Accuracy of non-contrast enhancement multiparametric MRI to diagnose liver tumor compared to conventional MRI sequences

Abstract

Magnetic resonance imaging is widely used to create detailed images of the organs and tissues. In daily clinical practice of diagnosing liver cirrhosis, it is still a challenge to find the best clinical diagnosis for detection hepatic lesions like hemangioma or hepatoma. In this study, we tried to compare the accuracy of method non-contrast enhancement multiparametric MRI (mpMRI) with conventional liver MRI. A total of 120 patients with liver cirrhosis were enrolled and were divided into two groups, 60 patients with hepatic hemangioma (HH, group A) and 60 patients with hepatocellular carcinoma (HCC, group B). MRI was first used for different diagnosis hepatic lesions without any invasive procedure or surgical treatment. All acquired MRI images were divided into 2 parts: the first part was the non-contrast enhancement multiparametric MRI (mpMRI), and the second part named the conventional liver MRI, was covered all images including non-contrast enhancement and dynamic contrast material-enhanced. The MRI images were interpreted by two experience gastrointestinal radiologists (I and II). Before analysis, the clinical manifestations, laboratory data, tissue biopsy results were not known. Paired samples t-test was used to analyze the significant difference of two-tailed probability (P < 0.05), and its Hemangioma and Hepatoma of receiver operating characteristic curve and area under the ROC curves values were evaluated. In the results, two-tailed probability of non-contrast enhancement mpMRI sequences and the conventional liver MRI in group A was found to be P = 0.109, while that in group B was found to be P = 0.115. After Wilcoxon sign-ranked test, the two-tailed probability reached 0.557 in group A, while that of group B reached 0.624; both showed not difference between the two groups. In conclusion, the non-contrast enhancement mpMRI that included the half-Fourier single-shot turbo spin-echo (HASTE) with fat suppression T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI), apparent diffusion coefficient (ADC) and susceptibility-weighted imagining (SWI) sequence can provide reliable information for effectively differentiate HH and HCC without using contrast materials. This study is expected to provide a reference for the application of MRI in clinical diagnose for a better enhanced method.