부산대학교 도서관

目的:评估术前与术后中性粒细胞-淋巴细胞比值(neutrophil-lymphocyte ratio,NLR)的组合对胃癌患者的预后预测价值.方法:回顾性收集2015年1月1日至2018年4月1日贵州医科大学附属医院胃肠外科进行根治性手术的初诊胃癌患者133例.收集患者术前NLR和术后NLR,使用受试者工作特征曲线(receiver operating characteristic curve,ROC)来分析确定炎症指标的最佳截断值,根据最佳截断值将所有患者分为术前NLR和术后NLR均低组、术前NLR低或术后NLR低组以及术前NLR和术后NLR均高组,并使用Kaplan-Meier法分析患者总生存期(overall survival,OS).采用Cox比例风险回归模型分析预后的独立危险因素.列线图模型由R Studio构建,并评估预测模型的效能.结果:术前NLR联合术后NLR对胃癌患者预后的预测效能(AUC=0.706 8)优于单独术前NLR(AUC=0.664 3)或术后NLR(AUC=0.593 3).术前NLR和术后NLR均低组的中位生存时间为61个月,5年生存率为62.07％;术前NLR低或术后NLR低组的中位生存时间为54个月,5年生存率为48.15％;术前NLR和术后NLR均高组的中位生存时间为12个月,5年生存率为22％.术前NLR和术后NLR均高组预后是最差的(P＜0.001).Cox回归分析显示手术方式(HR=0.465,P=0.001)、TNM分期(HR=3.387,P=0.006)及术前NLR联合术后NLR(HR=2.091,P=0.002)是胃癌患者预后的独立危险因素.结合独立危险因素构建列线图预测3年(AUC=0.859 3)和5年(AUC=0.885 2)预后,其预测效能优于单独术前NLR(AUC=0.664 3)和术后NLR(AUC=0.593 3).校准曲线提示该模型具有较高的一致性,决策曲线显示该模型具有良好的临床获益.结论:术前NLR联合术后NLR是胃癌患者的重要预后因素.术前NLR和术后NLR均高组的患者预后不良.基于这两项炎症指标和临床病理特征的列线图模型可被用于评估胃癌患者预后,其预测能力优于单独术前NLR和术后NLR.
Objective:To evaluate the prognostic predictive value of the combination of preoperative and postoper-ative neutrophil to lymphocyte ratio(NLR)in patients with gastric cancer.Methods:A total of 133 newly diagnosed gastric cancer patients who underwent radical surgery in the gastrointestinal surgery department,Affiliated Hospital of Guizhou Medical University from January 1,2015 to April 1,2018 were retrospectively collected.Preoperative NLR and postoperative NLR were collected,and the receiver operating characteristic curve(ROC)was used to analyze and determine the optimal cut-off value of inflammation indicators.According to the optimal cut-off value,all patients were divided into preoperative NLR and postoperative NLR low group,preoperative NLR low or postoperative NLR low group,and preoperative NLR and postoperative NLR high group.Kaplan-Meier method was used to analyze overall survival(OS).Independent risk factors for prognosis were analyzed by Cox proportional hazard regression model.The nomogram model was built by R Studio,and the efficacy of the prediction models was evaluated.Results:The pre-dictive efficacy of preoperative NLR combined with postoperative NLR for gastric cancer patients(AUC=0.706 8)was better than that of preoperative NLR alone(AUC=0.664 3)or postoperative NLR alone(AUC=0.593 3).The median survival time of patients with low preoperative NLR and postoperative NLR was 61 months,and the 5-year survival rate was 62.07％.The median survival time was 54 months and the 5-year survival rate was 48.15％in the preoperative NLR low or postoperative NLR low group.The median survival time was 12 months and the 5-year sur-vival rate was 22％in the high preoperative NLR and postoperative NLR group.The prognosis was the worst in the high preoperative NLR and postoperative NLR group(P＜0.001).Cox regression analysis showed that surgical meth-od(HR=0.465,P=0.001),TNM stage(HR=3.387,P=0.006)and preoperative NLR combined with postopera-tive NLR(HR=2.091,P=0.002)were independent risk factors for prognosis of gastric cancer patients.The predic-tion of 3-year prognosis(AUC=0.859 3)and 5-year prognosis(AUC=0.885 2)was better than that of preoper-ative NLR(AUC=0.664 3)and postoperative NLR(AUC=0.593 3).The calibration curve suggested that the model had a high consistency,and the decision curve showed a good clinical benefit from the model.Conclusion:Pre-operative NLR combined with postoperative NLR is an important prognostic factor in gastric cancer patients.Patients in high group of both preoperative NLR and postoperative NLR had a poor prognosis.The nomogram graph model based on these two inflammatory markers and clinicopathological features can be used to evaluate the prognosis of gas-tric cancer patients,with better predictive power than preoperative NLR and postoperative NLR alone.