Spinal dural arteriovenous fistula: correlation between radiological and clinical findings.

Abstract

The pathophysiology of spinal dural arteriovenous fistulas (SDAVFs) results in perimedullary venous congestion and in turn central cord congestion. Clinically, this presents with progressive neurological dysfunctions that, if diagnosed in a timely fashion, can be at least halted and in part reversed. In SDAVFs, imaging features on MRI and digital subtraction angiography (DSA) have not been studied in conjunction with clinical findings. The primary purpose of the present study was to test if severity of clinical presentation varies in relation to imaging. This retrospective cohort study identified 12 patients treated for SDAVF at the authors' institution. The extent of venous congestion and cord edema was quantified by the number of vertebral levels shown to be affected on DSA and MRI. A modified Aminoff-Logue Scale (ALS) score was assigned at the time of diagnosis and again after definitive therapy. The patients were divided into one of two groups: those with venous congestion < 7 and ≥ 7 vertebral levels seen on DSA and MRI and with central cord edema < 6 and ≥ 6 levels. A t-test was used to assess for a difference in the presenting ALS score between the groups. Patients with ≥ 7 levels of venous congestion reported greater functional disability (DSA: p ≤ 0.001, Cohen's d = 0.509; and MRI: p ≤ 0.001, d = 0.632). Patients with a greater extent of cord edema also reported worse functional disability (p ≤ 0.001, d = 2.31). There was a strong linear correlation between the post- and pretreatment ALS scores (R(2) = 0.86) for those with successful interventions (n = 9). In patients with an SDAVF, the severity of the neurological dysfunction may be predicted by the extent of DSA- and MRI-documented venous congestion and cord edema. There was a strong positive relationship between initial and posttreatment neurological dysfunction.