2026 年第 9 期 第 21 卷

基于肺炎支原体DNA和血清支原体抗体及EB病毒抗体构建儿童肺炎支原体脑炎预后评估模型

Construction of prognostic evaluation model for Mycoplasma pneumoniae encephalitis in children based on Mycoplasma pneumoniae DNA, serum mycoplasma antibodies and Epstein-Barr virus antibodies

作者：王倩李彦芳赵力搏张晓双徐茜茹王玉庆

英文作者：Wang Qian Li Yanfang Zhao Libo Zhang Xiaoshuang Xu Qianru Wang Yuqing

单位：河北省儿童医院检验科，石家庄050000

英文单位：Department of Laboratory Hebei Children′s Hospital Shijiazhuang 050000 China

关键词：肺炎支原体DNA；血清支原体抗体；EB病毒抗体；肺炎支原体脑炎；预后评估模型

英文关键词：MycoplasmapneumoniaeDNA;Serummycoplasmaantibodies;Epstein-Barrvirusantibodies; Mycoplasmapneumoniaeencephalitis;Prognosticevaluationmodel

• 摘要：

• 目的 基于肺炎支原体（MP）DNA和血清支原体抗体、EB病毒抗体构建儿童肺炎支原体脑炎（MPIE）列线图预后评估模型，以期为MPIE的早期诊断和预后评估提供更精准的工具。方法 回顾性选取河北省儿童医院2022年1月至2025年1月收治的MPIE患儿319例。患儿出院时采用格拉斯哥预后评分（GOS）评估预后情况，根据GOS评分将患儿分为预后良好组（197例，GOS评分4~5级）和预后不良组（122例，GOS评分1~3级）。比较2组MP DNA、MP-免疫球蛋白M抗体（IgM）、MP-IgG、EB病毒衣壳抗原IgA（VCA-IgA）、EB病毒早期抗原IgG抗体（EA-IgG）、EB病毒核抗原1 IgA（EBNA1-IgA）阳性检出率。采用多因素Logistic回归评估影响MPIE患儿预后不良的独立因素；构建列线图预后评估模型，评估列线图模型的预测效能、校准度以及临床收益。结果 预后不良组病程及意识障碍、言语障碍、癫痫发作、肌力及肌张力异常、脑电图异常、头颅磁共振成像异常、抗病毒治疗比例均长于/高于预后良好组（均P＜0.05）。预后不良组MP DNA、MP-IgM、MP-IgG、VCA-IgA、EA-IgG、EBNA1-IgA阳性检出率均高于预后良好组（均P＜0.001）。多因素Logistic回归分析结果显示，病程、抗病毒治疗、MP DNA、MP-IgM、EBNA1-IgA以及VCA-IgA均为MPIE患儿预后不良的独立危险因素（均P＜0.05）。基于上述因素构建的列线图预后评估模型受试者工作特征曲线下面积为0.855（95%置信区间：0.814~0.897），校正曲线拟合优度较好，且具有良好的阈值概率和临床净收益。结论 基于病程、抗病毒治疗、MP DNA、MP-IgM、EBNA1-IgA以及VCA-IgA构建的儿童MPIE列线图预后评估模型具有较好的预测效能、区分度以及临床实用性。

• Objective To establish a nomogram prognostic evaluation model for children with Mycoplasma pneumoniae encephalitis (MPIE) based on Mycoplasma pneumoniae (MP) DNA, serum mycoplasma antibodies and Epstein-Barr virus (EBV) antibodies, so as to provide a more accurate tool for early diagnosis and prognostic assessment of MPIE. Methods A total of 319 children diagnosed with MPIE admitted to Hebei Children′s Hospital from January 2022 to January 2025 were retrospectively enrolled. The Glasgow outcome scale (GOS) was used to evaluate the prognosis at discharge. The patients were divided into good prognosis group (197 cases, GOS score 4-5) and poor prognosis group (122 cases, GOS score 1-3). The positive rates of MP DNA, MP immunoglobulin M (IgM) antibody, MP-IgG, EBV viral capsid antigen IgA (VCA-IgA), EBV early antigen IgG (EA-IgG) and EBV nuclear antigen 1 IgA (EBNA1-IgA) were compared between the two groups. Multivariate Logistic regression analysis was performed to identify independent risk factors for poor prognosis in children with MPIE. A nomogram prognostic model was established, and its predictive efficiency, calibration and clinical benefit were evaluated. Results The disease course, incidence of consciousness disturbance, speech disorder, seizure, abnormal muscle strength and muscle tension, abnormal electroencephalogram, abnormal cranial magnetic resonance imaging and the proportion of antiviral treatment in the poor prognosis group were significantly higher than those in the good prognosis group (all P<0.05). The positive rates of MP DNA, MP-IgM, MP-IgG, VCA-IgA, EA-IgG and EBNA1-IgA in the poor prognosis group were all higher than those in the good prognosis group (all P<0.001). Multivariate Logistic regression analysis showed that disease course, antiviral treatment, MP DNA, MP-IgM, EBNA1-IgA and VCA-IgA were independent risk factors for poor prognosis in children with MPIE (all P<0.05). The area under the receiver operating characteristic curve of the nomogram model established based on the above factors was 0.855 (95% confidence interval: 0.814-0.897). The calibration curve presented good goodness of fit, and the model had favorable threshold probability and net clinical benefit. Conclusion The nomogram prognostic model for children with MPIE constructed based on disease course, antiviral treatment, MP DNA, MP-IgM, EBNA1-IgA and VCA-IgA possesses good predictive performance, discrimination ability and clinical practicability.