| Research background:Pancreatic cancer is a highly malignant tumor of the digestive system,and its incidence has increased significantly in recent years.Pancreatic cancer is often occult and is usually detected in the late sta ges of the disease or when it produces oppressive symptoms.The clinical and imaging manifestations of mass chronic pancreatitis and pancreatic cancer are very similar,so it is particularly important to distinguish them accurately in the early stage.The use of traditional imaging methods,such as abdominal ultrasound,CT,MRI and so on,can not accurately and effectively distinguish the two.In recent years,with the development of endoscopic ultrasound(EUS),especially endoscopic ultrasound-guided fine needle aspiration biopsy(EUS-FNA),the diagnostic effect of solid mass of pancreas has been significantly improved to some extent,but there are still limitations in the diagnosis of benign and malignant masses under certain conditions.Contrast-enhanced endoscopic ultrasonography(CE-EUS)is a new technique developed on the basis of endoscopic ultrasonography.Through real-time and dynamic display of microvessels and microstructure of pancreatic lesions,it can make a more accurate judgment of solid space occupying of pancreas.At present,many institutions at home and abroad have studied the application of this technique in pancreatic lesions,but it generally has some shortcomings,such as small sample size,obvious regional differences and so on.Objective:To systematically evaluate the value of CE-EUS in the differentiation of mass chronic pancreatitis and pancreatic cancer.Methods:The published Chinese and English literatures on the differential diagnosis of mass chronic pancreatitis and pancreatic cancer by CE-EUS were searched in China knowledge Network,Wanfang Medical Network,China Biomedical Literature Database,Pubmed,Embsae,cochrane library and other periodical full-text databases.Two researchers independently s creened the retrieved literature according to the screening criteria,and the QUADAS score scale was used to evaluate the quality of the included literature.The data were analyzed by Meta-Disc1.4 statistical software,and the relevant forest map was drawn,and the Deek funnel chart was drawn by stata12.0 software to detect whether there was publication bias.Results:1.The results of literature retrieval: by removing the weight of the preliminary included literature,excluding the literature of review,animal experiment,irrelevant literature,unable to extract four-grid table,and repeatedly published,a total of 20 art icles were included,including 19 English literatures and 1 Chinese literature.the study area is Europe and Asia.The publication period is from 2008 to 2018.All the QUADAS scores included in the study were ? 10.2.Analysis of statistical results: the aggregate sensitivity,specificity,positive likelihood ratio,negative likelihood ratio and diagnostic odds ratio of CE-EUS for differential diagnosis of mass chronic pancreatitis and pancreatic cancer were 86%(95% CIV 83% ·89%),93%(95% CIV 92% ·95%),12.01(95%CI:9.72~14.85),0.16(95%CI:0.10~0.26),respectively,and the sensitivity,specificity,positive likelihood ratio,negat ive likelihood ratio and diagnostic odds ratio for the differential diagnosis of mass chronic pancreatitis and pancreatic cancer were 86%,93%,12.01% and 0.16%,respectively.76.12(the area under 95%CI:54.50~106.31),SROC curve is 0.9655 SE(Q)= 0.0113).3.The results of threshold effect test,Meta regression analysis and sensitivity analysis: in this study,threshold effect test showed that there was no heterogeneity caused by threshold effect;Meta regression analysis showed that some factors that may affect the diagnosis were also non-heterogeneity sources;sensitivity analysis showed that a single article had little influence on the results of Meta analysis.Because of the obvious heterogeneity of sensitivity and negative likelihood ratio,after excl uding 4 articles,the heterogeneity decreased to an acceptable range,suggesting that the source of heterogeneity may be these four articles. |
PDF Full Text Request