儿童手足口病临床特征分析及风险列线图模型构建

Analysis of Clinical Characteristics and Construction of Risk Nomogram Model for Hand,Foot and Mouth Disease in Children

目的分析不同肠道病毒分型感染的手足口病(hand,foot and mouth disease,HFMD)患儿的临床特征,并构建风险列线图预测模型,为HFMD防治提供依据。方法收集2017~2021年山西省儿童医院(山西省妇幼保健院)收治的346例HFMD患儿的病例信息,按不同肠道病毒分为柯萨奇病毒A组16型组(n=26)、肠道病毒71型组(n=56)和肠道病毒通用型组(n=264),比较3组患儿的临床特征,构建儿童HFMD合并脑干脑炎风险列线图模型,对模型预测效果进行验证。结果HFMD患儿年龄主要分布在1~3岁(81.21%),男女比例为1.56∶1,发病主高峰为6~8月、次高峰为10~11月,2017年以肠道病毒71型感染为主,2018~2021年以肠道病毒通用型感染为主。3组之间年龄、住院天数、是否发热、发热分型、发热持续天数、疱疹、斑疹、口部皮疹、膝关节皮疹、白细胞计数、血小板计数、C反应蛋白、降钙素原比较,差异均有统计学意义(P<0.05)。儿童HFMD合并脑干脑炎的发生与住院天数及发热持续天数较长、丘疹及肠道病毒71型阳性呈正相关,与出疹天数较长呈负相关(P<0.05)。列线图模型显示,发生儿童HFMD合并脑干脑炎概率为85.6%,受试者工作特征曲线下面积为0.860。校正曲线显示,实际概率与预测概率一致性较好。决策曲线显示,模型基本对全阈值概率净收益明显。结论儿童HFMD季节性明显,呈双高峰,肠道病毒通用型是近年HFMD优势病原,感染不同肠道病毒的临床特征不同,风险列线图模型对儿童HFMD合并脑干脑炎具有良好的早期预测效果,临床应用价值高,对预防和诊治儿童HFMD有重要意义。

Objective To analyze the clinical characteristics of children with hand,foot and mouth disease(HFMD)infected with different enterovirus subtypes and construct a risk nomogram prediction model to provide basis for HFMD prevention and treatment.Methods The clinical data of 346 cases of children hospitalized with HFMD in Shanxi Children′s Hospital(Shanxi Maternal and Child Health Hospital)from 2017-2021 were collected and divided into Coxsackie virus A16 group(n=26),enterovirus 71group(n=56)and enterovirus general group(n=264)according to different enteroviruses,and the clinical characteristics were compared between the three groups to construct a risk nomogram model in children with HFMD combined with brainstem encephalitis to validate the prediction effect of the model.Results The age of children with HFMD was mainly distributed between 1 and 3 years(81.21%),with a male to female ratio of 1.56∶1,with a primary peak of incidence in June to August and a secondary peak in October to November,with enterovirus 71 type infections predominating in 2017 and enterovirus general infections in 2018-2021.The differences were statistically significant in age,days of hospitalization,presence of fever,fever typing,days of fever duration,herpes,maculopapular rash,oral rash,knee rash,white blood cell,platelet count,C-reactive protein and procalcitonin among the three groups(P<0.05).The occurrence of HFMD combined with brainstem encephalitis in children was positively correlated with longer days of hospitalization and fever duration,papular rash and enterovirus 71 positivity,and negatively correlated with longer days of rash(P<0.05).The nomogram model showed that the probability of occurrence of childhood HFMD combined with brainstem encephalitis was 85.6%,the area under the receiver operating characteristic curve was 0.860.The calibration curve showed that the actual probability was in good agreement with the predicted probability,and the decision curve showed that the net gain of the model essentially to the full threshold probability was obvious.Conclusion HFMD in children has obvious seasonality with double peaks,enterovirus general type is the dominant pathogen of HFMD in recent years,the clinical characteristics of infection with different enteroviruses are different,the risk nomogram model has good early prediction effect on HFMD combined with brainstem encephalitis in children,and has high clinical application value,which is important for the prevention,diagnosis and treatment of HFMD in children.