Spinal cord intradural arteriovenous fistulae: anatomic, clinical, and therapeutic considerations in a series of 32 consecutive patients seen between 1981 and 2000 with emphasis on endovascular therapy.

OBJECTIVE: To review our series of intradural spinal cord arteriovenous fistulas (SCAVFs), analyzing symptoms and related angioarchitecture, and to study the morphological and clinical results of embolization. METHODS: Thirty-two SCAVFs (in 22 adults and 10 children) were treated between 1981 and 2000. These lesions were classified as microarteriovenous fistulas (mAVFs) or macroarteriovenous fistulas (MAVFs) according to shunt morphology. Location, architecture, presenting symptoms, and age group were detailed. The selection of patients for endovascular versus surgical treatments was analyzed, as were the anatomic and clinical results obtained by embolization with n-butylcyanoacrylate. Clinical status was evaluated according to the Karnofsky Performance Scale score. RESULTS: Ten SCAVFs were found in the pediatric population (four mAVFs and six MAVFs). All four mAVFs presented with acute symptoms. Three mAVFs (two cervical and one thoracic) presented hematomyelia; in one patient with a thoracic AVF, subarachnoid hemorrhage was suspected. All six MAVFs were located in the thoracolumbar cord (five associated with hereditary hemorrhagic telangiectasias). Four of the six MAVFs presented with hemorrhage. In the adult population, there were 21 mAVFS (95%) and one MAVF (5%). Only two mAVFs were found in the cervical cord, all other shunts affecting the thoracolumbar region. Hemorrhage was present in 6 of the 22 cases seen in adults (27%). The symptoms of SCAVFs did not differ from those found in spinal cord arteriovenous shunts of nidus type. Pial venous reflux and congestion were the most frequently encountered features in both the adult and pediatric groups. Arterial aneurysms (different from false aneurysms) were not found in association with hemorrhagic presentation of SCAVFs. Mean follow-up in our series was 3.3 years. Of the MAVFs, 86% were embolized, with 67% cured. The others had more than 75% occlusion. All patients followed up improved significantly. Of the mAVFs, 48% were treated endovascularly. Successful embolization was performed in 75% of patients. One patient was not embolized because of vasospasm, whereas 67% percent of mAVFs were completely occluded, 22% were more than 90% occluded, and 11% were 75% occluded. Complementary surgery was deemed unnecessary. All patients with mAVFs improved significantly at follow-up. Transient complications occurred in 22% of all patients, with no permanent morbidity or mortality. No patient bled or rebled after embolization. Thirty-six percent of mAVFs were operated on because of anticipated technical difficulties for endovascular approach or distal localization of the shunt. CONCLUSION: Endovascular treatment of SCAVFs stabilizes, normalizes, or improves neurological symptoms in all patients at long-term follow-up, with no bleeds or rebleeds. Embolization of SCAVFs with glue is a safe treatment that compares favorably with other approaches and significantly improves the poor natural history of the disease.