直肠癌患者保肛术1年后发生重度低位前切除综合征列线图预测模型的构建

Establishment of Nomogram Model to Predict Major Low Anterior Resection Syndrome in Patients with Rectal Cancer One Year after Sphincter-preserving Surgery

目的探讨直肠癌患者保肛术1年后发生重度低位前切除综合征的危险因素,并构建预测模型。方法回顾性收集安徽省某三级甲等医院接受直肠癌保肛术1年后的患者,采用一般资料调查表、低位前切除综合征评分进行调查。通过二分类Logistic回归模型分析相关危险因素,采用R语言绘制列线图,构建风险预测模型。结果本研究共纳入319例患者,其中57例(17.9%)直肠癌患者在保肛术1年以后仍存在重度低位前切除综合征。Logistic回归分析结果显示,肿瘤距肛缘的距离、预防性造口、吻合口漏、术前放疗、术后放疗是直肠癌患者保肛术1年后重度低位前切除综合征发生的独立危险因素。基于回归分析结果构建列线图预测模型,C-index为0.849(95%CI:0.790~0.907),ROC曲线下面积(AUC)为0.843(95%CI:0.784~0.901),均显示该模型区分度较好,Brier得分和校正曲线均显示校准度较好,H-L检验(χ^(2)=9.313,P=0.231)表明模型拟合度较好。结论本研究构建的预测模型具有较好的预测效果,可直观、简便地甄别保肛术1年后仍存在重度低位前切除综合征的高危患者,为早期筛查和干预提供参考。

Objective To investigate the risk factors of major low anterior resection syndrome(LARS)in patients with rectal cancer one year after sphincter-preserving surgery,and to construct a predictive model.Methods The patients who underwent sphincter-preserving surgery for rectal cancer in a tertiary grade-A hospital in Anhui Province were recruited retrospectively,and they were surveyed with the general information questionnaire and Low Anterior Resection Syndrome Score.The binary logistic regression model was used to analyze the risk factors,and R language to draw the nomogram to construct a risk predictive model.Results A total of 319 patients were enrolled in this study,of which 57(17.9%)patients with rectal cancer still had major LARS one year after sphincter-preserving surgery.Logistic regression analysis showed that the distance from the tumor to the anal margin,prophylactic stoma,anastomotic leakage,preoperative radiotherapy and postoperative radiotherapy were independent risk factors for major LARS in patients with rectal cancer one year after sphincter-preserving surgery.The nomogram prediction model was constructed based on the logistic regression analysis.The C-index was 0.849(95%CI:0.790~0.907);the area under the ROC curve(AUC)0.843(95%CI:0.784~0.901),which showed that the model had good discrimination.The Brier score and calibration curve showed good calibration,and H-L test(χ^(2)=9.313,P=0.231)showed the model had good fitting.Conclusion The predictive model constructed in this study has a good prediction effect,which can intuitively identify high-risk patients with major LARS one year after sphincter-preserving surgery,and provide reference for early screening and intervention.