| 高静.基于肠道准备不良的危险因素构建儿童结肠镜检查肠道准备方案[J].安徽医药,待发表. |
| 基于肠道准备不良的危险因素构建儿童结肠镜检查肠道准备方案 |
|
| 投稿时间：2026-04-17 录用日期：2026-06-08 |
| DOI： |
中文关键词: 儿童 结肠镜检查 影响因素 预测模型 肠道准备不良 肠道准备方案? Developing a bowel preparation protocol for colonoscopy in children based on risk factors of poor bowel preparation
|
| 英文关键词: |
| 基金项目:河北省2026年度医学科学研究课题(编号:20261102) |
|
| 摘要点击次数: 134 |
| 全文下载次数: 0 |
| 中文摘要: |
| 目的? 探索儿童结肠镜检查肠道准备不良的危险因素并建立预测模型,构建个体化的肠道准备方案。方法? 选取我院2025年3月至2025年11月拟行结肠镜检查的患儿,按照7:3的比例分为训练集与验证集,根据肠道准备情况分为不良组与良好组,收集所有患儿的临床资料,筛选儿童结肠镜检查肠道准备不良的影响因素并建立预测模型,采用受试者工作特征(ROC)曲线、校正曲线及临床决策(DCA)曲线评价预测模型的价值,并构建儿童结肠镜检查肠道准备方案。结果578例拟行结肠镜检查的患儿中,肠道准备良好416例(72.04%),肠道准备不良162例(27.96%)。患儿年龄、服药后至检查前排便次数、功能性便秘、泻药服用剂量、患儿服药配合度、服药后首次排便时间、检查前最后一次大便性状均为患儿肠道准备不良的独立影响因素。基于上述7个独立影响因素构建儿童结肠镜检查肠道准备不良的预测模型,预测模型在训练集与验证集中的AUC为0.860(95%CI:0.797～0.922)、0.821(95%CI:0.722～0.919),校准度为0.844、0.842；DCA曲线分析,该预测模型在训练集与验证集中均在15%～80%的广泛阈概率范围内临床净获益显著,优于二元决策；据此建立低、高风险分级的个体化肠道准备方案。结论 患儿年龄、服药后至检查前排便次数、功能性便秘、泻药服用剂量、患儿服药配合度、服药后首次排便时间、检查前最后一次大便性状均为患儿肠道准备不良的独立影响因素,构建的预测模型能准确识别肠道准备不良高风险患儿。基于风险分级的儿童结肠镜检查个体化肠道准备方案,有助于实现精准分级管理,提高肠道准备质量。 |
| 英文摘要: |
| Objective To explore the risk factors of poor bowel preparation for pediatric colonoscopy, establish a predictive model, and construct an individualized bowel preparation protocol. Methods Children scheduled for colonoscopy in our hospital from March 2025 to November 2025 were selected and divided into a training set and a validation set at a ratio of 7:3. According to the quality of bowel preparation, they were divided into a poor group and a good group. Clinical data of all children were collected to screen the influencing factors of poor bowel preparation for pediatric colonoscopy and establish a predictive model. Receiver operating characteristic (ROC) curve, calibration curve and decision curve analysis (DCA) were used to evaluate the value of the predictive model, and an individualized bowel preparation protocol for pediatric colonoscopy was constructed. Results Among 578 children scheduled for colonoscopy, 416 cases (72.04%) had good bowel preparation and 162 cases (27.96%) had poor bowel preparation. Children"s age, number of defecations from medication to examination, functional constipation, laxative dosage, children"s medication compliance, time to first defecation after medication, and consistency of the last stool before examination were all independent influencing factors of poor bowel preparation in children. A predictive model for poor bowel preparation in pediatric colonoscopy was constructed based on the above 7 independent influencing factors. The AUC values of the predictive model in the training set and validation set were 0.860 (95%CI: 0.797–0.922) and 0.821 (95%CI: 0.722–0.919), with calibration degrees of 0.844 and 0.842, respectively. DCA curve analysis showed that the predictive model had significant clinical net benefits in both the training set and validation set within a wide threshold probability range of 15%–80%, which was superior to binary decision-making. Based on this, an individualized bowel preparation protocol with low and high risk classification was established. Conclusion Children"s age, number of defecations from medication to examination, functional constipation, laxative dosage, children"s medication compliance, time to first defecation after medication, and consistency of the last stool before examination are all independent influencing factors of poor bowel preparation in children. The constructed predictive model can accurately identify children at high risk of poor bowel preparation. The individualized bowel preparation protocol for pediatric colonoscopy based on risk classification is helpful to realize precision hierarchical management and improve the quality of bowel preparation. |
|
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
