Value of intravoxel incoherent motion diffusion-weighted imaging in differential diagnosis of benign and malignant hepatic lesions and blood perfusion evaluation

Abstract

Objective: To investigate the value of intravoxel incoherent motion diffusion-weighted imaging (IVIM-DWI) in the differential diagnosis and blood perfusion evaluation of benign and malignant hepatic lesions. Methods: A retrospective analysis was performed for 86 patients (96 lesions) with pathologically or clinically confirmed hepatic lesions or hepatic lesions diagnosed based on follow-up results, among whom 48 had malignant lesions (53 lesions) and 38 had benign lesions (43 lesions). The patients underwent conventional magnetic resonance (MR) plain scan, contrast-enhanced scan, and diffusion-weighted imaging (DWI) with different b values (b = 0, 50, 100, 150, 200, 400, 600, 800, 1 000, and 1 200 s/mm2) to determine the parameters of the double exponential model for intravoxel incoherent motion (IVIM): fast diffusion coefficient Dfast, slow diffusion coefficient Dslow, and percentage of fast-diffusion constituent F value. The patients were divided into groups according to the blood supply to lesions on conventional MR plain scan and contrast-enhanced scan, and there were 47 lesions in abundant blood supply group and 49 in poor blood supply group. The data for analysis were Dfast, Dslow, and F values of benign/malignant lesion groups and abundant/poor blood supply groups. The independent samples t-test was used for statistical analysis; the independent samples non-parametric test Mann-Whitney U test was used for the comparison of F value; the receiver operating characteristic (ROC) curve was used to evaluate the value of above parameters in the differentiation of benign and malignant lesions and blood supply evaluation. Results: Compared with the malignant lesion group, the benign lesion group had significantly higher Dslow, and F values (P< 0.001 orP= 0.001) and a higher Dfast value (P= 0.053). Compared with the poor blood supply group, the abundant blood supply group had significantly higher Dfast and F values (P< 0.001 orP= 0.001) and a higher Dslow value (P= 0.185). According to the ROC curve, the cut-off values of Dslow, Dfast, and F values in the diagnosis of benign/malignant hepatic lesions and evaluation of abundant/poor blood supply were 1.18×10-3mm2/s, 27.20×10-3mm2/s, 20.25%, 1.17×10-3mm2/s, 20.30×10-3mm2/s, and 17.80%, respectively, with sensitivities, specificities, accuracy, and areas under the ROC curve of 90.69%/92.45%/91.66%/0.938, 46.51%/73.58%/61.45%/0.589, 74.41%/50.94%/62.50%/0.653, 59.57%/57.14%/58.33%/0.559, 55.32%/63.26%/59.37%/0.618, and 93.61%/89.79%/90.62%/0.961, respectively. Conclusion: The parameter of the double exponential model for IVIM, Dslow value, has a certain value in the differential diagnosis of benign and malignant hepatic lesions, and F value can show blood perfusion in benign and malignant hepatic lesions without the need for contrast-enhanced scan, which provides a reference for the qualitative diagnosis of liver tumor.