Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Patients with iAVMs typically present after hemorrhage, seizure, or the appearance of focal neurologic deficits.
Comprehensive evaluation includes CT, MRI, MR angiography, CT angiography, and digital subtraction angiography, with additional studies as required and based on location.
If possible, surgery after rupture is usually delayed to allow for resolution of edema and swelling and for accurate nidus identification.
Treatment of symptomatic lesions often involves multiple interventions, including embolization, surgery, and/or stereotactic radiosurgery.
Surgical excision, with or without preoperative embolization, is recommended for Spetzler-Martin grade I–III iAVMs.
Intracranial arteriovenous malformations (iAVMs) are highly complex and heterogeneous lesions that account for 2% of all hemorrhagic strokes. The malformations vary in size, location, angioarchitecture, and hemodynamics—all of which factor into clinical presentation and rupture risk. Though iAVMs are rare, with an estimated prevalence of 0.2%, they are a leading cause of spontaneous, nontraumatic intracerebral hemorrhage (ICH) in young adults. Approximately 35%–50% of all patients with iAVMs present with symptomatic ICH, with seizures being the second most common presentation. Patients can also present with focal neurologic deficits (FNDs) or cognitive changes secondary to mass effect or ischemic steal. These symptoms, as well as the risk of rehemorrhage, have significant neurologic and functional consequences and merit careful consideration when determining clinical management. Intervention is warranted if the natural history risks outweigh the risks of treatment. Thus judicious management and accurate assessments of risk profiles are critical when triaging patients with symptomatic iAVMs.
Due to the rarity of these lesions, the actual incidence and prevalence of iAVMs are not yet fully known. The best estimates derived from population-based studies report an incidence ranging from 0.8 to 2.05 cases per 100,000 person-years, and an accepted prevalence of 0.2%. These numbers may be skewed by the biased detection of symptomatic lesions, though increasing availability of noninvasive cranial imaging has led to increased diagnosis of incidental, asymptomatic lesions.
Although iAVMs can be clinically silent, they are not benign lesions. Results from an early autopsy study estimated that 12% of affected persons develop clinical symptoms in their lifetime. ICH remains the most common presentation of symptomatic iAVMs. The first-time hemorrhage risk is approximately 2%, and the overall annual hemorrhage rate of untreated iAVMs is estimated at 2%–4%. AVM-associated aneurysms appear to be a major cause of rerupture within the first year. The risk of rerupture is low immediately after the initial event but rises to 6%–18% in the first year before dropping to an annual rate of 2%–7.9%. AVM-associated ICH, however, appears to have more favorable clinical outcomes than ICH from other causes. This may be due to younger age and fewer comorbidities in iAVM patients. Overall, ruptured iAVMs are associated with annual mortality rates of 0.7%–2.9%, with a 51% excess mortality in patients with untreated iAVMs compared to the general population.
Clinical presentations are often categorized as hemorrhagic vs nonhemorrhagic. While ICH is the dominant presentation for patients with iAVMs, approximately 15% of patients are asymptomatic at the time of AVM detection, and another 20% present with seizures. Specific symptoms are often reflective of underlying anatomic and hemodynamic characteristics and should be taken into careful consideration when discussing prognosis and treatment options.
Epileptic seizures occur in an estimated 20%–45% of iAVM patients and are the second most common clinical manifestation of iAVMs. Seizures can occur in the setting of unruptured or ruptured iAVMs and can also develop de novo after intervention. While posthemorrhagic seizures are attributed to hemosiderin deposits or gliotic scarring, the pathogenesis in patients with unruptured iAVMs is not yet well understood. Seizure risk factors include nidus size > 3 cm; long pial draining vein, venous outflow stenosis, venous ectasia, or superficial venous drainage; supratentorial and cortical localization, especially in frontal or temporal lobes; and arterial border zone location. Functional steal—or redirection of blood to the lesion and away from functioning brain tissue—may also be a contributing factor.
Generalized seizures occur in approximately one-third of patients with unruptured iAVMs, while focal seizures occur in approximately 10%. Though antiepileptic drugs (AEDs) offer a 40%–80% chance of achieving 2 seizure-free years in a 5-year follow-up period, epileptic seizures can have a significant impact on quality of life. The risks and benefits of intervention for seizure control remain controversial. A meta-analysis by Baranoski et al. demonstrated that 73% of patients remained seizure-free at the last follow-up after open surgery, compared to 62.9% after stereotactic radiosurgery (SRS) and 50% after endovascular embolization. Conversely, in a Scottish population study, Josephson et al. compared conservative management vs intervention and demonstrated no significant difference in the 5-year risk of unprovoked seizure. Thus it is unclear whether invasive intervention is justified for primary seizure control.
Fixed, temporary, or progressive FNDs are uncommon initial presentations in iAVM patients without underlying hemorrhage or seizures. The reported incidence is low: 7.2% of patients with fixed deficits and 1.3% with progressive deficits. Risk factors include increased nidus size, angiomatous change (or dilated pial-to-pial collaterals from surrounding territories), deep brain or brainstem localization, superficial drainage, and venous ectasias.
The likely pathophysiology of FNDs is mass effect or arterial steal. Mass effect is the mechanical compression of neighboring brain tissue, whereas steal is the shunting of blood into chronically dilated, low-pressure AVM vessels and away from surrounding brain tissue. Whether or not this sump effect is responsible for symptoms is yet unproven. Several cases of ischemic stroke in iAVM patients have been reported, with the authors citing arterial steal as the putative etiology. Though early transcranial Doppler sonography investigation showed no differences in arterial velocities and pulsatility, MRI and CT studies have shown perfusion deficits (as defined by decreased cerebral blood volume, decreased cerebral blood flow, and increased mean transit time) in patients with FNDs. Further studies are needed to determine whether treating the underlying mass effect or perfusion deficits improves neurologic outcomes.
Neurocognitive impairment is a poorly characterized and potentially underreported sequela of iAVMs. In 1948, Olivecrona and Riives reported mental deterioration and memory changes in 11 of 43 iAVM patients. Lower IQ, memory, attention, and other neuropsychologic deficits have since been reported in 24%–71.4% of iAVM patients. However, these studies are confounded by the inclusion of patients with ruptured iAVMs; reported rates decrease significantly when controlling for nonhemorrhagic presentations. For instance, Choi et al. reported cognitive deficits in 10 of 736 patients with untreated, unruptured iAVMs. Mass effect, ischemic steal, and venous hypertension are the most cited etiologies of iAVM-related cognitive changes. Evidence for steal includes findings of deficits associated with the hemisphere contralateral to the iAVM and dystrophic cortical changes remote from the iAVM. Few studies, however, report neurocognitive outcomes, so it remains unclear whether decompressive or obliterative treatment benefits this subset of patients.
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here