文章摘要
结直肠癌术后肠梗阻的危险因素分析及预测模型建立
Establishment of a nomogram model predicting risk of postoperative ileus after colorectal cancer surgery
  
DOI:10.12089/jca.2021.07.001
中文关键词: 术后肠梗阻  结直肠癌  危险因素  列线图  预测模型
英文关键词: Postoperative ileus  Colorectal cancer  Risk factors  Nomogram  Predicting model
基金项目:中国医师协会麻醉学医师分会青年麻醉医师科研基金(21800008);西安交通大学第一附属医院临床研究重点项目(XJTU1AF-CRF-2016-003)
作者单位E-mail
卜宁 710061,西安交通大学第一附属医院麻醉手术部  
赵敏 710061,西安交通大学第一附属医院麻醉手术部  
赵莎 710061,西安交通大学第一附属医院麻醉手术部  
杨岚 710061,西安交通大学第一附属医院麻醉手术部  
高媛 710061,西安交通大学第一附属医院麻醉手术部  
高巍 710061,西安交通大学第一附属医院麻醉手术部 gaowei2906@xjtufh.edu.cn 
摘要点击次数: 2714
全文下载次数: 1026
中文摘要:
      
目的 构建结直肠癌(CRC)术后肠梗阻(POI)的列线图风险预测模型并进行验证。
方法 回顾性收集2018年6月至2019年8月接受CRC手术患者413例的围术期临床资料,年龄≥18岁,ASA Ⅰ—Ⅲ级。通过LASSO回归和多因素Logistic回归分析筛选独立危险因素,以此建立列线图模型。通过C-index验证模型的区分度;通过Calibration校正曲线验证模型的一致性;并通过决策曲线分析(DCA)以确定模型的临床有效性。
结果 共有404例CRC患者纳入分析,其中POI患者74例(18.3%)。列线图风险预测模型中包括开腹手术、术中未用非甾体类抗炎药(NSAIDs)、术前白蛋白(Alb)<37.55 g/L和术前球蛋白(Glb)≥28.35 g/L。经内部验证,该模型的C-index为0.799(95%CI 0.746~0.852);Calibration校正曲线显示较好的一致性。DCA曲线表明当POI发生的风险阈值超过4%时,此列线图具有临床使用价值。
结论 基于开腹手术、术中未用NSAIDs、术前Alb<37.55 g/L和术前Glb≥28.35 g/L这4个预测因素构建的列线图预测模型对CRC患者发生POI风险有良好的预测性能。
英文摘要:
      
Objective To establish and validate a novel nomogram for the prediction of the risk of postoperative ileus (POI) after surgery for colorectal cancer (CRC).
Methods A total of 413 patients who underwent elective colorectal surgery from June 2018 to August 2019, aged ≥ 18 years, ASA physical status Ⅰ-Ⅲ were enrolled. LASSO model and multivariable logistic regression analysis was used to identify the independent variables. Then incorporating these factors, the nomogram model was established to predict the risk of POI. The discriminative ability of the nomogram was evaluated by C-index and the calibration ability of the nomogram was evaluated by calibration plot. Decision curve analysis (DCA) was performed to evaluate the clinical usage of the current model.
Results A total of 404 CRC patients were included, of which 74 (18.32%) had POI. Predictors contained in the prediction nomogram included open surgery, no use of nonsteroidal anti-inflammatory drugs (NSAIDs), preoperative albumin (Alb) < 37.55 g/L and preoperative globulin (Glb) ≥ 28.35 g/L. The nomogram model was internally validated and displayed good discrimination with a C-index of 0.799 (95% CI 0.746-0.852). The calibration curve displayed a good agreement and DCA demonstrated that the POI nomogram was clinically useful when intervention was decided at the POI possibility threshold of 4%.
Conclusion This novel POI nomogram incorporating the open surgery, no use of NSAIDs, preoperative Alb < 37.55 g/L and preoperative Glb ≥ 28.35 g/L shows good predictive performance for the risk of POI in CRC patients.
查看全文   查看/发表评论  下载PDF阅读器
关闭