Spinal dural arteriovenous fistulae: Analysis of outcome in relation to pretreatment indicators

OBJECTIVE: The outcome of treatment for a spinal dural arteriovenous fistula is unpredictable. In this study, we reviewed the outcome of patients treated for this condition, in relation to pretreatment indicators. METHODS: We reviewed the records of 37 consecutive patients treated either surgically or endovascularly in our department between 1989 and 2002. RESULTS: At follow-up, 78% of the patients reported improvement. Those with motor symptoms had the best outcome: improvement was achieved in more than 60% of the patients, amounting to 1 or 2 points on the Aminoff scale. Sensory disorders improved in 43% of cases, pain in 61%, and sphincter impairment in only 37%. Patients with a fistula in the lower thoracic region did better than those with fistulae in other locations. CONCLUSION: The clinical state before treatment is certainly relevant in predicting outcome. Age and duration of the symptoms before treatment have no direct relation to the chances of improvement with treatment, particularly when motor improvement is considered. The modality of onset of symptoms and location of the fistula play significant roles, also: a fistula in the lower thoracic segment was associated with more severe symptoms but tended to improve more (83%, P = 0.04) after treatment. Midthoracic and lumbar fistulae showed a lower incidence of improvement (<50%).