摘要
目的探讨老年2型糖尿病(T2DM)合并子宫内膜癌(EC)患者根治术后下肢淋巴水肿(LLL)的现状及其影响因素,构建术后LLL的预警模型。方法选择河北北方学院附属第一医院2022年1月至2023年12月收治的200例老年T2DM合并EC患者为研究对象,按照7∶3比例将患者分为训练集(n=140)与验证集(n=60)。患者均进行EC根治术,术后随访6个月,统计LLL发生率。根据是否发生LLL,将训练集患者分为LLL组(29例)和非LLL(non-NLLL,NLLL)组(111例)。采用SPSS 26.0统计软件进行数据处理,根据数据类型分别采用t检验或χ^(2)检验进行组间比较。采用多因素logistic回归分析构建老年T2DM合并EC患者术后LLL的预测模型,使用Hosmer-Lemeshow验证该模型的拟合度。绘制受试者工作特征(ROC)曲线,经曲线下面积(AUC)评价该模型的预测能力,以灵敏度与特异度评价该模型的实际应用效能。结果200例患者EC根治术后6个月内LLL发生率为19.50%(39/200),训练集为20.71%(29/140),验证集为16.67%)(10/60),两组患者LLL发生率比较,差异无统计学意义(χ^(2)=0.438,P=0.508)。多因素logisitc回归分析提示,国际妇产科联盟(FIGO)分期Ⅲ期(OR=4.473,95%CI 1.617~12.370)、肿瘤组织低分化(OR=2.998,95%CI 1.805~4.981)、淋巴结转移(OR=2.408,95%CI 1.676~3.461)、开腹手术(OR=2.192,95%CI 1.574~3.053)及淋巴结清扫数目>20个(OR=5.663,95%CI 2.303~13.924)是老年T2DM合并EC患者术后并发LLL的独立危险因素。行前哨淋巴结活检(OR=0.476,95%CI 0.257~0.882)及保留旋髂淋巴结(OR=0.626,95%CI 0.499~0.784)是其保护因素。根据回归方程中各变量的回归系数(β)进行赋值,风险预测模型为:Y=1.498×FIGO分期Ⅲ期+1.098×低分化+0.879×淋巴结转移+0.785×开腹手术-0.743×前哨淋巴结活检-0.469×旋髂淋巴结保留+1.734×淋巴结清扫数目(>20个)-4.491。绘制训练集的ROC曲线,结果显示其AUC为0.803(95%CI 0.723~0.883),模型的灵敏度及特异度分别为0.862,0.676。绘制验证集的ROC曲线,其AUC为0.760(95%CI 0.586~0.934),模型的灵敏度及特异度分别为0.700,0.800。采用Hosmer-Lemeshow拟合优度检验对训练集进行评价,结果显示,χ^(2)=1.897,P=0.771;对验证集进行评价,结果显示,χ^(2)=2.741,P=0.469。结论老年T2DM合并EC患者术后6个月内LLL发生率较高,基于多因素logistic回归分析结果可知,LLL预测模型具有良好的应用价值,可为临床医师进行高危LLL人群筛查提供依据。
Objective To explore the status quo and influencing factors of postoperative lower limb lymphedema(LLL)in elderly patients with type 2 diabetes mellitus(T2DM)complicated with endometrial cancer(EC),and to construct an early warning model of postoperative LLL.Methods A total of 200 elderly T2DM patients with EC admitted in our hospital from January 2022 to December 2023 were enrolled and divided into a training set(n=140)and a validation set(n=60)in a 7∶3 ratio.All patients received radical resection of EC.During six months of follow-up after surgery,the incidence of LLL was observed.And then the patients in the training set were assigned into a LLL subgroup(29 cases)and a non-LLL(NLLL)group(111 cases).SPSS 26.0 was used for data processing,and t test or Chi-square test was performed for between-group comparison depending on data type.Multivariate logistic regression analysis was used to construct a prediction model of postoperative LLL in elderly T2DM patients with EC.Hosmer-Lemeshow test was used to verify the fitting of the model.Receiver operating characteristic(ROC)curve was drawn and the area under the curve(AUC)was utilized to evaluate the predictive ability of the model.The sensitivity and specificity were used to evaluate the practical application efficiency of the model.Results Among the 200 eligible patients,19.50%(39/200)experienced LLL within six months after surgery.The incidence of LLL was 20.71%(29/140)in the training set and 16.67%(10/60)in the validation set,but no statistical difference was seen between them(χ^(2)=0.438,P=0.508).Multivariate logistic regression analysis suggested that International Federation of Gynecology and Obstetrics(FIGO)stageⅢ(OR=4.473,95%CI 1.617-12.370),poorly differentiated tumor tissue(OR=2.998,95%CI 1.805-4.981),lymph node metastasis(OR=2.408,95%CI 1.676-3.461),laparotomy(OR=2.192,95%CI 1.574-3.053)and dissected lymph nodes>20(OR=5.663,95%CI 2.303-13.924)were independent risk factors,and sentinel lymph node biopsy(OR=0.476,95%CI 0.257-0.882)and retention of circumflex iliac lymph nodes(OR=0.626,95%CI 0.499-0.784)were protective factors for postoperative LLL in elderly T2DM patients with EC.According to the regression coefficient(β)of each variable in the regression equation,the risk prediction model was shown as Y=1.498×FIGO stageⅢ+1.098×low differentiation+0.879×lymph node metastasis+0.785×laparotomy-0.743×sentinel lymph node biopsy-0.469×retention of circumflex iliac lymph node+1.734×number of lymph node dissection(>20)-4.491.ROC curve analysis revealed that the model had an AUC value of 0.803(95%CI 0.723-0.883),a sensitivity of 0.862 and a specificity of 0.676 in the training set,and an AUC value of 0.760(95%CI 0.586-0.934),a sensitivity of 0.700 and a specificity of 0.800 in the validation set.Hosmer-Lemeshow goodness of fit test revealed that the training set showedχ^(2)=1.897 and P=0.771,and the validation setχ^(2)=2.741 and P=0.469.Conclusion The incidence of LLL in elderly T2DM patients with EC is quite high within six months after surgery.Multivariate logistic regression analysis indicates that our prediction model for LLL occurrence has good application value,and can provide a basis for clinicians to screen high-risk LLL population.
作者
梁鹏娟
孙丽佳
王凤英
Liang Pengjuan;Sun Lijia;Wang Fengying(Operation Room,First Affiliated Hospital of Hebei North University,Zhangjiakou 075000,Hebei Province,China;Department of Gynaecology,First Affiliated Hospital of Hebei North University,Zhangjiakou 075000,Hebei Province,China)
出处
《中华老年多器官疾病杂志》
2025年第11期811-816,共6页
Chinese Journal of Multiple Organ Diseases in the Elderly
基金
河北省医学科学研究课题计划项目(20231459)
张家口市重点研发计划(2221173D)。
关键词
老年人
子宫内膜癌
糖尿病
术后下肢淋巴结水肿
预测模型
aged
endometrial cancer
diabetes mellitus
postoperative lower limb lymphedema
prediction model
