8Dural AV Fistula and Endovascular Treatment
A dural arteriovenous fistula can present as nothing more than a whoosh in one ear — yet some carry a real risk of haemorrhage. Catheter angiography defines it; endovascular embolisation cures it.
FAn abnormal short-circuit behind the ear
A dural arteriovenous fistula (dAVF) is an acquired direct connection between a dural artery — usually a branch of the external carotid system — and a dural venous sinus or vein, bypassing the normal capillary bed. The high-pressure arterial blood pours into a low-pressure venous channel, creating turbulence that is transmitted to the cochlea and heard as a pulse-synchronous noise. When the fistula sits at the transverse or sigmoid sinus, just behind the ear, pulsatile tinnitus is frequently the first and sometimes only symptom [2013].
Arteriovenous malformations (AVMs) are a related but congenital lesion with a true nidus, more often within brain parenchyma. The clinical point for the foundation learner is that an arterial pulsatile tinnitus — one that does NOT stop with neck compression and may have an audible cranial bruit — demands a vascular work-up, because the underlying lesion may be dangerous, not merely annoying [2008].
TCatheter angiography defines the lesion — and the risk
Non-invasive imaging (CT angiography, MR angiography with time-resolved sequences) can raise the suspicion, but digital subtraction angiography (DSA) remains the gold standard. It maps the arterial feeders, the fistulous point and — most importantly — the venous drainage pattern [2003].
The drainage pattern is what stratifies danger. The Cognard and Borden classifications grade a dAVF by whether it drains antegrade into a sinus (benign) or refluxes retrogradely into cortical veins (aggressive). Cortical venous reflux is the red flag: it predicts a meaningful annual risk of intracranial haemorrhage or progressive neurological deficit, and converts a fistula from a nuisance into a lesion that should be treated to prevent catastrophe [2008].
CEndovascular embolisation: the mainstay
For most dAVFs presenting with pulsatile tinnitus, endovascular embolisation is the first-line and often definitive treatment. Under general anaesthesia, a catheter is advanced from the femoral artery or vein to the fistula. The operator then occludes the abnormal connection — ideally the fistulous point itself and the proximal draining vein — using a liquid embolic agent such as Onyx (ethylene-vinyl alcohol copolymer) or n-butyl cyanoacrylate, sometimes combined with coils [2008].
The transarterial route delivers embolic material through the feeding arteries; the transvenous route, used when the affected sinus can be sacrificed, packs the diseased venous compartment directly. Surgical disconnection of the draining vein or, occasionally, stereotactic radiosurgery is reserved for fistulas that cannot be cured endovascularly or that have dangerous anatomy — radiosurgery is unsuitable for high-grade lesions because its effect takes one to three years, leaving the haemorrhage risk unmitigated in the interval [2009].
COutcomes: the noise usually goes for good
When embolisation achieves complete angiographic occlusion of the fistula, pulsatile tinnitus typically resolves immediately and permanently, and the haemorrhage risk associated with cortical reflux is abolished. Series of dAVFs presenting with pulsatile tinnitus report very high rates of symptom cure after successful endovascular obliteration [2009].
Residual or recurrent tinnitus signals incomplete occlusion or recanalisation and warrants repeat angiography. The complications — stroke from non-target embolisation, cranial nerve palsy, venous infarction — are uncommon in experienced neurointerventional hands but real, which is why these patients belong in a multidisciplinary neurovascular service rather than a general clinic [2008].
Which feature most strongly determines that this fistula should be treated rather than observed?
A pulsatile tinnitus that does NOT stop with ipsilateral neck compression and is accompanied by a cranial bruit most suggests:
What is the gold-standard investigation to define a dAVF and stratify its haemorrhage risk?
What is the mainstay of treatment for a dAVF presenting with pulsatile tinnitus?