2019 年第 8 期 第 14 卷

降钙素原联合炎症指标鉴别革兰阳性与阴性菌致血流感染的临床价值

Clinical value of procalcitonin combined with inflammatory indicators in differentiating bloodstream infection caused by Gram-positive and negative becteria

作者：牛新荣肖东张大权

英文作者：

单位：830001乌鲁木齐，新疆维吾尔自治区人民医院重症医学二科

英文单位：

关键词：血流感染；降钙素原；菌种

英文关键词：

• 摘要：

• 【摘要】目的    探讨降钙素原联合炎症指标鉴别革兰阳性与阴性菌致血流感染的临床价值。方法    选取2017年7月至2018年7月在新疆维吾尔自治区人民医院重症监护病房住院的病原菌血培养阳性且为单一菌株（真菌、混合菌感染除外）的80例患者为研究对象，其中革兰阳性菌感染28例（G+组），革兰阴性菌感染52例（G-组）。对所有患者分别于入院后24 h、3 d、5 d抽取静脉血进行降钙素原、血常规及C反应蛋白（CRP）检测；同时采用受试者工作特性曲线下面积计算各指标诊断临界值、敏感度和特异度。结果    入院后24 h、3 d、5 d，G-组血清降钙素原水平明显高于G+组［（19.0±3.5）μg/L比（7.1±2.0）μg/L、（32.4±4.7）μg/L比（14.0±3.1）μg/L、（58.8±5.4）μg/L比（16.4±5.0）μg/L］，而血小板计数（PLT）水平明显低于G+组，差异均有统计学意义（均P<0.05）。入院后24 h 2组患者的血清CRP水平差异无统计学意义（P>0.05）；入院后3 d、5 d, G-组的血清CRP水平明显高于G+组，差异均有统计学意义（均P<0.05）。G-组血清降钙素原浓度≥10 μg/L者所占比例高于G+组［51.9%（27/52）比10.7%（3/28）］，差异有统计学意义（P<0.01）。绘制ROC曲线，当以降钙素原≥4.3 μg/L、 CRP>15 mg/L和PLT<110×109/L作为临界值时，鉴别诊断革兰阳性菌、阴性菌感染的敏感度分别为70.37%、95.91%和67.43%；特异度分别94.41%、50.48%和90.13%。以上述3项指标同时超过临界值作为诊断标准，敏感度为96.43%（27/28），特异度为92.31%（48/52）。结论    血清降钙素原、CRP水平在细菌致血流感染后均明显升高，而PLT水平表现为下降，降钙素原和PLT的变化早于CRP；革兰阴性菌感染时降钙素原浓度≥10 μg/L占比增高，由此认为降钙素原是革兰阳性与阴性菌致血流感染早期理想的检测指标；相比单一指标测定，3项炎症指标联合测定可提高诊断敏感度与特异度。

• 【Abstract】Objective    To investigate the clinical value of procalcitonin combined with inflammatory indicators in differentiating bloodstream infection caused by Gram-positive and negative becteria. Methods    Eighty patients with positive blood culture of single pathogenic bacteria (except fungi and mixed bacteria) in Intensive Care Unit of People′s Hospital of Xinjiang Uygur Autonomous Region from July 2017 to July 2018 were selected; 28 cases were Gram-positive bacterial infection(G+ group) and 52 cases were Gram-negative bacterial infection(G- group). Venous blood was taken from all patients at 24 h, 3 d and 5 d after admission for procalcitonin, blood routine and C-reactive protein(CRP) detection. Diagnostic critical value, sensitivity and specificity of the indexes were analyzed using receiver operating characteristic（ROC） curve. Results    At 24 h, 3 and 5 d after admission, the levels of serum procalcitonin in G- group were significantly higher than those in G+ group［（19.0±3.5）μg/L vs （7.1±2.0）μg/L, （32.4±4.7）μg/L vs （14.0±3.1）μg/L, （58.8±5.4）μg/L vs （16.4±5.0）μg/L］; platelet count（PLT） in G- group was significantly lower than that in G+ group(all P<0.05). There was no significant difference of serum CRP level between groups at 24 h after admission(P>0.05); the levels of serum CRP at 3 d and 5 d in G- group were significantly higher than those in G+ group(both P<0.05). In patients with serum procalcitonin≥10 μg/L, the proportion of G- infection was higher than that of G+ infection［51.9%（27/52） vs 10.7%（3/28）］(P<0.01). ROC curve analysis showed that the sensitivities of procalcitonin(≥4.3 μg/L), CRP(>15 mg/L) and PLT(<110×109/L) were 70.37%, 95.91% and 67.43% in differential diagnosis of Gram-positive and negative bacterial infection; the specificities were 94.41%, 50.48% and 90.13%. Combined detection of the three indicators showed higher sensitivity and specificity, which were 96.43%(27/28) and 92.31%(48/52), respectively. Conclusions    Serum procalcitonin and CRP increased while PLT decreased in patients with bacterial infection; changes of procalcitonin and PLT are more significant and earlier than CRP. G- infection takes a large proportion in patients with serum procalcitonin level≥10 μg/L, suggesting that procalcitonin may be a early indicator in differentiating bloodstream infection caused by Gram-positive and negative bacteria. Combined detection of procalcitonin, CRP and PLT shows higher diagnostic sensitivity and specificity than individual indicators.