Objective To discuss the imaging findings and pathological features of primary pulmonary lymphoepithelioma-like carcinoma (LELC). Methods Imaging and pathological data of 13 patients (9 male and 4 female, age from 17 to 72 years, median age 52 years) with pathology-proven primary pulmonary LELC from January 2010 to March 2015 were retrospectively analyzed. Thirteen patients underwent X-ray examination, two of them underwent non-enhanced CT scan, eight underwent non-enhanced and enhanced CT scans, one underwent non-enhanced MR scan. All imaging data including location, number, shape, size and density of lesions were analyzed by two experienced chest radiologists respectively and reached an agreement after consultations. Surgery was performed in 11 cases and aspiration biopsy in 2 cases. HE staining was performed in all cases and immunohistochemistry staining in 6 cases. Results All thirteen primary pulmonary LELC were unilateral. Nine peripheral tumors and four central tumors were identified. Lobulations (n=11), spiculas (n=7), vessel convergences (n=3), vascular encasements (n=2), pleural indentations (n=4) and punctuate calcification sign (n=1) were seen in the CT images. Thirteen tumors had diameters ranging from 1.3—11.0 cm, average diameter of (4.4 ± 2.7) cm, and all showed homogeneous density in non-enhanced CT. CT value ranged from 22—48 HU, average value of(34± 10)HU, and during arterial phase it ranged from 33—70 HU, average value of(53 ± 13)HU;and during venous phase it ranged from 43—86 HU, average value of(66±14)HU;the tumor had long T1 and long T2 signal in non-enhanced MR scan, and signals were homogeneous. Pathologically, the tumor was from pleomorphic epithelial with large cells, syncytia in the infiltration of lymphocytes or accompanied with collagenzation. Large nucleus showed vacuole-shape. Immunochemistry CK(+), P63(+), CK5/6(+), CK14(-) supported the diagnosis of LELC. Conclusions There are certain imaging characteristics for primary pulmonary LELC. Histopathological and immunohistochemistry examination could provide the definitive diagnose.

Objective MRI and MR hydrogen proton spectroscopy (1H-MRS) were used to detect the abnormal signal and alteration of metabolites, in order to explore the efficacy of these method in evaluating the damages of central nervous system (CNS) induced by occupational manganese exposure.Methods Eighteen workers exposed to manganese without any manganism symptoms, 12 workers with slightly chronic manganese poisoning, and 19 healthy workers were scanned using routine MRI sequence and 1H-MRS.The blood manganese concentration was also collected for each subject.On cerebral axial T1 WI,the signal intensities of ipsilateral globus pallidus and frontal white matter were measured in the visually brightest area (try to select the signal homogeneous region), and the globus pallidus index (PI) was then calculated.The 1H-MRS data was calculated to get the values of the peak height of N-acetylaspartate (NAA), choline (Cho), inositol (mI) and creatine (Cr) and the ratios of NAA/Cr, Cho/Cr, and mL/Cr were also calculated.One way ANOVA was used to compare the values of PI, NAA/Cr, Cho/Cr, mI/Cr and MnB among the three groups, and the correlations between PI and the time span of manganese exposure or blood manganese concentration were analyzed by Pearson correlation analysis.Eight workers exposed to manganese were followed up one year, and their PI , NAA/Cr before and after follow-up were compared by t test.Results Fourteen of 18 cases exposed to manganese without any manganism symptoms showed symmetrically high intensity signal on T1 WI, while the T2 WI were normal.No high signal intensity was observed on T1WI in any of the healthy workers or manganese poisoning workers.We found that the average PI in manganese exposed group (1.16 ±0.09) was significantly higher (F =24.79 ,P =0.O00)than those of the poisoning ( 1.05 ± 0.07 ) and control groups ( 1.01 ± 0.05 ).The blood manganese concentration in manganese exposed group, the poisoning group and the control group were (0.051 ±0.024), (0.047 ±0.018 ), ( 0.043 ± 0.020 ) μg/ml respectively, which was not significantly different ( F = O.623, P =0.541 ) and did not exceed the upper limit of normal reference value ( < 0.10 μg/ml ).There was a significantly correlation between PI and the time span of manganese exposure ( r = 0.67, P = 0.002 ),however, there was no correlation between PI and blood manganese concentration ( r = 0.20, P = 0.427 ).Furthermore, the NAA/Cr ratio decreased variously in the manganese poisoning group ( 1.22 ± 0.07 ) which was significantly lower( F = 4.120, P = 0.023 ) than those of the poisoning( 1.33 ± 0.13 ) and control groups ( 1.31 ±0.13).No statistical significanees were found in the ratios of Cho/Cr and mI/Cr among these three groups(P>0.05).No obvious changes of the PI and NAA/Cr were found in the 8 manganese exposed workers after 1 year follow-up.Conclusion Manganese exposure could lead to the high intensity signal on T1 WI, therefore the increased PI may be the biomarkers of central nerve system damages caused by the occupational manganese exposure.

Objective ToinvestigatetheenhancedCTfindingsandpathologicalfeaturesoflocalized malignantperitonealmesothelioma (LMPM)andimprovetheaccuracyofdiagnosis.Methods Theimagingandclinicaldataof5casesofLMPMconfirmedbypathology werecollected,thefeaturesofCTenhancedimagingwereanalyzedretrospectivelyandcomparedwiththepathologicalresults.Results Allofthe5caseswereisolatedsolidandcysticmasswithirregularshape,andtherangeofmaximumdiameterofthelesionbeing 8.4-13.3cm,inwhich1casewaspolycystic,andtheother4casesweresolidandcysticmass.CTvalueofthecysticpartwasabout 13.0-27.8 HU,andnoenhancementshowedoncontrast-enhancedphase.Thethicknessofcysticwallandseparationwasuneven, whiletheCTvalueofsolidpartwasabout32.6-40.8HU,andmoderateenhancementwasseeninarterialphase(△CTvaluewasabout30.9-38.4HU ),followedbyslightlyincreaseofenhancementdegreeinvenousphase,andthedecreaseofenhancementdegreeinequilibriumphase. Separationandalotofvesselshadowwereseenin3caseswhileperitonealcavity,pelviceffusionandintraperitonealimplantmetastasiswereseen in1case.Forthepathologicaltypesofpatientsinthisgroup,1casewasepithelialtype,mainlycomposedofcysticcomponent,andthe other4caseswerefibroustypeorbiphasictype,mainlycomposedofcysticandsolid,withsmallcysticdegeneration,necrosis,fibrousseptum, bloodsupplyvessels,andetc.Somelesionsinvadedthesurroundingstructure.Conclusion Thepathologicaltypesoflocalizedmalignantperitoneal mesotheliomaaremostlyfibroustypeandbiphasictypewhichCTfindingsaremostlymanifestedascysticandsolidmasses.Lobulatedand wallnodules,uneventhicknessofcysticwallandseptumcanbeseen,andthedensityofcysticpartialisrelativelyhigher.Solidpartis moderateenhancement,andinvenousphaseismoreobviouswithenhancedvascularseenintheseptum.CTenhancementcombined withclinicalmanifestationsarehelpfultomakeinsuggestivediagnosis.