直肠癌Miles术后早期造口并发症列线图预测模型的构建与验证  

Construction and validation of a nomogram model for early ostomy complications after Miles surgery for rectal cancer

作　　者：张毅[1] 刘焱[2] 柴亚欣 刘晓静 Zhang Yi;Liu Yan;Chai Yaxin;Liu Xiaojing(Third General Surgery(Anorectal),Xinxiang Central Hospital,Xinxiang 453000,China;Oncology Surgery(First General Oncology Department),Xinxiang Central Hospital,Xinxiang 453000,China;Magnetic Resonance Diagnosis Center,Xinxiang Central Hospital,Xinxiang 453000,China;Oncology Surgery(Third General Oncology Department),Xinxiang Central Hospital,Xinxiang 453000,China)

机构地区：[1]新乡市中心医院普通外科三(肛肠),新乡453000 [2]新乡市中心医院肿瘤外科(普瘤一),新乡453000 [3]新乡市中心医院磁共振诊断中心,新乡453000 [4]新乡市中心医院肿瘤外科(普瘤三),新乡453000

出　　处：《中华现代护理杂志》2023年第32期4417-4422,共6页Chinese Journal of Modern Nursing

基　　金：2021年河南省医学科技攻关计划联合共建项目。

摘　　要：目的分析直肠癌Miles术后发生早期造口并发症的危险因素,并构建列线图预测模型。方法采用病例对照研究方法收集2019年1月—2022年1月于新乡市中心医院行直肠癌Miles术的205例患者的临床资料,按照是否发生早期造口并发症分为早期造口并发症组及无早期造口并发症组。采用单因素分析和多因素Logistic回归分析确定直肠癌Miles术后早期造口并发症发生的影响因素,采用R语言构建直肠癌Miles术后发生早期造口并发症的列线图预测模型。通过受试者工作特征曲线(ROC)下面积评估列线图预测模型的预测效能。通过绘制校准图进行一致性检验,使用Hosmer-Lemeshow检验判断模型的拟合优度,P>0.05表明模型的拟合优度较好。利用R软件随机抽取30%的病例作为验证组进行内部验证。结果本研究共纳入患者205例,男性115例,女性90例,其中37例发生早期造口并发症,早期造口并发症发生率为18.0%。多因素Logistic回归分析结果显示,未行快速外科康复护理(OR=2.779,95%CI:1.144~6.750)、年龄≥65岁(OR=2.862,95%CI:1.186~6.906)、白蛋白<35 g/L(OR=3.414,95%CI:1.373~8.488)和未行术前造口位置定位(OR=4.846,95%CI:1.488~15.776)是直肠癌Miles术后发生早期造口并发症的危险因素(P<0.05)。根据此结果构建预测模型,预测模型的ROC曲线下面积为0.801(95%CI:0.713~0.868),经内部验证后,验证组的ROC曲线下面积为0.807(95%CI:0.641~0.973)。建模组(P=0.980)与验证组(P=0.981)校正曲线显示,预测结果与实际结果有较好的重合度。结论未行快速外科康复护理、年龄≥65岁、白蛋白<35 g/L和未行术前造口位置定位是直肠癌Miles术后发生早期造口并发症的危险因素。本研究基于Logistic回归分析结果构建的列线图预测模型,可以有效预测直肠癌Miles术后发生早期造口并发症的概率。对于高风险患者,要及早进行个体化干预,降低造口并发症发生率,提高患者生Objective To analyze the risk factors for early ostomy complications after Miles surgery for rectal cancer and construct a nomogram model.Methods A case-control study method was used to collect clinical data of 205 patients who underwent Miles surgery for rectal cancer at Xinxiang Central Hospital from January 2019 to January 2022.Patients were divided into early ostomy complications group and no early ostomy complications group based on whether early ostomy complications occurred.Single factor analysis and multivariate Logistic regression analysis were used to determine the influencing factors for early ostomy complications after Miles surgery for rectal cancer.R language was used to construct a nomogram model for early ostomy complications after Miles surgery for rectal cancer.The area under the receiver operating characteristic curve(ROC)was used to evaluate the predictive performance of the nomogram model.By drawing a calibration map for consistency testing,the Hosmer-Lemeshow test was used to determine the goodness of fit of the model.P>0.05 indicated a good fit of the model.R software was used to randomly select 30%of cases as validation groups for internal validation.Results This study included a total of 205 patients,including 115 males and 90 females.Among them,37 had early ostomy complications,with an incidence of 18.0%.The results of multivariate Logistic regression analysis showed that failure to receive rapid surgical rehabilitation nursing[OR=2.779,95%CI(1.144,6.750)],age≥65 years old[OR=2.862,95%CI(1.186,6.906)],albumin<35 g/L[OR=3.414,95%CI(1.373,8.488)],and failure to locate the preoperative ostomy position[OR=4.846,95%CI(1.488,15.776)]were risk factors for early ostomy complications after Miles surgery for rectal cancer,and the differences were statistically significant(P<0.05).Based on this result,a prediction model was constructed,with an area under the ROC curve of 0.801[95%CI(0.713,0.868)].After internal validation,the area under the ROC curve of the validation group was 0.807[95%CI(0.641,

关 键 词：直肠肿瘤 MILES术 早期造口并发症 列线图预测模型 

分 类 号：R473.73[医药卫生—护理学]