Stefano Colagrande1, Linda Calistri2, Giulia Grazzini1, Cosimo Nardi1, Simone Busoni3, Giovanni Morana4, Luigi Grazioli5. 1. Department of Experimental and Clinical Biomedical Sciences, Radiodiagnostic Unit n. 2, University of Florence - Azienda Ospedaliero-Universitaria Careggi, Largo Brambilla 3, 50134, Florence, Italy. 2. Department of Experimental and Clinical Biomedical Sciences, Radiodiagnostic Unit n. 2, University of Florence - Azienda Ospedaliero-Universitaria Careggi, Largo Brambilla 3, 50134, Florence, Italy. lindacalistridott@gmail.com. 3. Medical Physics Department, University of Florence - Azienda Ospedaliero-Universitaria Careggi, Largo Brambilla 3, 50134, Florence, Italy. 4. Department of Radiology, General Hospital Ca' Foncello, Piazza Ospedale 1, 31100, Treviso, Italy. 5. Department of Radiology, University of Brescia "Spedali Civili", P.le Spedali Civili 1, 25123, Brescia, Italy.
Abstract
PURPOSE: Our retrospective study sought to describe the spectrum of magnetic resonance imaging (MRI) features of primary hepatic lymphoma (PHL) by analyzing its morphological aspects, signal intensity before and after contrast agent (CA) administration, and diffusion-weighted imaging (DwI) with the apparent diffusion coefficient (ADC) values. METHODS: A retrospective analysis was conducted on 25 patients with pathologically proven PHL who underwent MRI between January 2011 and December 2016. For the evaluation of the ADC, we used a control group of 87 patients (22 with hepatocellular carcinoma, 15 with cholangiocellular carcinoma, 23 with liver metastasis, 22 with focal nodular hyperplasia, and 5 with adenoma). Two radiologists evaluated the morphological features, the signal intensity before and after CA administration, and the DwI. The sensitivity and specificity of the ADC values in distinguishing the PHL lesions from other hepatic lesions were calculated by analyzing the receiver operating characteristic (ROC) curves. RESULTS: Twenty-one patients had non-Hodgkin's lymphoma (18 had diffuse large B-cell lymphoma and three had mucosa-associated lymphoid tissue) and four had Hodgkin's lymphoma (nodular sclerosis). The PHL had a variable morphologic distribution (17 focal mass and eight multiple nodules) and mainly an insinuative growth (24/25). Usually, PHL was usually hypointense on the T1-weighted images (23/25) and hyperintense on the T2-weighted images (21/25); non-specific dynamic enhancement was found after CA administration, but in the hepatobiliary phase, PHL is mainly hypointense (92%). All PHLs showed a signal restriction in the DwI. The sensitivity and specificity in the differential diagnosis between PHL and the other malignant lesions were respectively 81.7% and 100%, with applying an ADC cut-off value of 0.918 × 10-3 mm2/s. CONCLUSION: Although PHL is a rare disease and biopsy is still required, an MRI could be indicative. In our series, PHL showed an insinuative growth, hypointense signal in the hepatobiliary phase, signal restriction in the DwI, and an ADC value lower than that of the other hepatic lesions analyzed.
PURPOSE: Our retrospective study sought to describe the spectrum of magnetic resonance imaging (MRI) features of primary hepatic lymphoma (PHL) by analyzing its morphological aspects, signal intensity before and after contrast agent (CA) administration, and diffusion-weighted imaging (DwI) with the apparent diffusion coefficient (ADC) values. METHODS: A retrospective analysis was conducted on 25 patients with pathologically proven PHL who underwent MRI between January 2011 and December 2016. For the evaluation of the ADC, we used a control group of 87 patients (22 with hepatocellular carcinoma, 15 with cholangiocellular carcinoma, 23 with liver metastasis, 22 with focal nodular hyperplasia, and 5 with adenoma). Two radiologists evaluated the morphological features, the signal intensity before and after CA administration, and the DwI. The sensitivity and specificity of the ADC values in distinguishing the PHL lesions from other hepatic lesions were calculated by analyzing the receiver operating characteristic (ROC) curves. RESULTS: Twenty-one patients had non-Hodgkin's lymphoma (18 had diffuse large B-cell lymphoma and three had mucosa-associated lymphoid tissue) and four had Hodgkin's lymphoma (nodular sclerosis). The PHL had a variable morphologic distribution (17 focal mass and eight multiple nodules) and mainly an insinuative growth (24/25). Usually, PHL was usually hypointense on the T1-weighted images (23/25) and hyperintense on the T2-weighted images (21/25); non-specific dynamic enhancement was found after CA administration, but in the hepatobiliary phase, PHL is mainly hypointense (92%). All PHLs showed a signal restriction in the DwI. The sensitivity and specificity in the differential diagnosis between PHL and the other malignant lesions were respectively 81.7% and 100%, with applying an ADC cut-off value of 0.918 × 10-3 mm2/s. CONCLUSION: Although PHL is a rare disease and biopsy is still required, an MRI could be indicative. In our series, PHL showed an insinuative growth, hypointense signal in the hepatobiliary phase, signal restriction in the DwI, and an ADC value lower than that of the other hepatic lesions analyzed.