腹型过敏性紫癜患儿发生胃肠道出血的影响因素及logistic预测模型构建  

The influencing factors on the occurrence of gastrointestinal bleeding in children with abdominal allergic purpura and the construction of logistic prediction model

作　　者：李春玲 刘永立 曹丽 季米 陈曦 LI Chunling;LIU Yongli;CAO Li;JI Mi;CHEN Xi(The Second Ward of Integrative Medicine,Yongcheng Central Hospital,Yongcheng,Henan 476600,China)

机构地区：[1]永城市中心医院中西医结合二病区,河南永城476600

出　　处：《现代医药卫生》2025年第4期896-899,903,共5页Journal of Modern Medicine & Health

摘　　要：目的分析腹型过敏性紫癜(HSP)患儿发生胃肠道出血的影响因素,并构建logistic预测模型。方法选择2021年9月至2023年9月该院收治的腹型HSP患儿118例,根据胃肠道是否出血分为合并组52例和未合并组66例,收集2组患儿临床资料,比较2组患儿实验室指标,采用多因素logistic回归分析腹型HSP患儿发生胃肠道出血的影响因素。结果合并组患儿与未合并组在皮疹分布情况及呕吐患儿占比方面比较,差异有统计学意义(P<0.05)。合并组患儿白细胞(WBC)、血小板计数(PLT)、C反应蛋白(CRP)、中性粒细胞(NEU)、中性粒细胞/淋巴细胞值(NLR)明显高于未合并组,差异有统计学意义(P<0.05)。多因素logistic回归分析显示,皮疹分布、呕吐、NLR及CRP水平为腹型HSP患儿发生胃肠道出血的独立危险因素(P<0.05)。得到Nomogran风险预测模型,Logit(P)=3.152×X 1+2.158×X 2+2.785×X 6+2.855×X 7,Hosmer-Lemeshoe检验显示预测拟合度良好(χ^(2)=6.528,P=0.574)。受试者操作特征(ROC)曲线结果显示,腹型HSP患儿发生胃肠道出血风险预测模型预测胃肠道出血的曲线下面积(AUC)为0.899[95%可信区间(95%CI)0.848~0.951],灵敏度为81.11%,特异度为85.00%,提示该模型具有较好的预测效能(P<0.05)。结论腹型HSP发生胃肠道出血的危险因素包括皮疹分布、呕吐、NLR及CRP水平,构建了一个具有较高预测准确性的Nomogram预测模型。Objective To analyze the factors influencing the occurrence of gastrointestinal bleeding in children with abdominal purpura(HSP)and to construct a logistic prediction model.Methods A total of 118 children with abdominal HSP admitted to the hospital from September 2021 to September 2023 were selected and divided into 52 cases in the combined group and 66 cases in the non-combined group according to whether they had gastrointestinal bleeding or not,and the clinical data of the two groups of children were collected,and the laboratory indexes of the two groups of children were compared,and the influencing factors of the gastrointestinal bleeding occurring in children with abdominal HSP were analyzed by multiple logistic regression.Results The difference between the children in the combined group and the non-combined group were statistically significant in terms of the rash distribution and the percentage of children who vomited(P<0.05).The levels of WBC,PLT,CRP,NEU and NLR in the combined group were significantly higher than those in the non-combined group,and the differences were statistically significant(P<0.05).Multiple logistic regression analysis showed that rash distribution,vomiting,NLR and CRP levels were independent risk factors for the occurrence of gastrointestinal bleeding in children with abdominal HSP(P<0.05).A risk prediction model was obtained,Logit(P)=3.152×X 1+2.158×X 2+2.785×X 6+2.855×X 7,and the Hosmer-Lemeshoe test showed a good fit of the diagnostic model(χ^(2)=6.528,P=0.574).The receiver operator characteristic(ROC)curve showed that the area under the curve(AUC)of the risk prediction model for gastrointestinal bleeding in children with abdominal HSP for predicting was 0.899(95%CI 0.848-0.951),with a sensitivity of 81.11%and a specificity of 85.00%,suggesting that the model had a good predictive efficacy(P<0.05).Conclusion Risk factors for gastrointestinal bleeding in abdominal HSP include rash distribution,vomiting,NLR and CRP levels,and a Nomogram predictive model with high predictive acc

关 键 词：腹型过敏性紫癜 胃肠道出血 影响因素 Nomogram预测模型 

分 类 号：R554.6[医药卫生—血液循环系统疾病]