Pretreatment steroids and delayed intervention predict poor long-term outcomes in spinal dural arteriovenous fistulas: a cohort study.

Abstract

Spinal dural arteriovenous fistulas (SDAVFs) are frequently misdiagnosed as inflammatory myelitis, resulting in inappropriate steroid therapy that may trigger acute deterioration. However, the long-term prognostic significance of pretreatment steroid exposure is poorly understood. We aimed to analyze long-term outcomes and identify predictors of recovery in the largest SDAVF cohort to date. This retrospective study analyzed 516 patients with confirmed SDAVFs treated between October 2012 and December 2021. All patients were followed for a minimum of 3 years, with a mean follow-up of 79.38 months. The primary outcome was clinical improvement, defined as a ≥1-point decrease on the modified Aminoff-Logue Scale (mALS). Multivariable logistic regression was used to identify independent prognostic factors. Clinical improvement on the mALS was observed in 68.7% (355/516) of patients . The multivariable analysis revealed three independent prognostic factors: a longer disease course (OR 0.686; 95% CI 0.534 to 0.882) and pretreatment steroid application (OR 0.559; 95% CI 0.401 to 0.778) were associated with poorer prognosis. Conversely, a more severe pretreatment mALS score paradoxically predicted a higher likelihood of improvement (OR 2.458; 95% CI 1.805 to 3.347). Additionally, sensory deficits proved refractory to treatment in most patients, with only 26.1% experiencing relief. The long-term prognosis of treated SDAVFs is multifactorial. A higher mALS score was a strong predictor of greater improvement in patients. This study also highlights that a shorter disease course is strongly predictive of improvement, while pretreatment steroid application is linked to a poorer prognosis. Therefore, prompt and accurate diagnosis and treatment are the prerequisites for improving the prognosis of SDAVFs.