Objective To explore the value of MSCT in the differentiation of solid pseudopapillary tumor(SPN)from cystic-solid pancreatic carcinoma(PC).Methods A total of 32 pathologically confirmed SPN and 14 pathologically con-firmed cystic-solid pancreatic carcinoma were included in this study.Clinical data and CT findings,including tumor size,lo-cation,shape,margin,component,tumor contrast enhancement,tumor enhancement ratio,"floating cloud sign",calcifica-tion,pancreatic duct dilatation,pancreatic atrophy and invasion or metastases,were retrospectively analyzed and compared.Multivariate logistic regression analysis was used to establish the diagnostic model.The receiver operating characteristic curve(ROC)was used to evaluate the diagnostic efficacy of the model in the differential diagnosis of SPN and cystic-solid PC.Results Patients with SPN usually occurred in young women as compared to PC(mean age,32 vs.64,P<0.001).The tumor size of SPN group was significantly larger than that of cystic-solid PC group(5.41 cm vs.3.90 cm,P =0.017).SPNs usually showed"floating cloud sign",low metastatic rates compared with PCs(84.4%vs.42.9%,6.25%vs.42.9%,P<0.05 for both).There were no significant differences in gender,tumor location,margin,shape,component,tumor contrast enhancement,tumor enhancement ratio,calcification,pancreatic atrophy,or pancreatic duct dilatation between the two groups(P>0.05).Multivariate logistic regression analysis showed that"floating cloud sign"and invasion or me-tastases were independent predictors of SPN and cystic-solid PC.The areas under the curve,sensitivity,specificity and accu-racy of the combined model were 0.853(95%CI:0.714-0.991),78.6%,90.6%and 86.9%,respectively.Conclusion Patients in SPN group were younger than cystic-solid PC group.The tumor size of cystic-solid PC group was smaller than that of SPN group.SPN usually showed"floating cloud sign",well-defined margin and low metastatic rates as compared to cystic-solid PC.
Solid pseudopapillary neoplasm of pancreasCystic solid pancreatic carcinomaDifferential diagnosisMSCT
NETL
NSTL
万方数据
