all. The burden of proof rests on us. If we offer a curative treatment, it is our duty to prove that any risky preventive intervention will be beneficial. If we choose conservative management, there again we must base our decision on evidence.

For these reasons, we propose the TOBAS (Treatment of Brain AVMs) study. It is a pragmatic, prospective randomized study with an accompanying registry. Contrary to ARUBA, which started off with the hypothesis that conservative management was best, the hypothesis of TOBAS favors curative treatment. There are 3 main objectives: 1) to allow clinicians to manage their patients within a research protocol, which is probably the best way to justify our action when faced with uncertainty and is probably also the best way to minimize potential error in our decision making; 2) to determine the role of curative treatment in unruptured AVMs; and 3) to stratify the risks associated with the various treatment modalities in unruptured and ruptured AVMs, including whether or not to pre-embolize, since the potential benefit of presurgical or preradiosurgical embolization remains contentious.

These goals are essential to increase our knowledge and improve care. As stated by Arthur L. Day and colleagues, “ARUBA is not the end of research in brain AVMs; it is just the end of the beginning of new research into their management.”1

We know that we can injure patients while treating an AVM, and ARUBA reaffirmed this. However, this does not mean that we should stop treating unruptured AVMs, as Elhammady and Heros have noted. It does mean that we cannot continue doing what we were doing without taking a closer look at what we are doing.

We invite physicians treating patients with AVMs to join TOBAS (ClinicalTrials.gov, trial registration no.: NCT02098252; tdarsaut@ualberta.ca or jraymond.nri@gmail.com).



We appreciate Dr. Bojanowski and colleagues' expressed interest in our recent editorial. We certainly agree with their concerns that some may wrongly interpret the conclusions of the ARUBA study as indicating that no patient with an unruptured AVM should be treated. We also agree that it would be wrong to completely ignore the findings of the ARUBA study. At the very least we have learned from this study that any kind of intervention for unruptured AVMs can carry significant morbidity.

We agree with Bojanowski and colleagues that there is still some uncertainty about how and when to treat some unruptured cerebral AVMs and that continuing research is in order. However, we disagree with their statement that “we have been treating AVMs for decades, and we still do not know whether we should even treat them at all.” We do believe that through accumulated experience, we have learned a great deal about treating AVMs and which AVMs should be treated. The lack of confirmation by a randomized study does not mean that what we have learned from personal experience and from a careful following of the literature is worth nothing. When faced with having to advise a patient, it would be unethical for us to recommend a course of action other than that which we think is best for that patient. As an extension of this, it would also be improper, in our opinion, to tell a patient that we do not know which course of action is better and therefore that we are willing to randomize him or her if indeed we believe that one course of action is better than another. This will make it hard for experienced clinicians to randomize a large number of patients into another study such as the TOBAS study proposed by the authors. It would be difficult for us to randomize patients with Spetzler-Martin (SM) Grades I and II AVMs since we firmly believe that the results of excision in this group of patients, in competent hands, are far superior to the lesion's natural history. Likewise, it would be very difficult to randomize unruptured SM Grade IV or V malformations since we have learned that the morbidity of curative treatment of these lesions is too high to justify treatment given what we know about their natural history. We also believe that palliative treatment, such as partial embolization, for these lesions is ineffective. There may be an occasional Grade IV malformation, perhaps a very large lesion in the frontal or nondominant parietal lobe that does not directly involve eloquent regions of the brain, that we would randomize because we truly do not know if treatment, such as excision after embolization, is better than the natural history in these particular cases. I suspect we would randomize a number of Grade III malformations since we truly don't know if treatment is better than observation for many of these lesions. However, randomization would have to be specifically for the form of treatment that we believe would be best for that malformation. We would not, for example, randomize to radiosurgery a patient with a large Grade III malformation, nor would we want to randomize to excision a patient with a small deep malformation in an eloquent area of the brain that we believe would be best treated by radiosurgery.

Dr. Bojanowski and colleagues have been great contributors to the cerebrovascular field, and we are sure that if they lead the design of a new randomized study of AVM treatment, they would do so in a very thoughtful and most ethical way. We are just not very optimistic that a large enough number of patients can be recruited into such a study in a reasonable period of time.

If the inline PDF is not rendering correctly, you can download the PDF file here.

Article Information

INCLUDE WHEN CITING Published online March 6, 2015; DOI: 10.3171/2014.10.JNS142356.

DISCLOSURE The authors report no conflict of interest.

© AANS, except where prohibited by US copyright law.



  • 1

    Day ALDannenbaum MJung S: A Randomized trial of unruptured brain arteriovenous malformations trial: an editorial review. Stroke 45:314731482014

  • 2

    Elhammady MSHeros RC: Editorial. Management of incidental cerebral AVMs in the post-ARUBA era. J Neurosurg 121:101110142014

  • 3

    Meling TRProust FGruber ANiemela MRegli LRoche PH: On apples, oranges, and ARUBA. Acta Neurochir (Wien) 156:177517792014

  • 4

    Mohr JPParides MKStapf CMoquete EMoy CSOverbey JR: Medical management with or without interventional therapy for unruptured brain arteriovenous malformations (ARUBA): a multicentre, non-blinded, randomised trial. Lancet 383:6146212014

  • 5

    Pierot LFiehler JCognard CSöderman MSpelle L: Will a randomized trial of unruptured brain arteriovenous malformations change our clinical practice?. AJNR Am J Neuroradiol 35:4164172014

  • 6

    Raymond JDarsaut TEAltman DG: Pragmatic trials can be designed as optimal medical care: principles and methods of care trials. J Clin Epidemiol 67:115011562014

  • 7

    Russin JSpetzler R: Commentary: The ARUBA Trial. Neurosurgery 75:E96E972014. (Letter)




All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 167 167 4
PDF Downloads 134 134 2
EPUB Downloads 0 0 0


Google Scholar