Spontaneous formation of an arteriovenous fistula produced by a ruptured anterior communicating artery aneurysm.

Abstract

An 86-year-old patient presented to the Emergency Department with sudden onset of severe headache. The patient was confused (GCS 13) without objective motor deficit at the time of presentation. The patient's related past history included hypertension, previous stroke and atrial fibrillation. A CT scan revealed a subarachnoid haemorrhage (Figs. 1 and 2). The distribution of the haemorrhage was consistent with rupture of the patient's previously diagnosed anterior communicating artery (ACoA) aneurysm that had been managed conservatively. Two head CT scans prior to rupture had not reported a significant increase in the size of the aneurysm. Digital subtraction angiogram (DSA) showed a low flow arteriovenous (AV) fistula (Figs. 3 and 4) at the dome of the aneurysm draining into an adjacent vein. The low-flow fistula and lack of involvement of other veins were consistent with recent development of the fistula as a result of rupture of the aneurysm. Considering the age and comorbidities of the patient, after discussion with the family, no further intervention or aneurysm repair was performed. The patient did not survive the subarachnoid haemorrhage. The morphologic features of the AV fistula in this patient suggest rupture of the dome of the ACoA aneurysm into an adjacent vein. When a cerebral aneurysm ruptures, bleeding commonly does not breach natural barriers. Rupture of totally intracavernous aneurysms through the lateral wall of the cavernous sinus has been reported.1 Lee et al. believe that aneurysm erosion through the dura of the wall of the cavernous sinus can happen and could cause subarachnoid haemorrhage (SAH).2 According to Shigematsu et al., the aneurysm fills the cavernous sinus, and expands its dura. If the aneurysm ruptures before the subarachnoid space is entered, bleeding may result in acute subdural hematoma.3 In our case, one possible explanation is that adhesions to adjacent structures including veins, as a result of previous aneurysmal haemorrhages, could have allowed the jet of the blood from the point of rupture to enter an adjacent vein leading to development of a fistula. Similar pathology in spontaneous rupture of aortic aneurysm into the vena cava and formation of aortocaval fistula has been reported. In those cases, arterial pulsation causing necrosis of the arterial wall may be the pathophysiologic mechanism. This process leads an adventitial inflammation and subsequent adherence of adjacent veins.4-6 In our patient, chronic pulsation of the aneurysm in presence of adhesions possibly from a previous haemorrhage may have contributed to the formation of the arteriovenous fistula. As far as the authors know similar cases have not been previously reported. Open access publishing facilitated by The University of Sydney, as part of the Wiley - The University of Sydney agreement via the Council of Australian University Librarians. Nathaniel Deboever: Investigation; writing – original draft. Geoffrey Parker: Conceptualization; investigation; validation; writing – review and editing. Behzad Eftekhar: Investigation; supervision; validation; writing – review and editing.