Clinical application value of CT and MRI examination in preoperative evaluation of adjacent organ invasion for periampullary carcinomas / 中华消化外科杂志

ABSTRACT

Objective:To investigate the clinical application value of computed tomography (CT) and magnetic resonance imaging (MRI) examination in preoperative evaluation of adjacent organ invasion for periampullary carcinomas (PACs).Methods:The retrospective and descriptive study was conducted. The clinicopathological data of 81 patients with PACs who were admitted to the Affiliated Wuxi No.2 People′s Hospital of Nanjing Medical University from September 2013 to June 2019 were collected. There were 52 males and 29 females, aged from 41 to 80 years, with an average age of 62 years. Observation indicators: (1) surgical and pathological outcomes; (2) evaluation of adjacent organ invasion on CT and MRI examination for PACs; (3) comparison of diagnostic accuracy between CT and MRI examination in assessing adjacent organ invasion for PACs; (4) auxiliary and feature images of adjacent organ invasion for PACs; (5) comparison between CT and MRI examination in assessing adjacent organ invasion for PACs. Measurement data with skewed distribution were represented as M (range), and count data were described as absolute numbers or percentages. Comparison between groups was analyzed using the Mann-Whitney U test.The receiver operating characteristic curve and area under curve were used to evaluate diagnostic accuracy between CT and MRI examination in assessing adjacent organ invasion for PACs. Consistency was compared using the κ test. Results:(1) Surgical and pathological outcomes: of the 81 patients, 76 underwent pancreatoduodenectomy, 5 underwent palliative gastrojejunostomy or biliary drainage combined with biopsy, including the pancreas, duodenum, or lymph nodes. Of the 81 patients, 35 had pancreatic head carcinoma including 26 with duodenal invasion and 9 without duodenal invasion; 23 had ampullary carcinoma including 17 with duodenal invasion, 4 with both duodenal invasion and pancreatic invasion, and 2 without duodenal invasion or pancreatic invasion; 17 had distal bile duct carcinoma (including papillary type in 4 patients and periductal infiltrative type in 13 patients), of which 8 had duodenal invasion, 1 had duodenal invasion and pancreatic invasion (pathological classification of the 9 patients was periductal infiltrative type), 8 had neither duodenal invasion nor pancreatic invasion; 6 had duodenal carcinoma including 4 with pancreatic invasion and 2 without pancreatic invasion. (2) Evaluation of adjacent organ invasion on CT and MRI examination for PACs: of the 35 patients with pancreatic head carcinoma, duodenal invasion was identified in 25 patients and no duodenal invasion in 10 patients on both CT and MRI examination. Of the 23 patients with ampullary carcinoma, duodenal invasion, pancreatic invasion, both duodenal invasion and pancreatic invasion, and neither duodenal invasion nor pancreatic invasion were identified in 17, 1, 4, and 1 patients on CT examination, respectively; the above indicators were identified in 15, 2, 4, and 2 patients on MRI examination. Of the 17 patients with distal bile duct carcinoma, pancreatic invasion, both duodenal invasion and pancreatic invasion, and neither duodenal invasion nor pancreatic invasion were identified in 8, 1, and 8 patients on CT examination, respectively; the above indicators were identified in 9, 1, and 7 patients on MRI examination. Of the 6 patients with duodenal carcinoma, pancreatic invasion and no pancreatic invasion were identified in 3 and 3 patients on both CT and MRI examination.(3) Comparison of diagnostic accuracy between CT and MRI examination in assessing adjacent organ invasion for PACs: two reviewers had good agreement in assessing adjacent organ invasion on CT examination for pancreatic head carcinoma, ampullary carcinoma, and distal bile duct carcinoma ( κ=0.868, 0.701, 0.881, P<0.05), but they had poor agreement for duodenal carcinoma ( κ=0.333, P>0.05). Meanwhile, two reviewers had good agreement in assessing adjacent organ invasion on MRI examination for pancreatic head carcinoma and ampullary carcinoma( κ=0.860, 0.747, P<0.05), and moderate agreement for distal bile duct carcinoma ( κ=0.643, P<0.05), but they had poor agreement for duodenal carcinoma ( κ=0.333, P>0.05). (4) Auxiliary and feature images of adjacent organ invasion for PACs: for the 25 patients who had pancreatic head carcinoma with duodenal invasion on CT and MRI examination, based on well filling in duodenum, 12 patients showed locally morphological change of lumen and flattened or disappeared duodenal mucosal folds on negative contrast CT cholangiopancreatography; 14 patients showed similar signs on T2 weighted imaging or magnetic resonance cholangiopancreatography. The 17 patients who had distal bile duct carcinoma with pancreatic invasion on CT and MRI examination were periductal infiltrative type. Pancreatic invasion manifested as local thickenness of ductal wall with marked enhancement and narrowed ductal lumen, which was indistinguishable from the pancreas, and the pancreatic parenchyma showed hyperdense or hyperintense signs similar with the lesion, like a "transmural" sign. One patient with both duodenal invasion and pancreatic invasion showed locally thickened and enhanced duodenal wall on both CT and MRI examination. Four patients, who had papillary type distal bile duct carcinoma with neither duodenal invasion nor pancreatic invasion, showed intraductal growing mass which had a discernible boundary to the pancreas and slighter enhancement than infiltrative type on both CT and MRI examination. (5) Comparison between CT and MRI examination in assessing adjacent organ invasion for PACs: CT examination evaluating adjacent organ invasion for pancreatic head carcinoma, ampullary carcinoma, distal bile duct carcinoma, and duodenal carcinoma had a sensibility of 92.3%, 90.5%, 88.9%, 75.0%, a specificity of 88.9%, 50.0%, 87.5%, 100.0%, an accuracy of 0.906, 0.702, 0.882, 0.875, respectively. MRI examination evaluating adjacent organ invasion for pancreatic head carcinoma, ampullary carcinoma, distal bile duct carcinoma, and duodenal carcinoma had a sensibility of 88.5%, 85.7%, 88.9%, 75.0%, a specificity of 77.8%, 50.0%, 75.0%, 100.0%, an accuracy of 0.831, 0.679, 0.819, 0.875. There was no significant difference in sensibility for pancreatic head carcinoma, distal bile duct carcinoma, or duodenal carcinoma between CT and MRI examination( χ2=3.140, 0.141, 0.444, P>0.05), while there was a significant difference in sensibility for ampullary carcinoma ( χ2=13.263, P<0.05). There was no significant difference in specificity for pancreatic head carcinoma, ampullary carcinoma, or distal bile duct carcinoma between CT and MRI examination( χ2=0.321, 2.000, 3.429, P>0.05). There was no significant difference in accuracy for pancreatic head carcinoma, ampullary carcinoma, distal bile duct carcinoma, or duodenal carcinoma between CT and MRI examination( Z=0.967, 0.273, 0.559, 0.000, P>0.05). Conclusion:CT and MRI examination can be used for preoperative evaluation of adjacent organ invasion for periampullary carcinoma, with similar performance in specificity and accuracy, however, CT examination has a higher sensibility for ampullary carcinoma.