结肠癌术后复发转移的影响因素及列线图预测模型的建立  

Influencing factors of postoperative recurrence and and establishment of Nomogram prediction model

作　　者：仝传志[1] 马瑞斌[2] 程坤[3] TONG Chuan-zhi;MA Rui-bin;CHENG Kun(Department of General Sursery,Xinjiang Medical University Affiliated Traditional Chinese Medicine Hospital,Wulumuqi Xinjiang 830000,China;Department of General Sursery,The Xinjiang Armed Police Corps Hospital,Wulumuqi Xinjiang 830099,China;Department of General Sursery,The First Affiliated Hospital of Xinjiang Medical University,Wulumuqi Xinjiang 830011,China)

机构地区：[1]新疆医科大学附属中医医院普通外科,新疆乌鲁木齐830000 [2]武警新疆总队医院普外科,新疆乌鲁木齐830099 [3]新疆医科大学第一附属医院普外科,新疆乌鲁木齐830011

出　　处：《临床和实验医学杂志》2024年第24期2612-2616,共5页Journal of Clinical and Experimental Medicine

基　　金：新疆维吾尔自治区自然科学基金课题(编号:2021D01C240)。

摘　　要：目的分析结肠癌术后复发转移的影响因素,并构建列线图预测模型。方法回顾性选取2016年1月至2021年12月新疆医科大学附属中医医院和新疆医科大学第一附属医院普通外科收治的结肠癌患者为研究对象。根据术后随访期间是否发生复发转移,分为复发转移组(n=63)和非复发转移组(n=204)。收集所有患者的临床资料[性别、年龄、肿瘤部位、组织分化程度、病理类型、肿瘤直径、肿瘤肠壁侵袭程度、淋巴结转移、TNM分期、肿瘤数目、浸润深度、术后癌胚抗原(CEA)、术后接受化疗、术后p53蛋白和术后癌抗原(CA)19-9]。将单因素分析变量纳入多因素Logistic回归分析风险预测模型,通过R Studio4.2.2软件构建结肠癌术后复发转移列线图预测模型,使用受试者操作特征(ROC)曲线下面积评估预测模型的区分度,采用Hosmer-Lemeshow拟合优度检验P值评估预测模型的校准度。结果两组患者的性别构成比、年龄、肿瘤部位、病理类型、肿瘤直径、肿瘤肠壁侵袭程度、肿瘤数目、浸润深度、术后接受化疗、术后p53蛋白比较,差异均无统计学意义(P>0.05);两组患者的组织分化程度、淋巴结转移、TNM分期、术后CEA、术后CA19-9比较,差异均有统计学意义(P<0.05)。多因素Logistic回归分析结果显示,组织分化程度(低分化)、淋巴结转移、TNM分期(Ⅲ~Ⅳ期)、术后CEA(≥5.0μg/L)、术后CA19-9(≥37 U/mL)是影响结肠癌患者发生术后复发转移的独立危险因素(P<0.05)。ROC曲线结果显示,曲线下面积为0.913(95%CI:0.896~0.979),提示结肠癌术后复发转移列线图预测模型具有良好的精准度和区分度;通过Hosmer-Lemeshow拟合优度检验进行评估,最大偏移量为0.096,最小偏移量为0.028(P=0.429),模型校正曲线趋近于对角线标准曲线,预测值同实测值基本一致。结论术后CA19-9、组织分化程度、术后CEA、淋巴结转移、TNM分期是术后复发转移的独�Objective To analyze the influencing factors of postoperative recurrence and establish a Nomogram prediction model.Methods The patients with colon cancer admitted to the Department of General Surgery of Xinjiang Medical University Affiliated Traditional Chinese Medicine Hospital and the First Affiliated Hospital of Xinjiang Medical University from January 2016 to December 2021 were retrospectively selected as the research objects.According to whether recurrence and metastasis occurred during postoperative follow-up,the clinical data of all patients were respectively in the recurrence and metastasis group(n=63)and the non-recurrence and metastasis group(n=204).The clinical data[gender,age,tumor location,degree of tissue differentiation,pathological type,tumor diameter,degree of tumor intestinal wall invasion,lymph node metastasis,TNM stage,tumor number,depth of invasion,postoperative carcinoembryonic antigen(CEA),postoperative chemotherapy,postoperative p53 protein and postoperative cancer antigen(CA)19-9]of all patients were collected.Statistically significant variables in the univariate analysis were incorporated into multivariate Logistic regression analysis and a risk prediction model was constructed.R Studio4.2.2 software was used to construct a Nomogram prediction model for postoperative recurrence and metastasis of colon cancer.The discrimination of the prediction was evaluated using the receiver operating characteristic(ROC)curve area under curve,and the Hosmer-Lemeshow goodness of fit P-value was used to evaluate the calibration.Results There were no statistically significant differences in gender composition ratio,age,tumor location,pathological type,tumor diameter,tumor intestinal wall invasion degree,tumor number,invasion depth,postoperative chemotherapy and postoperative p53 protein between the two groups(P>0.05).There were statistically significant differences in the degree of tissue differentiation,lymph node metastasis,TNM stage,postoperative CEA and postoperative CA19-9 between the two groups(P<0.

关 键 词：结肠癌手术 复发转移 影响因素 列线图预测模型 

分 类 号：R73[医药卫生—肿瘤]