结肠镜检查患者肠道准备不佳风险列线图模型的构建和验证
| 作 者：钱月琴, 董文红, 胡敏, 王岚, 邹静, 吴姗 , 庄谦, 田波, 宛新建, 陆群峰 |
| 单 位：上海交通大学医学院附属第六人民医院 |
| 基金项目：国家自然科学基金青年科学基金项目 |
| 摘 要： |
| 背景：良好的肠道准备是高质量结肠镜检查的关键。评估受检者的肠道准备不佳风险有助于制订个体化的肠道准备方案。目的：基于真实世界样本数据探索结肠镜检查肠道准备质量的影响因素，建立可视化风险预测模型。方法：回顾性收集2023年9月—12月上海交通大学医学院附属第六人民医院消化内镜中心符合标准的结肠镜检查数据4 221例，建模队列3 068例，验证队列1 153例。肠道准备不佳定义为Boston肠道准备量表（BBPS）总分<6分或任一肠段评分<2分。以Logistic回归模型分析肠道准备不佳的危险因素并构建风险预测模型，以列线图展示模型，并进行内部和外部验证。结果：纳入研究的受检者中573例（13.6%）肠道准备不佳。多因素Logistic回归分析显示，住院（OR=8.03, 95% CI: 5.94~10.86, P<0.001）、便秘（OR=4.20, 95% CI: 3.29~5.35, P<0.001）、腹部/盆腔手术史（OR=5.73, 95% CI: 4.36~7.53, P<0.001）、年龄≥60岁（OR=3.99, 95% CI: 3.15~5.05, P<0.001）和超重/肥胖[体质指数（BMI）25~30 kg/m2, OR=1.49, 95% CI: 1.13~1.97, P=0.005; BMI >30 kg/m2, OR=4.41, 95% CI: 2.63~7.40, P<0.001]是肠道准备不佳的独立危险因素。肠道准备不佳风险列线图高危评分为≥90分。建模队列和验证队列的ROC曲线下面积（AUC）分别为0.822（95% CI: 0.798~0.845）和0.777（95% CI: 0.719~0.835），校准曲线显示模型预测的概率与实际观察到的概率之间的一致性在可接受范围内。结论：研究构建的结肠镜检查肠道准备不佳风险列线图模型预测效能良好，相应小程序可用于指导制订个体化的肠道准备方案。 |
| 关键词：结肠镜检查；肠道准备；危险因素；风险预测模型；列线图 |
Construction and Validation of A Nomogram Model for Risk of Inadequate Bowel Preparation in Patients Undergoing Colonoscopy
| Author's Name: |
| Institution: |
| Abstract: |
| Background: Adequate bowel preparation leads to high-quality colonoscopy. Assessment of the risk of inadequate bowel cleansing can provide individualized bowel preparation plans. Aims: To identify the risk factors associated with inadequate bowel preparation for colonoscopy based on data of real-world samples, and develop a visualized risk prediction model. Methods: A retrospective observational study was performed among patients undergoing colonoscopy at the Digestive Endoscopy Center, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine from September to December 2023 (development cohort: 3 068 cases, validation cohort: 1 153 cases). Inadequate bowel preparation was defined as Boston bowel preparation scale (BBPS) score <6 points or any individual segment score <2 points. The risk factors for inadequate bowel preparation were explored by Logistic regression model to form a risk prediction model. The prediction model was presented as a nomogram and subjected to both internal and external validation. Results: Inadequate bowel preparation was observed in 573 (13.6%) cases. Multivariate Logistic regression analysis showed that hospitalization (OR=8.03, 95% CI: 5.94-10.86, P<0.001), constipation (OR=4.20, 95% CI: 3.29-5.35, P<0.001), history of abdominopelvic surgery (OR=5.73, 95% CI: 4.36-7.53, P<0.001), age ≥60 years (OR=3.99, 95% CI: 3.15-5.05, P<0.001) and overweight/obesity [body mass index (BMI) 25-30 kg/m2, OR=1.49, 95% CI: 1.13-1.97, P=0.005; BMI >30 kg/m2, OR=4.41, 95% CI: 2.63-7.40, P<0.001] were independent risk factors for inadequate bowel preparation. The high-risk score of the nomogram for inadequate bowel preparation was ≥90 points. The area under the ROC curve (AUC) of the development cohort and validation cohort were 0.822 (95% CI: 0.798-0.845) and 0.777 (95% CI: 0.719-0.835), respectively. The calibration curve demonstrated acceptable agreement between the predicted and actual probabilities. Conclusions: The nomogram model developed to predict the risk of inadequate bowel preparation for colonoscopy showed favorable predictive performance. The nomogram-based web-applet can be utilized to guide the development of individualized bowel preparation plans. |
| Keywords: Colonoscopy; Bowel Preparation; Risk Factors; Risk Prediction Model; Nomogram |
| 投稿时间：2025-04-13 |
|
|
|
查看pdf文件
|
|
