On this page
After bleeding, the first question is immediate care
A ruptured AVM has caused hemorrhage. Sudden severe headache, new weakness, speech or vision disturbance, seizure or collapse needs local emergency assessment. The immediate team assesses the bleeding and its effects; stabilizing that situation is a different task from choosing the later definitive AVM plan. Timing and treatment depend on the clinical condition and anatomy, rather than a universal rule that every rupture requires the same procedure immediately.123
Specialists may need to distinguish bleeding from the nidus from an associated aneurysm or another vascular feature. In selected ruptured cases, targeted embolization of a feature thought responsible for hemorrhage may be considered. That goal is different from complete elimination of the whole AVM. Definitive planning should explain what has been treated and what remains.3
With an unruptured AVM, prevention can itself cause harm
An unruptured lesion has not presented with a recognized hemorrhage. The discussion may concern future bleeding prevention, seizures or other symptoms, and the potential effects of an intervention. Prior hemorrhage, nidus and drainage anatomy, age, neurological function, general health and personal priorities affect the balance. No one characteristic or published group rate supplies a decision for everyone.43
Active observation is a care strategy
Observation means deliberately managing symptoms and arranging reassessment with the treating team while leaving the AVM untreated procedurally. It can avoid the immediate hazards of intervention. It also leaves the lesion present, so ongoing hemorrhage and symptom risks must be discussed. A plan should explain whom to contact about changing symptoms and how uncertainty will be reassessed. Observation need not mean neglect or that later treatment is impossible.45
For an adult with an unruptured lesion, ARUBA provides randomized evidence that medical management alone led to fewer deaths or symptomatic strokes than medical management plus varied procedures during its reported follow-up. The final report supports taking observation seriously. It does not show that every AVM should remain untreated forever, and its population did not include children or ruptured lesions.6
The newer MATCH analysis associated intervention with better estimated hemorrhage-free survival using observational target-trial emulation. It did not randomly assign patients and its main outcome did not cover the entire spectrum of procedural neurological harm. Hemorrhage-free survival is not disability-free survival. The research comparison explains why this association and ARUBA's result cannot be treated as measurements of the same endpoint.5
Comparing the purposes and timing
The options address different goals. This qualitative comparison cannot rank success rates or select a treatment for a particular scan.3
| Option | Purpose and time to effect | Risks and evidence limits |
|---|---|---|
| Active observation | Manages symptoms and reassesses over time; doesn't aim to close the AVM. | Avoids procedural harms while the AVM's own risks continue. ARUBA supports discussing this option for adults with unruptured AVMs6. |
| Microsurgery | Aims for complete removal; the result can be immediate once removal is confirmed. | Bleeding and neurological injury; anatomy and patient selection matter3. |
| Embolization | May prepare for another procedure, target a feature, aim for complete closure or ease symptoms. Timing depends on the goal and stages. | Bleeding or stroke during the procedure; partial reduction isn't complete closure3. |
| Radiosurgery | Aims for gradual closure over years, confirmed by imaging. | Bleeding while waiting for closure, failure to close, and radiation effects3. Results in selected low-grade AVMs don't apply to all7. |
Individualized multidisciplinary selection
A useful discussion includes a cerebrovascular surgeon, endovascular specialist and radiosurgery expertise where relevant, with neurological and rehabilitation support according to need. The team should explain observation as well as each realistic procedural option. Combination treatment may have a purpose, but more procedures also mean additional exposures and a longer treatment course. Local experience matters alongside published evidence.3
The desired endpoint should be explicit: complete shunt obliteration, addressing a particular feature, reducing symptoms, or assisting another treatment. Procedural completion, angiographic obliteration, preserved function and seizure control are separate achievements. Ask how each will be measured, and when a residual lesion would change the plan.3
Questions that keep the decision grounded
Ask why the team recommends this strategy for this age, rupture status and anatomy. Ask what could favor observation, how procedural risks compare with the relevant untreated course, and which outcomes are missing from the evidence cited. Discuss priorities such as independence and symptoms alongside the imaging goal.
References
- US Centers for Disease Control and Prevention. Signs and Symptoms of Stroke. Patient information. Source ↩
- NHS. Epilepsy. Patient information; reviewed 6 March 2025. Source ↩
- De Leacy R et al. J NeuroIntervent Surg. 2022;14:1118–1124. doi:10.1136/neurintsurg-2021-018632. Source ↩
- Derdeyn CP et al. Stroke. 2017;48:e200–e224. doi:10.1161/STR.0000000000000134. Official Top Things to Know summary. Source ↩
- Han H, Chen Y, Ma L et al. JAMA Netw Open. 2025;8:e2543408. doi:10.1001/jamanetworkopen.2025.43408. Source ↩
- Mohr JP et al. Lancet Neurol. 2020;19:573–581. doi:10.1016/S1474-4422(20)30181-2. PubMed ↩
- Graffeo CS et al. Neurosurgery. 2020;87:442–452. doi:10.1093/neuros/nyaa004. Source ↩