屈光异常弱视患儿矫正治疗无效的列线图风险模型建立及验证  

Establishment and Validation of A Nomogram Risk Model for the Failure of Corrective Treatment in Children with Amblyopia

作　　者：魏宁 WEI Ning(Department of Ophthalmology,Jianfeng Eye Hospital of Zhengzhou,Zhengzhou Henan 450099)

机构地区：[1]郑州尖峰眼科医院眼科,河南郑州450099

出　　处：《医学临床研究》2024年第12期1874-1877,1881,共5页Journal of Clinical Research

摘　　要：【目的】构建并验证屈光异常弱视患儿矫正治疗无效的列线图风险模型。【方法】回顾性分析2020年1月至2023年6月本院收治的129例屈光异常弱视患儿的临床资料,按照8∶2随机分为训练集(n=104)和验证集(n=25)。所有患儿均给予常规矫正治疗,治疗3个月后,检查屈光状态评估治疗效果,将显效、有效患儿纳入有效组,无效患儿纳入无效组。分析屈光异常弱视患儿矫正治疗效果的影响因素,建立屈光异常弱视患儿矫正治疗无效的预测模型,并进行模型的验证及效能评估。【结果】训练集中23例(22.12%)矫正治疗无效,验证集中7例(28.00%)矫正治疗无效。Logistic回归分析显示:病情程度、视网膜神经纤维层(RNFL)差值、黄斑区视网膜厚度(MRT)差值是屈光异常弱视患儿矫正治疗效果的影响因素(P<0.05)。以上述影响因素为预测变量建立列线图预测模型,C-index指数为0.804(95%CI:0.756~0.828),预测屈光异常弱视患儿矫正治疗无效的校正曲线趋近于理想曲线(P>0.05)。训练集受试者工作特征(ROC)曲线分析结果显示:列线图模型预测屈光异常弱视患儿矫正治疗无效的灵敏度78.26%,特异度为81.48%,曲线下面积(AUC)为0.844;验证集ROC曲线分析结果显示:列线图模型预测屈光异常弱视患儿矫正治疗无效的灵敏度为71.43%,特异度为88.89%,AUC为0.873。【结论】病情程度、RNFL差值、MRT差值是屈光异常弱视患儿矫正治疗效果的影响因素,基于此构建的列线图模型预测屈光异常弱视患儿矫正治疗无效效能良好。【Objective】To establish and validate the risk model of the failure of correction therapy in children with amblyopia.【Methods】The clinical data of 129 children with amblyopia treated in our hospital from January 2020 to June 2023 were retrospectively analyzed and randomly divided into a training set(n=104)and a validation set(n=25)according to 8∶2.All the children were given routine corrective treatment.After 3 months of treatment,the refractive status was checked to evaluate the treatment effect.The effective and effective children were included in the effective group,and the ineffective children were included in the ineffective group.The factors influencing the effect of correction treatment in children with amblyopia were analyzed,and the prediction model of correction treatment ineffectiveness was established,and the model was validated and evaluated.【Results】There were 23 cases(22.12%)in the training set and 7 cases(28.00%)in the verification set.Logistic regression analysis showed that the degree of disease,the difference of retinal nerve fiber layer(RNFL)and the difference of retinal thickness(MRT)in macular area were the factors affecting the effect of correction treatment in children with amblyopia(P<0.05).Using the above influencing factors as predictive variables,a nomogram prediction model was established.The C-index was 0.804(95%CI:0.756-0.828),and the correction curve for predicting the failure of correction treatment in children with amblyopia with abnormal refractive error was close to the ideal curve(P>0.05).ROC curve analysis of the training set showed that the sensitivity,specificity and AUC of the nomogram model were 78.26%,81.48%and 0.844 to predict the failure of corrective treatment in the children with amblyopia.ROC curve results of validation set showed that the sensitivity,specificity and AUC of the nomogram model to predict the failure of correction therapy in children with abnormal refractive amblyopia were 71.43%,88.89%and 0.873.【Conclusion】The degree of illness,RNFL

关 键 词：屈光不正 弱视 儿童 医疗无效 比例危险度模型 

分 类 号：R777.44[医药卫生—眼科]