目的 探讨血浆中性粒细胞胞外网状陷阱（NETs）及相关标志物在儿童社区获得性肺炎（CAP）诊断中的价值。方法 选择收治的160例CAP患儿为CAP组，另选50例健康体检儿童为对照组。CAP组根据病情严重程度分为轻度CAP亚组（n=137）和重度CAP亚组（n=23）；根据病原体不同分为细菌性肺炎亚组（n=78）、支原体肺炎亚组（n=35）和病毒性肺炎亚组（n=47）。检测各组血浆NETs及相关标志物抗菌肽-37（LL-37）、胞外游离DNA （cfDNA）、DNA酶Ⅰ（DNaseⅠ）水平，采用受试者工作特征曲线（ROC）分析各指标对CAP及其病情严重程度的诊断价值。结果 与对照组相比，CAP组NETs、LL-37、cfDNA水平显著增加，DNaseⅠ活性显著降低（P < 0.05）。与轻度CAP亚组相比，重度CAP亚组NETs、LL-37、cfDNA水平显著增加，DNaseⅠ活性显著降低（P < 0.05）。细菌性肺炎亚组、支原体肺炎亚组、病毒性肺炎亚组间NETs、LL-37、cfDNA水平及DNaseⅠ活性差异无统计学意义（P > 0.05）。CAP组NETs与WBC、中性粒细胞百分比、C-反应蛋白（CRP）、降钙素原、肿瘤坏死因子-α呈正相关（分别r=0.166、0.168、0.275、0.181、0.173，P < 0.05）；LL-37、cfDNA与WBC （分别r=0.186、0.338）、CRP （分别r=0.309、0.274）呈正相关（P < 0.05），DNaseⅠ活性与CRP呈负相关（r=-0.482，P < 0.05）。ROC曲线分析显示，NETs、LL-37、cfDNA、DNaseⅠ诊断CAP的曲线下面积（AUC）分别为0.844、0.648、0.727、0.913；最佳截断值分别为182.89、46.26 ng/mL、233.13 ng/mL、0.39 U/mL；灵敏度分别为88.12%、35.63%、54.37%、91.25%；特异度分别为74.00%、92.00%、86.00%、76.00%。鉴别CAP病情严重程度的AUC分别为0.873、0.924、0.820、0.778；最佳截断值分别为257.7、49.11 ng/mL、252.54 ng/mL、0.29 U/mL；灵敏度分别为83.21%、86.96%、78.26%、95.65%；特异度分别为78.26%、83.94%、76.64%、56.93%。结论 CAP患儿血浆NETs及相关标志物水平对CAP诊断及病情严重程度具有一定预测价值。

Objective To study the significance of plasma neutrophil extracellular trap (NET) and its markers in the diagnosis of community-acquired pneumonia (CAP) in children. Methods A total of 160 children with CAP were enrolled as the CAP group, and 50 healthy children were enrolled the control group. According to disease severity, the CAP group was further divided into a mild CAP subgroup with 137 children and a severe CAP subgroup with 23 children. According to the pathogen, the CAP group was further divided into a bacterial pneumonia subgroup with 78 children, a Mycoplasma pneumonia subgroup with 35 children, and a viral pneumonia subgroup with 47 children. The levels of plasma NET and its markers[antibacterial peptide (LL-37), extracellular free DNA (cfDNA), and deoxyribonuclease I (DNase I)] were measured. Receiver operating characteristic (ROC) curve was used to analyze the value of each index in diagnosing CAP and assessing its severity. Results Compared with the control group, the CAP group had significant increases in the levels of NET, LL-37, and cfDNA and a significant reduction in the activity of DNase I (P < 0.05). Compared with the mild CAP subgroup, the severe CAP subgroup had significantly higher levels of NET, LL-37 and cfDNA and a significantly lower activity of DNase I (P < 0.05). There were no significant differences in the levels of NET, LL-37, and cfDNA and the activity of DNase I among the bacterial pneumonia, Mycoplasma pneumonia, and viral pneumonia subgroups (P > 0.05). In the CAP group, plasma NET levels were positively correlated with white blood cell count (WBC), percentage of neutrophils, and serum levels of C-reactive protein (CRP), procalcitonin and tumor necrosis factor-α (r=0.166, 0.168, 0.275, 0.181 and 0.173 respectively, P < 0.05); LL-37 and cfDNA levels were positively correlated with WBC (r=0.186 and 0.338 respectively, P < 0.05) and CRP levels (r=0.309 and 0.274 respectively, P < 0.05); the activity of DNase I was negatively correlated with CRP levels (r=-0.482, P < 0.05). The ROC curve analysis showed that NET, LL-37, cfDNA, and DNase I had an area under the ROC curve (AUC) of 0.844, 0.648, 0.727, and 0.913 respectively in the diagnosis of CAP, with optimal cut-off values of 182.89, 46.26 ng/mL, 233.13 ng/mL, and 0.39 U/mL respectively, sensitivities of 88.12%, 35.63%, 54.37%, and 91.25% respectively, and specificities of 74.00%, 92.00%, 86.00%, and 76.00% respectively. In the assessment of the severity of CAP, NET, LL-37, cfDNA, and DNase I had an AUC of 0.873, 0.924, 0.820, and 0.778 respectively, with optimal cut-off values of 257.7, 49.11 ng/mL, 252.54 ng/mL, and 0.29 U/mL respectively, sensitivities of 83.21%, 86.96%, 78.26%, and 95.65% respectively, and specificities of 78.26%, 83.94%, 76.64%, and 56.93% respectively. Conclusions Plasma NET and its related markers have a certain value in diagnosing CAP and assessing its severity in children.