直肠癌保肛根治术后低位前切除综合征危险因素分析及列线图预测模型构建  

A nomogram model for predicting risk factors of low anterior resection syndrome after anus-preserving radical resection for rectal cancer

作　　者：卜旻淳 曹先东[1] 周波[1] Bu Minchun;Cao Xiandong;Zhou Bo(Dept of General Surgery,The First Affiliated Hospital of Anhui Medical University,Hefei 230022)

机构地区：[1]安徽医科大学第一附属医院普外科,合肥230022

出　　处：《安徽医科大学学报》2021年第10期1632-1636,共5页Acta Universitatis Medicinalis Anhui

基　　金：国家自然科学基金(编号:81801952)。

摘　　要：目的探讨直肠癌保肛根治术后发生低位前切除综合征(LARS)的列线图风险预测模型的构建及验证。方法分析于安徽医科大学第一附属医院普外科行直肠癌前切除术或保肛手术并顺利出院患者的临床资料,在随访过程中根据患者的LARS量表评分结果,将轻度及重度患者分入LARS组,无症状患者分入无LARS组,各项临床指标的组间比较采用χ2检验,将差异有统计学意义的指标进行多因素Logistics回归分析,筛选出术后LARS发生的独立危险因素,以此构建风险预测模型,并通过内部验证和ROC曲线下面积(AUC)来评价模型的预测性能与准确度。结果体质量指数≥24 kg/m 2(OR=2.041,95%CI:1.038~4.013)、术后恢复时间≤6个月(OR=2.456,95%CI:1.339~4.505)、肿瘤距肛缘距离≤7 cm(OR=2.735,95%CI:1.480~5.055)、新辅助治疗(OR=3.772,95%CI:1.109~12.832)、吻合口瘘(OR=5.537,95%CI:1.103~27.791)是直肠癌保肛根治术后LARS发生的独立危险因素。利用上述危险因素构建预测术后LARS发生风险的列线图模型,模型的ROC曲线下面积为0.754(95%CI:0.689~0.819),Bootstrap法的内部验证结果显示该模型的C-index值为0.750,且校正曲线与理想曲线拟合良好。结论基于5项危险因素所构建的模型对直肠癌保肛根治术后LARS的发生概率具有较好的预测能力,有助于早期识别高风险人群,制定临床干预措施。Objective To investigate the risk factors of low anterior resection syndrome after anus-preserving radical resection for rectal cancer and establish a predictive nomogram model.Methods This was a retrospective case-control study.Patients who had undergone anus-preserving radical resection for rectal cancer at department of general surgery of the first affiliated hospital of Anhui medical university completed a LARS score scale.Then the nomogram model was established according to the risk factors of LARS which were assessed by univariate and multivariate analyses.Results Body mass index(BMI)≥24 kg/m 2(OR=2.041,95%CI:1.038-4.013),recovery time≤6 months(OR=2.456,95%CI:1.339-4.505),the distance from tumor to anus≤7 cm(OR=2.735,95%CI:1.480-5.055),neoadjuvant therapy(OR=3.772,95%CI:1.109-12.832),anastomotic leak(OR=5.537,95%CI:1.103-27.791)were independent risk factors of LARS.Based on the 5 selected risk factors,a nomogram model was established to predict the risk factors of LARS after anus-preserving radical resection for rectal cancer.The area under ROC curve of the nomogram model was 0.754(95%CI:0.689-0.819).After internal verification by Bootstrap self-sampling method,the C-index value of the model was 0.750 and the calibration curve fitted well with the ideal curve.Conclusion The nomogram model based on the above risk factors can better predict the probability of LARS after anus-preserving radical resection for rectal cancer,which is helpful for early identification of hign-risk population and development of clinical intervention measures.

关 键 词：直肠癌 低位前切除综合征 危险因素 列线图 预测模型 

分 类 号：R735.37[医药卫生—肿瘤]