Search

Aim: To evaluate the value of gadoxetic acid-enhanced T1-weighted (T1W) magnetic resonance cholangiography (MRC) versus conventional T2-weighted (T2W) MRC compared to endoscopic retrograde cholangiopancreatography (ERCP) in patients with primary sclerosing cholangitis (PSC). Materials and methods: Based on T1W MRC, PSC patients were classified into a regular (RG) and a delayed (DG) excreting group, with an absence of gadoxetic acid in the common bile duct at 20 min. Beading, pruning, and gradation of central bile duct stenosis, evaluated by T1W and T2W MRC, were compared to ERCP. Liver parenchymal enhancement was measured in both study groups and compared to a reference group (n = 20) without a history of liver disease. Two readers performed all measurements. Results: Based on beading and pruning of the peripheral bile ducts, sensitivities, specificities, and accuracies for reader 1 were 0.17/0.43, 0/0.17, and 0.15/0.31 for T1W MRC, and 0.83/0.86, 1/0.83, and 0.85/0.85 for T2W MRC (p = 0.004). For reader 2 sensitivities, specificities, and accuracies were 0.25/0.57, 0/0.33, and 0.23/0.46 for T1W MRC, and 0.92/1, 1/0.83, and 0.92/0.92 for T2W MRC (p = 0.012). Compared to ERCP, central bile duct stenoses were significantly overestimated (p < 0.001) by T2W MRC. A significantly lower parenchymal enhancement was found in the DG (n = 7) compared to the RG (n = 13), and compared to the reference group (p < 0.001). Conclusion: The combined performance of T2W and T1W MRC may provide a comprehensive imaging workup of PSC, including morphological and functional information resulting in optimal management. [Copyright &y& Elsevier]

Abstract: Using conventional MRI methods, the differentiation of primary cerebral lymphomas (PCNSL) and other primary brain tumors, such as glioblastomas, is difficult due to overlapping imaging characteristics. This study was designed to discriminate tumor entities using normalized vascular intratumoral signal intensity values (nVITS) obtained from pulsed arterial spin labeling (PASL), combined with intratumoral susceptibility signals (ITSS) from susceptibility-weighted imaging (SWI). Thirty consecutive patients with glioblastoma (n =22) and PCNSL (n =8), histologically classified according to the WHO brain tumor classification, were included. MRIs were acquired on a 3T scanner, and included PASL and SWI sequences. nVITS was defined by the signal intensity ratio between the tumor and the contralateral normal brain tissue, as obtained by PASL images. ITSS was determined as intratumoral low signal intensity structures detected on SWI sequences and were divided into four different grades. Potential differences in the nVITS and ITSS between glioblastomas and PCNSLs were revealed using statistical testing. To determine sensitivity, specificity, and diagnostic accuracy, as well as an optimum cut-off value for the differentiation of PCNSL and glioblastoma, a receiver operating characteristic analysis was used. We found that nVITS (p =0.011) and ITSS (p =0.001) values were significantly higher in glioblastoma than in PCNSL. The optimal cut-off value for nVITS was 1.41 and 1.5 for ITSS, with a sensitivity, specificity, and accuracy of more than 95%. These findings indicate that nVITS values have a comparable diagnostic accuracy to ITSS values in differentiating glioblastoma and PCNSL, offering a completely non-invasive and fast assessment of tumoral vascularity in a clinical setting. [Copyright &y& Elsevier]