Study on the application of multimodal MRI findings in the diagnosis of fatty liver with focal nodular lesions

Abstract

Abstract: Objective To compare the differences between focal nodular hyperplasia (FNH) and hepatocellular carcinoma (HCC) in terms of MRI proton density fat fraction (PDFF), diffusion parameters, and dynamic contrast-enhanced perfusion characteristics, and to evaluate their diagnostic performance. Methods A total of 92 patients who underwent liver MRI between December 2022 and September 2025 and were pathologically confirmed to have fatty liver with focal nodular lesions were retrospectively enrolled. According to pathological diagnosis, 55 cases were classified as FNH and 37 cases as HCC. Fat quantification parameters, diffusion parameters, and dynamic enhancement characteristics were compared between the two groups, and their diagnostic performance was evaluated. Results In FNH group, the PDFF in the lesion and the surrounding liver parenchyma were significantly lower than those in HCC group [(9.3±2.1)% vs. (16.7±2.9)%, (18.6±6.2)% vs. (24.9±7.8)%, all P<0.05). In terms of diffusion parameters, ADC and pure diffusion coefficient D in FNH group were significantly higher than those in HCC group [(1.45±0.18) × 10^-3 mm²/s vs. (1.12±0.21) × 10^-3 m²/s; (1.18±0.14) vs. (0.89±0.16), all P<0.05), while the micro-perfusion fraction F was significantly lower than that in HCC group [(18.2±4.1)% vs. (25.7±5.4)%, P<0.05). In the aspect of dynamic enhancement, the enhancement amplitude of FNH group in arterial phase was significantly lower than that of HCC group [(78.6±22.4)% vs. (112.3±28.5)%, P<0.05], while the relative enhancement ratio of portal vein phase and delayed phase was higher [(0.92±0.15) vs. (0.68±0.14), (0.88±0.13) vs. (0.56±0.12), all P<0.05]. ROC analysis showed that the AUC of ADC, enhancement ratio of delay period, elution rate and enhancement mode were 0.90, 0.91, 0.93 and 0.94, respectively, and the AUC of multi-parameter joint model reached 0.97, with sensitivity of 94.5% and specificity of 91.9%. Conclusion Multimodal MRI can effectively distinguish FNH from HCC in the background of fatty liver by assessing fat deposition, tissue structure, cellular density, and perfusion characteristics. Its comprehensive diagnostic performance is significantly superior to any single sequence or parameter.

Key words: Fatty liver, Focal nodular hyperplasia, Hepatocellular carcinoma, Proton density fat fraction

ZHANG He-yi, LI Xin-ran, LIU Xiao-xiao. Study on the application of multimodal MRI findings in the diagnosis of fatty liver with focal nodular lesions[J]. Chinese Hepatolgy, 2026, 31(4): 528-531.

References

[1] Fromenty B, Roden M. Mitochondrial alterations in fatty liver diseases[J]. J Hepatol, 2023, 78(2):415-429.
[2] Hu Y, Hu X, Jiang L, et al. Microbiome and metabolomics reveal the effect of gut microbiota on liver regeneration of fatty liver disease[J]. EBioMedicine, 2025, 111:105482.
[3] Lou T W, Yang R X, Fan J G. The global burden of fatty liver disease: the major impact of China[J]. Hepatobiliary Surg Nutr, 2024, 13(1):119-123.
[4] Rinella M E, Lazarus J V, Ratziu V, et al. A multisociety Delphi consensus statement on new fatty liver disease nomenclature[J]. Ann Hepatol, 2024, 29(1):101133.
[5] Huang Z, Lin X J, Li S S, et al. Differentiating atypical hepatocellular carcinoma from focal nodular hyperplasia: the value of Kupffer phase imaging with sonazoid-contrast-enhanced ultrasound compared to gadodiamide-enhanced MRI[J]. Eur J Radiol, 2025, 184:111991.
[6] Ding Z, Lin K, Fu J, et al. An MR-based radiomics model for differentiation between hepatocellular carcinoma and focal nodular hyperplasia in non-cirrhotic liver[J]. World J Surg Oncol, 2021, 19(1):181.
[7] Le Gout J D, Bolan C W, Bowman A W, et al. Focal nodular hyperplasia and focal nodular hyperplasia-like lesions[J]. Radiographics, 2022, 42(4):1043-1061.
[8] Feng F, Zhao Y. Hepatocellular carcinoma: prevention, diagnosis, and treatment[J]. Med Princ Pract, 2024, 33(5):414-423.
[9] Xia T, Du M, Li H, et al. Association between liver MRI proton density fat fraction and liver disease risk[J]. Radiology, 2023, 309(1):e231007.
[10] Reeder S B, Starekova J. MRI proton density fat fraction for liver disease risk assessment: a call for clinical implementation[J]. Radiology, 2023, 309(1):e232552.
[11] Liang W, Shao J, Liu W, et al. Differentiating hepatic epithelioid angiomyolipoma from hepatocellular carcinoma and focal nodular hyperplasia via radiomics models[J]. Front Oncol, 2020, 10:564307.
[12] 中国研究型医院学会肝病专业委员会, 中国医师协会脂肪性肝病专家委员会, 中华医学会肝病学分会脂肪肝与酒精性肝病学组, 中华医学会内分泌学分会肝病与代谢学组. 脂肪性肝病诊疗规范化的专家建议(2019年修订版)[J]. 实用肝脏病杂志, 2019, 22(6):787-792.
[13] Li W, Li R, Zhao X, et al. Differentiation of hepatocellular carcinoma from hepatic hemangioma and focal nodular hyperplasia using computed tomographic spectral imaging[J]. J Clin Transl Hepatol, 2021, 9(3):315-323.
[14] Yu Y, Lin X, Chen K, et al. Hepatocellular carcinoma and focal nodular hyperplasia of the liver: differentiation with CT spectral imaging[J]. Eur Radiol, 2013, 23(6):1660-1668.
[15] Gatti M, Calandri M, Bergamasco L, et al. Characterization of the arterial enhancement pattern of focal liver lesions by multiple arterial phase magnetic resonance imaging: comparison between hepatocellular carcinoma and focal nodular hyperplasia[J]. Radiol Med, 2020, 125(4):348-355.
[16] Kitao A, Matsui O, Yoneda N, et al. Differentiation between hepatocellular carcinoma showing hyperintensity on the hepatobiliary phase of gadoxetic acid-enhanced MRI and focal nodular hyperplasia by CT and MRI[J]. AJR Am J Roentgenol, 2018, 211(2):347-357.
[17] 郎也, 马春琼, 邓礼巍, 等. 利用磁共振成像多模态参数鉴别肝胆期高信号的肝细胞癌与局灶性结节增生[J]. 中华肝脏病杂志, 2022, 30(10):1069-1073.
[18] 伏林, 张龄尹, 魏倩. MRI在脂肪肝背景下对局灶性结节性病变定性诊断价值分析[J]. 实用肝脏病杂志, 2025, 28(1):140-143.
[19] 李明通. 肝脏局灶性结节性增生和甲胎蛋白阴性肝细胞癌患者的影像学特点分析[J]. 临床肝胆病杂志, 2015(5):729-732.
[20] Mao H Y, Shen B Q, Zhang J Y, et al. Gd-EOB-DTPA enhanced MRI nomogram model to differentiate hepatocellular carcinoma and focal nodular hyperplasia both showing iso- or hyperintensity in the hepatobiliary phase[J]. BMC Med Imaging, 2024, 24(1):211.