High-Resolution Steady State Magnetic Resonance Angiography of the Carotid Arteries: Are Intravascular Agents Necessary?

Abstract

To prospectively evaluate the potential of gadobenate dimeglumine for high-resolution steady-state (SS) contrast-enhanced magnetic resonance angiography (CE-MRA) of the carotid arteries as an adjunct to conventional first-pass (FP) MRA, with computed tomography angiography (CTA) and digital subtraction angiography (DSA) as reference. Institutional ethics committee approval and written informed consent were obtained. Forty consecutive patients underwent conventional FP MRA with 15 mL gadobenate dimeglumine, using a conventional 3D FLASH sequence (14 sec acquisition time). Immediately afterward, SS images were obtained using a high resolution coronal 3D FLASH sequence (240 sec acquisition time). All patients also underwent CTA and conventional DSA within 8 +/- 3 days. Three experienced radiologists assessed FP and SS image quality and calculated sensitivity, specificity, accuracy, and predictive values for stenosis grade and length, plaque morphology, and tandem lesions using DSA as reference. Detected stenoses were quantified and compared (Spearman rank correlation coefficient, [R(s)]; McNemar test) with DSA and CTA findings. Inter-read variability was assessed using kappa (kappa) statistics. The impact of SS acquisitions on diagnostic confidence and patient management was assessed. MRA FP and SS image quality was excellent in 63 (78.8%) and 46 (57.5%) vessels, adequate in 11 (13.8%) and 20 (25.0%) vessels, and poor in 6 (7.5%) and 14 (17.5%) vessels, respectively. Area under the curve analysis revealed no significant differences between MRA FP, MRA FP + SS, and CTA for the grading of stenoses (P = 0.838; accuracy values of 97.4%, 97.4%, and 98.7%, respectively). Greater accuracy (P < 0.001) was noted for FP + SS images over FP images alone for the assessment of plaque morphology (96.1% for FP + SS images vs. 83.3% for FP). Increased diagnostic confidence was noted for 49 (61.3%) vessels because of additional SS images whereas an impact on final diagnosis was noted in 8 (10%) cases. Good correlation was noted between SS image quality and impact on final diagnosis (R(s) = 0.7; P < 0.0001). SS imaging of the carotid arteries is feasible with gadobenate dimeglumine. The increased spatial resolution attainable allows improved evaluation of stenoses and plaque irregularity, yielding comparable diagnostic performance to that of CTA and DSA.