| Objective To provide a statistical basis for clinical diagnosis and treatment by performing a systematic review(meta-analysis) on the accuracy of Synovial C-reactive Protein in the diagnosis of Prosthetic joint infection after arthroplasty,compared with that of Serum C- reactive protein.Methods Retrieve Pub Med( 2010-2016.03)、 EMBase( 2010-2016.03)、 Ovid(2010-2016.03)、Scopus(2010-2016.03), two researchers independently carry on literature screening according to the inclusion criteria and exclusion criteria, evaluate the literature according to the quality assessment tool(QUADAS-2), use Rev Man5.3 to draw the map of risk assessment, use Stata software to draw Deeks funnel plot which shows publication bias. Then extract result data and use Meta-Disc to Meta analysis, which consists of heterogeneity test, combine sensitivity、specificity、positive likelihood ratio、negative likelihood ratio and diagnostic odds ratio, draw a summary receiver operating characteristic(SROC) curve, calculate curve area(SROC AUC) and Q index(Q value).Finally use Z test to compare diagnostic value of Synovial CRP with that of Setum CRP.Results A total of 7 studies with 469 patients meet the selection criteria of the analysis. The combined Sensitivity, specificity, positive likelihood ratio, negative likelihood ratio and diagnostic odds ratio is 0.89(0.82,0.93)、0.91(0.88,0.94)、10.52(6.6,16.77)、0.13(0.08,0.21)、91.22(44.97,185.04), respectively. The area under the curve(AUC) for the CRP assay is 0.9570(0.9445,9695), Q* value is 0.9007. The diagnostic value of Synovial CRP is better than that of Serum CRP, which has statistically significant.Conclusion The diagnostic value of Synovial CRP in PJI is good, and compared with that of Serum CRP,the diagnostic value is better. In the process of screening and diagnosis of PJI, this systematic review provides statistical basis for clinicians to choose the diagnostic method, which has guiding significance. |
PDF Full Text Request