Introduction Spinal dural arteriovenous fistulas (SDAVFs) are rare vascular malformations that are difficult to diagnose which could result in delayed treatment. This meta-analysis aims to quantify misdiagnosis rates and limitations of current imaging modalities in detecting SDAVFs.

Methods A systematic literature search was conducted through September 2023 across several databases. Studies describing SDAVF diagnostic techniques were included. Meta-analyses estimated pooled proportions for misdiagnosis rate, diagnostic value of CT/CTA and MRI[AB1]/MRA, and accuracy of digital subtraction angiography (DSA).

Results The meta-analysis included 4055 patients with spinal dural AVFs [AB2] from 105 studies. The mean time from symptom onset to diagnosis was 16.88 months. The rate of dilated vein finding in CT/CTA myelography was 90% with an overall diagnosis rate of 75% (95% CI 68% - 81%). T2-weighted MRI demonstrated flow voids in 87% of cases and hyperintense spinal cord signal in 88%. Contrast-enhanced MRI demonstrated intramedullary enhancement in 72%. Diagnostic digital subtraction angiography accurately identified the lesion in the initial study in almost 100% of cases. Comparisons found flow voids, intramedullary enhancement, and central hyperintensity on MRI to have lower diagnostic accuracy than overall MRA interpretation or DSA (diagnostic odds ratios 0.25, 0.37, and 0.99 respectively). Network meta-analysis detected no significant differences in diagnostic performance between CT, MRI/MRA and DSA.

Conclusion This meta-analysis reveals high SDAVF misdiagnosis rates and limitations of widely-used imaging techniques like MRI and DSA. Development of new non-invasive imaging methods through technological and analytical advances is critical for earlier SDAVF detection and treatment. Enhancing diagnostic precision promises significant improvements in patient outcomes.

Disclosures M. Essibayi: None. A. Mortezaei: None. A. Azzam: None. S. Al-Tous: None. A. Baker: None. D. Altschul: None.