Diagnostic Efficiency Of Diffusion-weighted Imaging In Differentiating Recurrence Of Glioma From Radiation-induced Brain Injury:A Meta-analysis

PurposeA Meta-analysis was used to evaluate the diagnostic ability of magnetic resonance diffusion weighted imaging(DWI)in differentiating glioma recurrence from radiation-induced brain injury,and to explore whether it can provide some help for clinical differential diagnosis.Materials and MethodsWe systematically searched the English and Chinese databases PubMed,Cochrane Library,EMBASE,CNKI,CBM,Wanfang and Weipu from January 2000 to December 2018,and searched the literature about diffusion-weighted imaging(DWI)in differentiating recurrence and radiation brain injury after brain glioma surgery.The literature was screened according to pre-set literature inclusion and exclusion criteria,and the quality of the included literature was evaluated according to the diagnostic accuracy study quality assessment tool QUADAS-2(Quality Assessment of Diagnostic Accuracy Studies-2).The true positive(TP),true negative(TN),false positive(FP),and false negative(FN)values of each study were extracted from the literature for statistical analysis.The literature quality evaluation and data extraction were independently conducted by two researchers.When the opinions of the two researchers were not uniform,they discuss and reached an agreement.If there were still differences,they discuss with the third researcher to get the results.Statistical analysis and related charting were performed using Stata15.1and RevMan5.3 software.The heterogeneity test was performed using a chi-square test.According to the meta-analysis of the diagnostic test,a bivariate model was used,and then the sensitivity(SEN),specificity(SPE),negative and positive likelihood ratio(-LR,+LR),and diagnostic odds ratio(DOR)were combined,to draw a summary receiver operating characteristic curve(SROC)and calculate the area under the curve(AUC).ResultsThrough literature retrieval,a total of 9 articles,3 in Chinese and 6in English were included,including 358 patients and 360 lesions.RevMan5.3 software was applied to evaluate the quality of literature.The results showed that the quality of 9 literature was good.The chi-square test was used to test the heterogeneity of the literature.The heterogeneity was expressed by I~2.The results of heterogeneity test were Q=2.292,df=2,p=0.159,I~2=13.95%,and 95%confidence interval(CI)was(0-100).The results showed that there was a slight heterogeneity.The combined results of SEN,SPE,+LR,-LR and DOR were 0.85(95%CI 0.78,0.91),0.90(95%CI 0.73,0.96),8.2(95%CI 2.8,24.0),0.16(95%CI 0.10,0.27),50(95%CI 11,221),and the area under SROC curve was 0.92.ConclusionThis meta-analysis showed that MR diffusion-weighted imaging had high sensitivity and specificity in differentiating glioma recurrence from radiation-induced brain injury,and it had high diagnostic accuracy.It was helpful to differentiate glioma recurrence from radiation-induced brain injury in clinical work.