术前Th1/Th2、TNF-α、ALB与老年胃癌根治术后肺部感染关系及预测效能

Relationship of preoperative Th1/Th2 ratio,TNF-α,and ALB with pulmonary infection in elderly patients after radical surgery for gastric cancer and predictive efficacy of a nomogram based on these factors

背景老年人群胃癌根治术术后易发生肺部感染,机体免疫系统、炎症反应发挥重要作用.辅助性T细胞1(helper T cells 1,Th1)/辅助性T细胞2(helper T cells 2,Th2)可反映免疫动态平衡状态.术前Th1/Th2及炎症反应相关因子可能对其术后肺部感染有预测价值.目的探讨术前Th1/Th2、肿瘤坏死因子-α(tumor necrosis factor-α,TNF-α)、白蛋白(albumin,ALB)与老年胃癌根治术后肺部感染关系及预测效能.方法选取2020-04/2022-06我院收治的135例老年胃癌根治术后肺部感染患者(感染组)及135例未感染患者(对照组),比较两组一般人口学资料、手术相关情况、合并疾病及术前血清Th1/Th2、TNF-α、ALB水平,采用二元Logistic回归分析老年胃癌根治术后肺部感染的相关影响因素,运用R语言绘制预测老年胃癌根治术后肺部感染的列线图模型,获取列线图的预测风险能力指数(concordance index,C-index)评价其预测能力,采用Bootstrap法绘制预测曲线、校准曲线、理想曲线评价列线图与实际观测结果的一致性,绘制决策曲线分析(decision curve analysis,DCA)评价列线图模型的临床效用.结果感染组手术时间、术后留置胃管时间长于对照组,术中出血量高于对照组,糖尿病患者多于对照组(P<0.05);感染组术前Th1/Th2、ALB低于对照组,TNF-α高于对照组(P<0.05);二元Logistic回归分析显示,手术时间、术中出血量、糖尿病、术后留置胃管时间、TNF-α均是老年胃癌根治术后肺部感染的相关危险因素,Th1/Th2、ALB均是老年胃癌根治术后肺部感染的相关保护因素(P<0.05);构建预测术后肺部感染的列线图模型,手术时间、术中出血量、术后留置胃管时间、TNF-α越大,预测风险越大,Th1/Th2、ALB越大,预测风险越小,其预测风险能力指数(C-index)达0.985,处于较高水平;外部验证显示,该列线图校准度为0.826,模型与实际观测结果有较好的一致性;绘制DCA曲线显示,该列线图模型具有明显的正向净收益,在预测术后肺部感染方面拥有良好的临床效用.结论术前Th1/Th2、TNF-α、ALB均与老年胃癌根治术后肺部感染有关,基于基线相关资料与以上三指标所构建的模型具有较好的预测能力和临床效用,可作为预测术后肺部感染的一个潜在方案,指导临床进行围手术期管理.

BACKGROUND The immune system and the inflammatory response play an important role in the development of lung infections in the elderly population after radical gastric cancer surgery.Helper T cell 1(Th1)/helper T cell 2(Th2)phenotype reflects the dynamic immune homeostasis.Preoperative Th1/Th2 ratio and inflammatory response-related factors may have predictive value for postoperative lung infection.AIM To investigate the relationship of Th1/Th2 ratio,tumor necrosis factor-α(TNF-α),and albumin(ALB)with pulmonary infection in elderly patients after radical gastrectomy for gastric cancer and their predictive efficacy for postoperative pulmonary infection.METHODS A total of 135 patients with lung infection after radical gastric cancer surgery(infection group)and 135 uninfected patients(control group)admitted to our hospital from April 2020 to June 2022 were included in this study.The general demographic data,surgery-related conditions,combined diseases,and preoperative serum Th1/Th2 ratio,TNF-α,and ALB levels were compared between the two groups.R was used to draw a nomogram for predicting lung infection after radical gastrectomy in elderly patients,and the concordance index(C-index)of the nomogram was obtained to evaluate its prediction ability.The bootstrap method was used to draw the prediction curve,calibration curve,and ideal curve to evaluate the consistency between the nomogram and the actual observation results,and decision curve analysis(DCA)was performed to evaluate the clinical efficacy of the nomogram.RESULTS The infection group had longer operative time and postoperative gastric tube indwelling time,higher intraoperative bleeding,and more diabetic patients than the control group(P<0.05).Preoperative Th1/Th2 ratio and ALB were lower and TNF-αwas higher in the infection group than in the control group(P<0.05).Binary logistic regression analysis showed that operative time,intraoperative bleeding,diabetes mellitus,postoperative gastric tube indwelling time,and TNF-αwere risk factors for pulmonary infection in elderly patients after radical surgery for gastric cancer,and Th1/Th2 ratio and ALB were protective factors(P<0.05).A nomogram for predicting postoperative pulmonary infection was developed,and its C-index reached 0.985,which was at a high level.External validation showed that the calibration degree of the nomogram was 0.826,and there was good agreement between the model and the actual observation.DCA showed that the nomogram had obvious positive net benefit and possessed good clinical utility in predicting postoperative infection.CONCLUSION Preoperative Th1/Th2 ratio,TNF-α,and ALB were associated with pulmonary infection in elderly patients after radical surgery for gastric cancer.The nomogram developed based on baseline correlation data with the above three indicators has good predictive ability and clinical utility,and can be used as a potential tool to predict postoperative infection and guide clinical management in the perioperative period.