| Background:Severe acute pancreatitis has rapid progress,poor prognosis,organ dysfunction in the early stage and death caused by secondary infection in the later stage.Systemic inflammatory response plays an important role in the progression of pancreatitis.In recent years,s TREM-1 has made great progress in the study of inflammatory diseases,and is expected to become a new diagnostic index of inflammation.Therefore,people begin to pay attention to the diagnostic value of s TREM-1 in severe acute pancreatitis.Objective:To perform a meta-analysis of the current literature to evaluate the diagnostic value of s TREM-1 in severe acute pancreatitis.Methods:An electronic literature search of Pub Med,EMBASE,Cochrane database,Google academic,Web of science,China National Knowledge Infrastructure(CNKI)and WANFANG database was performed.The two researchers independently screened the literature,extracted the data.Used QUADAS-2 tools(Revman5.4)to evaluate the quality and risk bias of the included study,and used Metadisc1.4 and Stata15 software for statistical analysis.Results:1.A total of 9 non-randomized studies were included in this meta-analysis.733patients were diagnosed with pancreatitis,including severe acute pancreatitis were found in 293 cases and non-severe acute pancreatitis were present in 437 cases.2.The combined analysis results of SEN,SPE,PLR,NLR,DOR and their 95%confidence intervals are 0.84[95%CI(0.79-0.88)],0.84[95%CI(0.80-0.87)],5.49[95%CI(4.22-7.14)],0.20[95%CI(0.15-0.26)]and 29.77[95%CI(19.08-46.43)]respectively.3.The SROC curve showed that the area under the curve AUC is 0.915,close to 1,indicating that s TREM-1 is a good diagnostic index for severe acute pancreatitis.4.The correlation coefficient of spearman is 0.617,P value is 0.077(>0.05),which is statistically significant.There is no typical"shoulder-arm"point distribution in SROC curve,the results indicated that there is no threshold effect in the included study.The range of I~2value obtained by Q test is 0%50%,suggesting that there is mild heterogeneity in the study.5.According to the order of grouping P value from large to small,Meta regression analysis shows that P>0.1,indicating that there is no heterogeneity in regression analysis.6.In the subgroup analysis,the DOR of the year group changed from 29.77 to22.94,suggesting that the published year may be the source of heterogeneity,and the AUC was 0.9161 in the>2015 group and 0.8963 in the<2015 group,suggesting that s TREM-1 is more effective in the diagnosis of severe acute pancreatitis in the study after 2015.7.By changing the effect model and one by one elimination method for sensitivity analysis,the effect amount and I~2 value did not change significantly,indicating that the Meta-analysis result is robust and reliable.8.The P value of Deeks,funnel chart is 0.76(>0.5),which is statistically significant,suggesting that there is no publication bias in the included study.9.The prior probability of the Fagan map is set to 20%,and the post-test probability becomes 56%and 5%,suggesting that s TREM-1 has high accuracy in the diagnosis of severe acute pancreatitis and great clinical effectiveness.Conclusion:This study is the first time to perform a meta-analysis to evaluate the value of s TREM-1 in diagnosis of severe pancreatitis.s TREM-1 is a good diagnostic index of severe pancreatitis.It can accurately evaluate the progression of pancreatitis.It has high clinical effect and is worth popularizing. |
PDF Full Text Request