Purpose
Spinal dural arteriovenous fistulas (SDAVFs) represent the most common indication for a spinal angiography. The diagnostic reference level (DRL) for this specific endovascular procedure is still to be determined. This single-center study provides detailed dosimetrics of diagnostic spinal angiography performed in patients with SDAVFs.
Methods
Retrospective analysis of all diagnostic spinal angiographies between December 2011 and January 2021. Only patients with an SDAVF who had baseline magnetic resonance angiography (MRA), diagnostic digital subtraction angiography (DSA), treatment and follow-up at this institution were included. Dose area product (DAP, Gy cm2) and fluoroscopy time were compared between preoperative and postoperative angiographies, according to SDAVF locations (common versus uncommon), MRA results at baseline (positive versus negative) and DSA protocols (low-dose, mixed-dose, normal-dose). The 75th percentile of the DAP distribution was used to define the local DRL.
Results
A total of 62 spinal angiographies were performed in 25 patients with SDAVF. Preoperative angiographies (30/62, 48%) yielded a significantly higher DAP and longer fluoroscopy time when compared to postoperative angiographies (32/62, 53%) (p < 0.01). The local DRL was 329.41 Gy cm2 for a nonspecific (n = 62), 395.59 Gy cm2 for a preoperative and 138.6 Gy cm2 for a postoperative spinal angiography. Preoperative angiography of uncommonly located SDAVFs yielded a significantly longer fluoroscopy time (p = 0.02). The MRA-based fistula detection had no significant impact on dosimetrics (p > 0.05). A low-dose protocol yielded a 61% reduction of DAP.
Conclusion
The results of the present study suggest novel DRLs for spinal angiography in patients with SDAVF. Dedicated low-dose protocols enable radiation dose optimization in these procedures.