Surgical treatment of spinal dural arterio-venous fistulae: long-term results and analysis of prognostic factors

Summary   Background. Spinal dural arterio-venous fistulae (SDAVF) are slow-flow extramedullary vascular lesions which account for 75–80% of all spinal vascular malformations. At present there is no agreed view with regard to the best therapeutic option being surgical or endovascular, and several reports favour one or other form of management. This is so because of lack of consistent literature, as well as knowledge, concerning the long-term clinical outcome of the patients. The objective of this study is to retrospectively analyse the results obtained with patients operated for a SDAVF at the Department of Neurosurgery of Verona during a 15-year period and to evaluate possible prognostic factors related to neurological outcome. Patients and methods. Between January 1987 and May 2002, 29 patients with SDAVF were operated at the Department of Neurosurgery of Verona. For 25 of these patients we were able to obtain a clinical follow-up using telephone interviews. The patients were evaluated with the Aminoff and Logue’s scale and subsequently stratified into three classes of disability. An overall score (gait and micturition, G + M) of 0–3 indicates a mild disability, a score between 4 and 5 indicates a moderate disability and a score between 6 and 8 a severe disability. All patients underwent surgical treatment which was mainly the first therapeutic option. Following surgery, the patients were re-evaluated with the same neurological scale. We also investigated with statistical analysis the possible impact on clinical outcome of the major clinical, neuroradiological and surgical variables. Results. The epidemiological, clinical, radiological and pathological features of our group of patients are very similar to those previously described in the literature. For 10 patients surgery consisted simply of the interruption of the intradural arterialised draining vein (with or without closure of the small extradural arterial afferents), whereas in the remaining 15 patients coagulation or excision of the fistolous dura was also accomplished. At the last follow-up (mean 7.3 years; in 19 patients longer than 5 years), 10 patients had improved (40%), 11 were stable (44%) and 4 had deteriorated (16%). We determined that only the pre-operative neurological status, described by the G value in the Aminoff and Logue’s scale and the class of disability, had an impact on clinical outcome. Conclusions. This retrospective study confirms that the surgical treatment results of SDAVF are satisfactory even if evaluated after many years. Given these results, and in accordance with the majority of the literature, we concur that surgery should be the first choice treatment for these spinal vascular lesions in order to avoid a dangerous delay and consequently further neurological deterioration. In our group of patients the only prognostic factor statistically related to clinical outcome was the pre-treatment neurological status, particularly the grade of paraparesis and the class of disability. Correspondence: Dr. Paolo Cipriano Cecchi, Operative Unit of Neurosurgery, Regional General Hospital, Via Boehler 5, 39100 Bolzano, Italy.