摘要
目的:探究腺病毒肺炎患儿临床特点,分析腺病毒肺炎感染影响因素,并构建感染发生预测模型。方法:回顾性分析2018年1月至2020年10月本院收治的104例腺病毒肺炎患儿(研究组)及72例非腺病毒肺炎患儿(对照组)的临床资料,分析腺病毒肺炎患儿临床特点及腺病毒肺炎发生的危险因素,构建危险因素预测模型。结果:腺病毒肺炎患儿主要为男性(64.40%)、年龄≤3岁(79.81%),发病季节主要集中于春夏季(75.96%),症状多表现为发热(87.50%)与咳嗽(75.96%),多存在肺部湿啰音(69.23%),胸部CT可见肺实质变化(71.15%),超过50%患儿出现血红蛋白下降、C反应蛋白上升、中性粒细胞计数下降、淋巴细胞计数下降;多元Logistic回归分析结果显示,血红蛋白<120 g/L、谷草转氨酶>40 U/L、乳酸脱氢酶>300 U/L为腺病毒肺炎发生的危险因素(均P<0.05);危险因素预测模型为:Logit(P)=-3.713+0.622×血红蛋白(1≤120 g/L、0≥120 g/L)+0.408×谷草转氨酶(1≥40 U/L、0≤40 U/L)+0.509×乳酸脱氢酶(1≥300 U/L、0≤300 U/L)。受试者工作特征(ROC)曲线下面积为0.802,95%CI为0.716~0.893,模型阈值、灵敏度、特异度分别为0.512、86.3%、79.5%。结论:儿童腺病毒肺炎的发生易受血红蛋白、谷草转氨酶、乳酸脱氢酶等因素影响,由此构建的危险因素预测模型区分度优异。
Objective: To investigate the clinical characteristics of children with adenovirus pneumonia, analyze the factors influencing adenovirus pneumonia infection, and construct a prediction model for the occurrence of infection. Methods: The clinical data of 104 children with adenovirus pneumonia(study group) and 72 children without adenovirus pneumonia(control group) admitted to our hospital from January 2018 to October 2020 were retrospectively analyzed to analyze the clinical characteristics of children with adenovirus pneumonia and risk factors for its occurrence and to construct a risk factor prediction model. Results: The children with adenovirus pneumonia were mainly male(64.40%), aged ≤3 years(79.81%), and the onset of symptoms was mainly in spring and summer(75.96%), and the most common symptoms were fever(87.50%) and cough(75.96%), wet rales(69.23%), and pulmonary parenchymal changes on chest CT(71.15%);more than 50% of the children had decreased hemoglobin, increased C-reactive protein, decreased neutrophil count, and decreased lymphocyte count. Multiple logistic regression analysis showed that hemoglobin <120 g/L, glutamate transaminase >40 U/L, and lactate dehydrogenase>300 U/L were risk factors for the occurrence of adenovirus pneumonia(all P <0.05);the risk factor prediction model was: Logit(P) =-3.713 + 0.622 × hemoglobin(1 ≤ 120 g/L, 0 ≥ 120 g/L)+ 0.408 × glutamate transaminase(1 ≥ 40 U/L, 0 ≤ 40 U/L) + 0.509 × lactate dehydrogenase(1 ≥ 300 U/L, 0 ≤300 U/L). The area under the receiver operating characteristic(ROC) curve was 0.802, with 95% CI 0.716-0.893,and the threshold, sensitivity, and specificity of the model were 0.512, 86.3%, and 79.5%, respectively. Conclusion: The occurrence of adenovirus pneumonia in children is related to factors such as hemoglobin, glutamate transaminase, and lactate dehydrogenase, and the resulting risk factor prediction model has excellent discrimination.
作者
何蕾
江剑民
陈建愉
陈玉婷
吴金花
林丹
He Lei;Jiang Jianming;Chen Jianyu;Chen Yuting;Wu Jinhua;Lin Dan(Guangzhou Maternal and Children Health Hospital,Guangzhou 510120,China)
出处
《广西医科大学学报》
CAS
2022年第9期1414-1418,共5页
Journal of Guangxi Medical University
基金
广东省科技计划项目(No.2017ZC0383)。
关键词
儿童腺病毒肺炎
临床特点
影响因素
预测模型
adenovirus pneumonia in children
clinical characteristics
influencing factor
prediction model
