The Evaluation Of Functional MRI In Normal Pancreas And Pancreatic Adenocarcinoma

Part I:Quantitative analysis of normal pancreas and pancreatic adenocarcinoma with dynamic contrast-enhanced MR imaging on a 3.0T systemPurpose:To quantify the perfusion parameters of normal pancreas and pancreatic carcinoma with three-dimension (3D) fast spoiled gradient echo dynamic contrast enhanced (DCE) MRI on 3.0T MR system, and to assess the value of 3D DCE-MRI in the diagnosis of pancreatic carcinoma.Materials and methods:Forty-three patients with pathology verified pancreatic carcinoma and thirty-seven control subjects with normal pancreas (without pancreatic diseases) underwent DCE-MRI with 3D LAVA sequence of ten phases. The data were processed on ADW4.2 workstation. The perfusion parameters of the head, body and tail of normal pancreas, together with lesion and non-lesion area of pancreatic carcinoma were measured and statistically analyzed, including signal enhancement ratio at 30 seconds after injection (SER30), signal enhancement ratio at 90 seconds after injection(SER9o), positive enhancement integral (PEI), time to peak (TTP) and maximum slope of increase (MSI).Results:There was no significant perfusion difference among head, body or tail of normal pancreas. The difference of SER30, PEI, TTP and MSI between lesion and non-lesion region of carcinous pancreas was significant. The TTP between normal pancreas and the non-lesion region of carcinous pancreas was significantly different.Conclusion:Normal pancreas has no regional perfusion difference. The data from DCE-MRI provide reliable information for the diagnosis of pancreatic cancer, and for the assessment of the invasion of pancreatic carcinoma. The difference in TTP between the normal pancreas and non-lesion region of carcinous pancreas suggest the existing of potential lesions. Part II:The metabolic analysis of normal pancreas and pancreatic adenocarcinoma by in vivo proton magnetic resonance spectroscopy at 3.0 TPurpose:To analyze metabolic features of normal pancreas and pancreatic adenocarcinoma by in vivo proton MRS at 3.0 T.Materials and methods:27 control subjects with normal pancreas (without pancreatic diseases) and 29 patients with pathology verified pancreatic adenocarcinoma and matching age and sex underwent single-voxel1H-MRS on a 3.0T MR system (GE Healthcare, HDxt) with 8-channel body coil. Breath-hold PRESS with TE/TR= 35ms/1500ms was used. The peaks area at 1.3 (lipid, Lip),3.2 (total choline, t-Cho) and 5.4 (fatty acids, FA) ppm of the head and body-tail in normal pancreas, together with lesion and non-lesion area in pancreatic carcinoma were measured by SAGE and their ratio to the peak area of none saturated (internal water, InW) at 4.7 ppm was calculated. Statistic analysis was made between different locations.Results:1) In normal pancreas, there were no statistical differences in the ratios of FA/InW and Lip/InW respectively, but t-Cho/InW of body-tail area was greater than that of head.2) In the pancreatic carcinoma, there was significant difference of the ratios in FA/InW and Lip/InW between lesion and non-lesion region respectively. There was no difference in the ratios of t-Cho/InW between lesion and non-lesion region in pancreatic body-tail cance. But in pancreatic head cancer, the ratio of t-Cho/InW in carcious region was smaller than that in non-lesion region.3) There were no statistical differences in the ratios of FA/InW,t-Cho/InWand Lip/InW between normal pancreas and non-lesion region in pancreatic cancer (head vs head, body-tail vs body-tail).Conclusion:The metabolic features of the pancreatic carcinoma including:a) in normal pancreas, t-Cho of body-tail was greater than that of head. b) the FA, t-Cho and Lip were decreased in the carcious region. c) no difference can be found between the normal pancrease and the non-leision region of pancreatic carcinoma.