Evaluation of combined detection of GR,NO and IL-6 in children with viral myocarditis
Objective To improve the diagnostic efficiency of viral myocarditis(VMC)in chil-dren by combining several immune mediated serological indicators,to provide early diagnostic value for clini-cal use,and to avoid adverse outcomes caused by missed detection.Methods 45 eligible VMC cases who were hospitalized in the Pediatric Department of Sihong Hospital from January 2018 to June 2022 were collect-ed and included in the VMC group,45 children with simple virus infection were included in the NVMC group,and 50 normal children who underwent physical examination were included in the control group.Gener-al information of the research subjects was collected,including interleukin-6(IL-6),glutathione reductase(GR),nitric oxide(NO),as well as myocardial enzyme spectrum and troponin results.The value of individu-al indicator of the receiver operating characteristic(ROC)curve was used to analyze the value of VMC,and a logistic regression model was used to build a model to predict VMC.Results The contents of CK-MB,AST,LDH,cTnI,NO and IL-6 in the VMC group were significantly higher than those in the NVMC group and the control group,while the contents of GR were significantly lower than those in the latter two groups;Using each single indicator to analyze the ROC curve of VMC,the results showed that the largest AUC area was NO(0.883),the highest specificity was 0.907 of cTnI,and the highest sensitivity was 0.912 of IL-6;When the di-agnostic cut-off value of the regression model jointly established by five indicators(cTnI,CK-MB,GR,NO,IL-6,etc.)is 0.55,the sensitivity is 0.977,the specificity is 0.982,the maximum Yoden index is 0.959,and the AUC is 0.976.Conclusion The specificity and sensitivity of the regression model constructed with NO,GR,IL-6 combined with cTnI and CK-MB in the diagnosis of VMC are significantly higher than other single indica-tors.It can provide a powerful basis and assistance for clinicians to diagnose VMC in children as early as possible.
NETL
NSTL
万方数据
维普
