联合检测降钙素原和血小板相关参数对革兰阴性菌血流感染的早期诊断价值

The Early Diagnostic Value of Combined Detection of Procalcitonin and Platelet-Related Parameters in Gram-Negative Bacterial Bloodstream Infections

目的 探讨降钙素原(PCT)和血小板相关参数联合检测对革兰阴性菌血流感染(GN-BSI)的早期诊断价值并构建Logistic预测模型。方法 回顾性选取佛山市顺德区乐从医院2019年1月至2021年4月246例血流感染患者为研究对象,根据分离病原菌种类划分为革兰阴性菌血流感染组(GN-BSI)145例和革兰阳性菌血流感染组(GP-BSI)101例,统计分析两组患者血培养阳性前24h内血清PCT水平及外周血血小板相关参数的差异,单因素和多因素Logistic回归分析筛选GN-BSI的独立预测因素并构建预测模型,采用校准曲线及受试者工作特征曲线(ROC)评价模型的预测效能。结果 与GP-BSI组相比,GN-BSI组PCT、PDW、MPV/PLT、PDW/PLT显著升高(均P<0.05),PLT、Plateletcrit、PLT/PCT、MPV/PCT、PDW/PCT、Plateletcrit/PCT显著降低(均P<0.01);单因素和多因素Logistic回归分析显示GN-BSI的独立预测因素为PCT、PLT、PDW/PCT、MPV/PLT和PDW/PLT,预测模型ROC曲线下面积(AUC)为0.826,最佳截断值0.610,95%置信区间(0.776~0.877),灵敏度71.7%,特异性76.3%。结论 GN-BSI与GP-BSI在PCT和血小板相关参数上存在显著差异,基于PCT和血小板相关参数构建的预测模型对GN-BSI具有较好的早期诊断价值,可为临床经验性抗生素治疗提供参考依据。

Objective To investigate the early diagnostic value of combined detection of serum procalcitonin(PCT)and platelet related parameters in Gram-negative bloodstream infections and to develop a logistic prediction model.Methods 246 patients with bloodstream infections in The Lecong Hospital of Shunde Foshan from January,2019 to April,2021 were retrospectively selected and divided into Gram-negative bloodstream infections group(GN-BSI,n=145)and Gram-positive bloodstream infections group(GP-BSI,n=101)according to the types of isolated pathogens.The differences of serum PCT level and peripheral blood platelet related parameters within 24h before the blood cultures positive time between the two groups were statistically analyzed.Univariate and multivariate logistic regression analysis were applied to screen independent predictors of GN-BSI and construct a prediction model.Calibration curve and receiver operating characteristic curve(ROC)were drawn to evaluate the prediction efficiency of the model.Results Compared with the GP-BSI group,PCT,platelet distribution width(PDW),mean platelet volume-to-platelet count ratio(MPV/PLT)and platelet distribution width-to-platelet count ratio(PDW/PLT)in the GN-BSI group were significantly higher(all P<0.05),while platelet count(PLT),plateletcrit,platelet count-to-procalcitonin ratio(PLT/PCT),mean platelet volume-to-procalcitonin ratio(MPV/PCT),platelet distribution width-to-procalcitonin ratio(PDW/PCT),plateletcrit-to-procalcitonin ratio(Plateletcrit/PCT)were significantly lower(all P<0.01).Univariate and multivariate logistic regression analysis revealed that the independent predictors of GN-BSI were PCT,PLT,PDW/PCT,MPV/PLT and PDW/PLT.The areas under the curve(AUC)of this prediction model was 0.826,with the best cut-off value of 0.610,95%confidence interval(0.776-0.877),sensitivity of 71.7%and specificity of 76.3%.Conclusion There are significant differences in terms of PCT and platelet related parameters between GN-BSI and GP-BSI;The prediction model based on PCT and platelet related parameters has a good early diagnostic value for GN-BSI,which can provide reference basis for clinical empirical antibiotic treatment.