Subscribe to eTOC


Last month, when South Dakota Sen. Tim Johnson had surgery for a bleeding brain arteriovenous malformation (AVM), the American public learned more about this rare but life-threatening condition. None of the science was of course new to neurologists, but the incident highlighted how little is still known about the natural history of AVMs and what to do with those that have not yet bled.

Several new studies, however, aim to address these lingering questions. In partnership with the medical school of Columbia University in New York City, the NINDS has established a long-term Arteriovenous Study Group. One study to come out of this collaboration, A Randomized multi-center clinical trial of Unruptured Brain AVMs (ARUBA), will involve 100 centers and up to 800 patients to compare outcomes of untreated and treated unruptured AVMs.


Sen. Tom Carper (D-DE) updated reporters on the medical condition of Sen. Tim Johnson (inset) last month after he had surgery for an AVM that hemorrhaged. Coverage of Sen. Johnsons condition elicited discussion in the lay press about treatment of AVMs.

“The natural history of brain AVMs discovered prior to hemorrhage is not well enough understood to certify the need to rid patients of the AVM before they have bleeding,” JP Mohr, MD, principal investigator for the study site at Columbia, said. “The reason for that is that AVMs are in, not on brain tissue. And in the process of eradicating them by any method, there is the possibility of unintentional disturbance of healthy brain tissue adjacent to the malformation, which may create symptoms of stroke that no one wishes.”

Study participants will be randomly assigned to receive either symptomatic medical management alone or invasive therapies including endovascular embolization, stereotactic radiotherapy, and surgical excision (or any combination of these).

Dr. Mohr, director of the Doris and Stanley Tananbaum Stroke Center at Columbia, noted that cerebral AVMs are rare and many are discovered incidentally when patients have imaging studies for something else. In fact, according to the NIH, AVMs affect about 300,000 Americans and only about 12 percent of these people have symptoms related to the AVM. Each year about 1 percent die as a direct result of an AVM.

He said patients such as Sen. Johnson — who was probably unaware of his AVM until it caused intracerebral bleeding — are exactly the type the study hopes to enroll. “It's not clear from what we know at the moment whether it's preferable to intervene or to delay intervention until the occurrence of hemorrhage. That's where we're stuck.”


Current treatment for AVMs is decided on a case-by-case basis, said Robert D. Brown Jr., MD, professor and chair of neurology at the Mayo Clinic College of Medicine in Rochester, MN. “Essentially, four treatment options exist when you see someone with an AVM,” he said. One includes leaving the AVM alone, which sometimes is considered for patients who have not yet had a hemorrhage. In this case, “the natural history is compared to the risk of the treatment options,” he said.


Dr. Edgar Kenton III: “We need to know a little bit more about AVMs because the numbers are so small in terms of defining best treatment.”

The ideal treatment would be to remove the AVM in its entirety, but this must be weighed against the risk of complications. Microsurgery may be considered for AVMs of all sizes and locations, but is most safely performed for AVMs that are small (usually less than 3 cm in diameter) and superficial, meaning they are near the surface of the brain. The problem is that many AVMs are located deep inside the brain, near important brain structures that are at risk of being damaged or destroyed during surgery, Dr. Brown said. In those cases, other treatments are considered.

One common treatment is endovascular embolization, in which a surgeon guides a micro-catheter inside the blood vessels that supply the AVM. Once the catheter reaches the AVM, chemicals are injected to form an embolus, or blood clot. These materials can include liquid tissue adhesives, micro-coils, and particles used to stop blood flowing to the AVM.

Another option is radiosurgery, in which a focused beam of radiation is used to degenerate and eventually close the vessels within the vascular malformation.

Since AVMs are complex vascular lesions, treatment decisions should be made with input from a multidisciplinary team of neurologists, neuroradiologists, neurosurgeons, and radiotherapists, Dr. Brown said. According to a 2006 study, success rates for treatments are difficult to compare because strategies vary by center, country, and continent. The evidence for invasive AVM treatment is based on uncontrolled, highly selective cohorts, based at single centers, and subject to referral and treatment selection bias (Curr Opin Neurol 19:63–68). But one study found that endovascular embolization can result in clinical neurological deficits of 14 percent (Stroke 2002;33:1816–1820). And the major disadvantage of stereotactic radiotherapy, according to the 2006 study, is a persistent hemorrhage risk of up to 10 percent until and even after the lesion disappears.

About 50 percent of AVMs first become symptomatic with a hemorrhage. The remaining 50 percent can become manifest by seizures, headaches, or neurologic deficit (Lancet Neurol 2005;4(5):299–308). Dr. Brown said neurologic deficits include weakness, numbness, difficulty speaking, and impaired vision — resulting from effects on surrounding brain tissue. A smaller percentage of patients, he said, note pulsating tinnitus caused by blood rushing through the malformation.


Dr. J. P. Mohr: “The natural history of brain AVMs discovered prior to hemorrhage is not well enough understood to certify the need to rid patients of the AVM before they have bleeding.”


“We need to know a little bit more about AVMs because the numbers are so small in terms of defining best treatment,” said Edgar Kenton III, MD, a spokesperson for the AAN and former director of the Stroke Prevention/Intervention Program at the Morehouse School of Medicine in Atlanta, GA. “We are unclear in terms of the cause, epidemiology, and natural history [of AVMs]. We need more research.”

An important consideration is that AVMs tend to be more benign in children than in adults, he said. For example, one study notes that although children with brain AVMs are more likely than adults to present with hemorrhage, they are not at increased risk for a subsequent intracranial hemorrhage (Stroke 2005;36:2099–2104). If scientists better understood what it is in children that may be protective, this could guide future treatments, he said.

Hemorrhagic AVM presentation, increasing age, deep brain location, and exclusive deep venous drainage seem to be independent predictors for intracranial hemorrhage in untreated AVM patients, according to a 2006 study in Neurology (66:1350–1355). Sex, AVM size, and presence of associated aneurysms, on the other hand, showed no significant effect.

Researchers led by Stephan Sidney, MD, MPH, associate director for clinical research at Kaiser Permanente Northern California, hope to build on these findings. The NINDS-sponsored researchers are currently studying large populations of patients with AVMs to determine trends in incidence of AVMs and the rate of re-bleeding.

Dr. Sidney is also leading a study to test the association of genetic risk factors with intracranial hemorrhage (ICH) occurring after the diagnosis of a cerebral AVM in Kaiser Permanente patients.

Genetic factors may one day be used to decide whether to perform surgery on an AVM. In a study published in 2006 (Neurosurgery 58(5):838–843), Dr. Sidney and colleagues reported that apolipoprotein E genotype may increase the risk of intracranial hemorrhage in the natural course of brain AVM. “The identification of genetic predictors of ICH risk may facilitate estimation of AVM natural history risk and individualize clinical decision-making and therapeutic recommendations,” they wrote.


Typical site for an AVM.

Another study published last year suggested that two common antibiotics — minocycline and doxycycline — might be useful to reduce the risk of bleeding from AVMs (J Cereb Blood Flow Metab 26(9):1157–1164). In mice, these drugs lowered enzyme activities of matrix metalloproteases (MMPs). MMPs, which are normally involved in tissue remodeling, degrade tissue and thereby may increase the risk of the vessel wall rupturing and causing spontaneous bleeding. Previous trials had shown that oral tetracycline-class drugs decreased MMPs in abdominal aortic aneurysms and carotid plaques. In this study, both drugs showed dose-dependent inhibition of matrix metalloproteinase mRNA expression in the mouse brain.

Dr. Brown emphasized that the incidence of detection of AVMs will probably increase as more are discovered through CT and MRI. He worked on a population-based study in Olmsted County, MN, which showed that detection rates increased from 1965 to 1992 (Neurology 1996;46(4):949–52), most likely because of the introduction of imaging. Forty-eight intracranial vascular malformations were detected among Olmsted County residents during the 27-year period, including 26 AVMs. From 1965 to 1974, the detection rate of AVMs was 1.24 in 100,000 patients per year. From 1985 to 1992, however, the rate jumped to 3.21 in 100,000 patients per year.

And as more AVMs are found, the major issue will be what to do with them. “For patients identified without a hemorrhage, in terms of clarifying who should be treated and who should be left alone, we still have a good deal of work to do,” he said.


  • ✓ Following the hemorrhage of Sen. Tim Johnson's AVM, cerebrovascular experts discuss the current research and treatment options for AVMs.


Are neurologists subject to liability risk if their patients with epilepsy — or other conditions involving lapses in consciousness — cause a car accident? Legal experts help clarify the intricacies of state laws and requirements on page 18 in Neurology Today's new monthly column, Legal-ease, so named to answer common medical-legal questions that affect neurologists.

Have a neurology-related medical-legal question that you'd like our experts to answer? Submit your questions for our consideration to [email protected].


• Stapf C, Mohr JP, Choi JH, et al. Invasive treatment of unruptured brain arteriovenous malformations is experimental therapy. Curr Opin Neurol 2006;19:63–68.
    • Hartmann A, Pile-Spellman J, Stapf C, et al. Risk of endovascular treatment of brain arteriovenous malformations. Stroke 2002;33:1816–1820.
      • Choi JH, Mohr JP. Brain arteriovenous malformations in adults. Lancet Neurol 2005;4:299–308.
        • Fullerton HJ, Achrol A, Johnston SC, et al. Long-term hemorrhage risk in children versus adults with brain arteriovenous malformations. Stroke 2005;36:2099–2104.
          • Staph C, Mast H, Mohr JP, et al. Predictors of hemorrhage in patients with untreated brain arteriovenous malformation. Neurology 2006;66:1350–1355.
            • Pawlikowska L, Poon KY, Achrol AS, et al. Apolipoprotein E epsilon 2 is associated with new hemorrhage risk in brain arteriovenous malformations. Neurosurgery 2006;58(5):838–843.
              • Lee CZ, Yao JS, Huang Y, et al. Dose-response effect of tetracyclines on cerebral matriz metalloproteinase-9 after vascular endothelial growth factor hyperstimulation. J Cereb Blood Flow Metab 2006; 26(9):1157–1164.
                • Brown RD Jr, Wiebers DO, Torner JC, O'Fallon WM. Incidence and prevalence of intracranial vascular malformations in Olmsted County, Minnesota, 1965 to 1992. Neurology 1996;46(4):949–952.