血清降钙素原、中性粒细胞与淋巴细胞比值联合脑电图监测与儿童热性惊厥严重程度的相关性研究

Correlation of serum procalcitonin, neutrophil-to-lymphocyte ratio, and electroencephalogram monitoring with the severity of pediatric febrile seizures

摘要:

目的 探讨血清降钙素原（PCT）、中性粒细胞与淋巴细胞比值（NLR）联合脑电图（EEG）监测在儿童热性惊厥严重程度评估中的应用价值，并构建预测模型以提高临床早期识别能力。方法 选取2023年1月—2026年1月扬州大学医学院附属扬州市妇幼保健院收治的98例热性惊厥患儿，分为单纯组（62例）与复杂组（36例）。比较两组血清PCT、NLR水平，并比较两组患儿EEG监测指标；通过多因素一般Logistic回归分析筛选儿童复杂性热性惊厥的独立影响因素，构建联合预测模型，并通过受试者工作特征（ROC）曲线评价其预测效能。结果 复杂组血清PCT水平和NLR水平均高于单纯组（P <0.05）。复杂组EEG分类正常率、背景节律正常率和EEG动态变化改善率均低于单纯组（P <0.05）。多因素一般Logistic回归分析结果显示：血清PCT水平高［O^R=2.798（95% CI：1.801，4.347）］、NLR水平高［O^R=2.896（95% CI：1.507，5.565）、EEG总体分类为非特异性异常［O^R=6.391（95% CI：1.093，37.367）］、EEG总体分类为特异性异常［O^R=11.942（95% CI：1.175，121.332）］、背景节律减慢或解体［O^R=8.478（95% CI：1.082，66.430）］、EEG动态变化异常［O^R=0.151（95% CI：0.027，0.853）］均为影响复杂性热性惊厥发生的危险因素（P <0.05）。ROC曲线结果显示，联合模型的预测效能最佳，敏感性为94.4%（95% CI：0.813，0.993），特异性为85.5%（95% CI：0.742，0.931）。结论 血清PCT、NLR及EEG监测指标与儿童热性惊厥严重程度密切相关，三者联合构建的预测模型具有较高的预测效能，可用于儿童复杂性热性惊厥的早期识别。

Abstract:

Objective To investigate the value of serum procalcitonin (PCT), neutrophil-to-lymphocyte ratio (NLR), and electroencephalogram (EEG) in evaluating the severity of pediatric febrile seizures and to develop a predictive model for early risk identification.Methods A total of 98 children with febrile seizures admitted to Yangzhou Maternal and Child Health Hospital Affiliated to Yangzhou University School of Medicine from January 2023 to January 2026 were enrolled and divided into a simple group (n = 62) and a complex group (n = 36). Differences in serum PCT and NLR levels, as well as EEG monitoring indicators, were compared between the two groups. Multivariable logistic regression analysis was employed to identify independent influencing factors for complex febrile seizures, construct a combined prediction model, and evaluate its predictive efficacy using the receiver operating characteristic (ROC) curve.Results Serum PCT and NLR levels were higher in the complex group than in the simple group (P < 0.05). The rates of normal EEG classification, normal background rhythm, and improvement in dynamic EEG changes were lower in the complex group (P < 0.05). Multivariable logistic regression analysis showed that elevated serum PCT [O^R = 2.798 (95% CI: 1.801, 4.347) ], elevated NLR [O^R = 2.896 (95% CI: 1.507, 5.565) ], nonspecific EEG abnormalities [O^R = 6.391 (95% CI: 1.093, 37.367) ], specific EEG abnormalities [O^R = 11.942 (95% CI: 1.175, 121.332) ], slowed or disorganized background rhythm [O^R = 8.478 (95% CI: 1.082, 66.430) ], and abnormal dynamic EEG changes [O^R = 0.151 (95% CI: 0.027, 0.853) ] were independently associated with complex febrile seizures (P < 0.05). ROC curve analysis demonstrated that the combined model had the best predictive performance, with a sensitivity of 94.4% (95% CI: 0.813, 0.993) and a specificity of 85.5% (95% CI: 0.742, 0.931).Conclusion PCT, NLR, and EEG parameters are closely associated with disease severity in pediatric febrile seizures. Their integration provides a reliable model for early identification of high-risk cases.