| Objective:(1)To discuss the clinical value of procalcitonin in neonatal sepsis.(2)To evaluate the diagnostic value of procalcitonin or neutrophil CD64 for neonatal sepsis.Methods:(1)100 children(aged 3-28 days)were collected from the Third Affiliated Hospital of Zunyi Medical University and the Affiliated Hospital of Zunyi Medical University from October 2019 to may 2020.According to the diagnostic criteria of neonatal sepsis,The study population were divided into two groups,one of the sepsis group(n=45),the other is non sepsis group(n=55).Then,PCT,WBC,CRP and blood culture were measured,before antibiotics were used.SPSS19.0 software was used for statistical analysis.(2)The databases of Cochrane Library,PubMed,Embase,Web of Science CNKI,Wangfang,China Biomedical Literature Database(CBM),VIP were searched to retrieve studies about estimating the diagnostic value of Neutrophil CD64 or procalcitonin for neonatal sepsis from inception databases to Decemer 2019.Literature screening was performed in accordance with the formulation of inclusion and exclusion criteria.The language is limited to Chinese and English.According to the established inclusion and exclusion criteria,seleted literature and extracted data by two researchers independently.If the results are inconsistent,a third-party researcher will be involved in the discussion.QUADAS tools were used for quality evaluation of the studies.The data were analyzed using Meta-Disc1.4.and Stata14 softwareand.The overal pooled sensitivity(SEN),specificity(SPE),positive likelihood ratio(PLR),negative likelihood ratio(NLR),diagnostic odds ratio(DOR),calculated area under curve(AUC)and Q~*index.drawed summary receiver operating characteristic(SROC)and forest plot.If I~2>50%,it indicates that there is obvious heterogeneity,then the random effect model is chosen to combine the statistics,otherwise the fixed effect model is chosen.In order to explore the source of heterogeneity,Meta-regression,subgroup analysis and sensitivity analysis were performed.Using Deeks'funnel to assess publication bias.Fagan's chart was used to evaluate the clinical application value.Results:(1)(1)When PCT?0.5ng/ml,the PCT level of the sepsis group was significantly different from that the non sepsis group(P<0.05),the sensitivity,specificity,positive predictive value and negative predictive value of PCT are 71.11%,92.72%,88.89%,79.69%;(2)when CRP?10mg/L,there were significant differences of the CRP level between two groups(P<0.05),and the sensitivity,specificity,positive predictive value and negative predictive value of CRP were 48.89%,89.09%,78.57%and 68.06%;(3)the white blood cell count of the sepsis group was compared with the non septic group,P>0.05,that is no significant difference;(4)The sensitivity,specificity,positive predictive value and negative predictive value of PCT were better than those of CRP and WBC.(2)(1)A total of 31 studies were enrolled about CD64,The result showed[SEN=0.82,95%CI(0.80~0.83)],[SPE=0.84,95%CI(0.82~0.85)],[PLR=5.59,95%CI(4.40~7.12)],[NLR=0.18,95%CI(0.13~0.24)],[DOR=36.05,95%CI(23.27~55.87)],AUC=0.9236,Q~*=0.8576.The heterogeneity test indicated that there was a significant non threshold effect.Meta analysis showed that gestational age,type of sepsis,expression mode of detection method,country of study,diagnostic standard and sample number may be the factors leading to heterogeneity.The results of subgroup analysis and sensitivity analysis have no essential change.Viewing the Deeks'funnel plot of CD64,that was publication bias.CD64 positive posterior probability were 81%,negative posterior probability were 10%.(1)A total of 53 studies were enrolled aboutPCT,The result showed,[SEN=0.78,95%CI(0.77~0.80)],[SPE=0.76,[PLR=3.44,95%CI(2.90~4.07)],[NLR=0.24,95%CI(0.19~0.31)],[DOR=15.97,95%CI(11.43~22.33)],AUC=0.8698,Q~*=0.8003.The heterogeneity test showed that there was non threshold effect,I~2>50%.Meta-regression,subgroup analysis and sensitivity analysis did not find the factors leading to heterogeneity Viewing the Deeks'funnel plot of PCT,that was no publication bias.PCT positive posterior probability were 74%,negative posterior probability were 14%.Conclusion:(1)PCT and CRP can be used as reference indexes in the diagnosis of neonatal sepsis,PCT the diagnostic value is better than that of CRP,which can better assist the diagnosis of neonatal sepsis.(2)PCT indicating a medium diagnostic value;CD64 has a high accuracy in the diagnosis of neonatal sepsis,indicating diagnostic value better of CD64 than PCT,and can be used as a reliable indicator for the diagnosis of neonatal sepsis,that is worthy of clinical application. |
PDF Full Text Request