降钙素原联合感染可能性评分对肝衰竭患者并发感染的预测价值

Value of procalcitonin combined with Infection Probability Score in predicting infection in patients with liver failure

目的评价降钙素原(PCT)联合感染可能性评分(IPS)对肝衰竭患者并发感染可能性的预测作用。方法回顾性分析2015年1月-2018年6月苏州大学附属第一医院收治的肝衰竭患者的临床资料,根据临床诊断将其分为感染组和非感染组。比较2组患者临床特征、常见的实验室指标以及IPS评分、序贯器官衰竭估计(SOFA)评分。计量资料2组间比较采用t检验或Wilcoxon秩和检验;计数资料2组间比较采用χ^2检验。采用多因素logistic回归分析感染发生的影响因素,并建立PCT联合IPS评分诊断模型,利用受试者工作特征曲线(ROC曲线)分析PCT联合IPS评分对患者发生感染的预测效能。结果最终纳入179例肝衰竭患者,并发感染者123例(68.72%),其中肺部感染99例(80.49%)、腹腔感染49例(39.84%),合并2个及以上部位感染40例(32.52%)。多因素logistic回归分析显示PCT[比值比(OR)=3.822,95%可信区间(95%CI):1.714~8.523,P=0.001]和IPS评分(OR=1.125,95%CI:1.030~1.230,P=0.009)为肝衰竭合并感染的独立预测因素。ROC曲线分析显示,PCT联合IPS评分对肝衰竭并发感染的预测能力最佳[ROC曲线下面积(AUC)=0.857,95%CI:79.7~90.5],显著高于IPS评分(AUC=0.803)和PCT(AUC=0.802),差异均有统计学意义(P值均<0.05),且PCT联合IPS评分的阳性似然比较高(3.40),阴性似然比最低(0.28)。结论 PCT和IPS评分均能预测肝衰竭患者并发感染,其预测效能相当,而二者联合则具有更好的预测价值。

Objective To investigate the value of procalcitonin( PCT) combined with Infection Probability Score( IPS) in predicting the possibility of infection in patients with liver failure.Methods A retrospective analysis was performed for the clinical data of patients with liver failure who were admitted to The First Hospital Affiliated to Soochow University from January 2015 to June 2018.According to the clinical diagnosis,the patients were divided into infection group and non-infection group,and the two groups were compared in terms of clinical features,common laboratory markers,IPS score,and sequential organ failure assessment( SOFA) score.The t-test and the Wilcoxon rank-sum test were used for comparison of continuous data between two groups,and the chi-square test was used for comparison of categorical data between two groups.A multivariate logistic regression analysis was performed to analyze the influencing factors for infection and establish a diagnostic model of PCT combined with IPS score.The receiver operating characteristic( ROC) curve was used to assess the predictive efficiency of PCT combined with IPS score in infection.Results A total of 179 patients with liver failure were enrolled,among whom 123( 68.72%) experienced infection.Among the 123 patients with infection,99( 80.49%) had pulmonary infection,49( 39.84%) had abdominal infection,and 40( 32.52%) had infections at 2 or more sites.The multivariate logistic regression analysis showed that PCT( odds ratio [OR]= 3.822,95% confidence interval [CI]: 1.714-8.523,P = 0.001) and IPS score( OR = 1.125,95% CI: 1.030-1.230,P = 0.009) were independent predictive factors for liver failure with infection.The ROC curve analysis showed that PCT combined with IPS score had the strongest predictive efficiency in liver failure with infection with an area under the ROC curve( AUC) of 0.857( 95% CI:79.7-90.5),with a significantly higher predictive efficiency than IPS alone and PCT alone( 0.857 vs 0.803/0.802,both P<0.05),and PCT combined with IPS score had a positive likelihood ratio as high as 3.40 and a negative likelihood ratio as low as 0.28.Conclusion Both PCT and IPS score can predict infection in patients with liver failure with similar predictive efficiency,while a combination of PCT and IPS score has a better predictive value.