Intraindividual comparison of 3-T and 5-T gadoxetic acid-enhanced MRI for evaluating HCC: initial results

Background We aimed to evaluate the utility of 5-T gadoxetic acid (Gd-ethoxybenzyl-diethylenetriaminepentaacetic acid, Gd-EOB-DTPA)-enhanced magnetic resonance imaging (MRI) by intraindividual comparison with 3-T, focusing on image quality and diagnosis of hepatocellular carcinoma (HCC). Methods We prospectively enrolled 28 patients with suspected HCC who underwent dynamic Gd-EOB-enhanced MRI using both 5-T and 3-T scanners. Artificial intelligence-assisted compressed sensing (ACS) and parallel imaging (PI) were both used for hepatobiliary phase (HBP) imaging at 5-T. Two radiologists performed the qualitative and quantitative assessments of image quality, and the evaluation of imaging features. Wilcoxon signed-rank, paired χ ² , and Cochran Q test as well as intraclass correlation coefficients and Cohen κ were used. Results All subjective image quality scores were rated as good to excellent. The subjective scores of contrast-enhanced phases at 5 T were higher than those at 3 T ( p  ≤ 0.016) except for image artifacts. Quantitative measures were also greater at 5 T ( p  ≤ 0.019). Subjective and quantitative assessment of HBP imaging were higher with ACS ( p  ≤ 0.046). The detection rate of enhancing HCC capsule was higher at 5 T ( p  = 0.031), as well as the peritumoral hypointensity on the HBP image at 5 T using ACS ( p  = 0.039). Conclusions Liver dynamic Gd-EOB-DTPA-enhanced 5-T MRI demonstrated superior image quality for contrast-enhanced phases and greater sensitivity in detecting the HCC enhancing capsule compared with 3-T MRI. The integration of 5-T MRI and ACS technology holds the potential to further improve image quality and the assessment of imaging features. Relevance statement Gd-EOB-DTPA-enhanced 5-T MRI provides promising potential for accurate HCC evaluation.