Comparison of 3.0T MRI with 3D LAVA sequence and digital subtraction angiography for the assessment of accessory hepatic veins in Budd-Chiari syndrome

Abstract

Purpose: To compare 3D liver acceleration volume acquisition (LAVA) and digital subtraction angiography (DSA) for evaluating the presence of accessory hepatic veins (AHV) in Budd-Chiari syndrome (BCS). Materials and Methods: This was a retrospective study in 228 patients with BCS who underwent 3.0T magnetic resonance imaging (MRI) with the 3D LAVA sequence. Two reviewers noted AHV: openings located in the inferior vena cava (IVC), caliber, and the angle of entering into the IVC. MRI results were compared to DSA. Kappa statistics were calculated to quantify intrareader variability in detecting AHVs. Results: On MRI, 63 patients demonstrated no AHV on LAVA images, 70 had one AHV, 62 had two AHVs, 26 patients had three AHVs, six patients had four AHVs, and one patient had five AHVs (P < 0.05 vs. DSA). The mean caliber of the AHVs was 8.3 ± 4.0 mm compared to 9.9 ± 3.2 for DSA (P < 0.001). Among the 301 AHVs, there were 140 with acute angles (46.5%), 71 with right angles (23.6%), and 90 with obtuse angles (29.8%). The prevalence of AHVs on DSA was 54.8% (125/228), while MRI demonstrated 301 AHVs in 165 patients, for a prevalence of 72.4% (165/228) compared to 54.8% for DSA (P = 0.001). The two methods were concordant in only 116/228 (50.9%) patients. The kappa coefficient demonstrated good intrareader consistency for all documented MRI findings of AHVs (κ = 0.626 for caliber and κ = 0.65 for angles). Conclusion: More AHVs were visible on MRI LAVA sequences than on conventional DSA. Level of Evidence: 4. J. Magn. Reson. Imaging 2017;45:401–409.