GLIM在胃癌病人营养不良诊断和预后评估中的价值研究及生存预测列线图模型的构建

Evaluation of the Global Leadership Initiative on Malnutrition Criteria(GLIM)for diagnosing malnutrition and predicting survival of patients with gastric cancer

目的:评估全球(营养)领导人营养不良诊断标准共识(GLIM)诊断胃癌病人营养不良的准确性以及其对胃癌病人术后远期生存的预后价值,并建立包含GLIM共识的胃癌病人术后的生存预测模型。方法:采用病例对照研究方法,回顾性收集因胃癌行胃切除术病人的临床病理信息和生存随访资料,采用GLIM和患者主观整体评估(PG-SGA)进行营养评定,采用Cohen’s kappa系数分析比较两种诊断方法的一致性。用Kaplan-Meier生存曲线分析GLIM营养不良病人的累积生存率和中位生存时间,并进行Log Rank检验统计分析。采用LASSO和Cox多因素分析筛选出的独立预后因素建立列线图模型。采用一致性指数(C-index)、受试者工作特征曲线(ROC)、校准曲线以及临床决策曲线(DCA)从区分度、校准度和临床应用3个方面验证该模型的性能。结果:根据GLIM营养不良诊断标准,31.4%(312例)的胃癌病人合并有营养不良,其中10.6%(105例)的病人为重度营养不良(2期),20.8%(207例)的病人为中度营养不良(1期)。GLIM和PG-SGA诊断胃癌合并营养不良具有较强的一致性(K=0.691,P<0.001)。胃癌合并营养不良术后生存时间(OS)较营养正常组短(营养不良组中位OS为45.2月,营养正常组中位OS未达到,P=0.0025)。多因素分析表明TNM分期晚(II和III)、年龄≥70岁和GLIM营养不良是胃癌切除术后预后的独立危险因素(P<0.05)。相反,根治性切除是胃癌术后预后的独立保护因素(P<0.001)。列线图生存预测模型对胃癌术后5年生存率具有良好的预测价值。结论:GLIM标准不仅可以准确诊断胃癌病人营养不良及其程度,而且可用于预测胃癌术后生存情况。包含GLIM标准的胃癌术后生存预测列线图模型具有良好的预测性能和临床应用价值。

Objective:To verify the validity of the Global Leadership Initiative on Malnutrition Criteria(GLIM) for identifying malnutrition,investigate the prognostic value of GLIM-defined malnutrition,and create a nomogram for survival prediction in patients who underwent gastrectomy for gastric cancer(GC).Methods:A case-control study was conducted to retrospectively analyze the clinicopathological and survival data of 994 consecutive patients who underwent gastrectomy for GC.Nutrition assessment was performed on all patients according to the GLIM criteria and PG-SGA.Kappa statistic was used to evaluate the agreement between two methods.Kaplan-Meier curves were used to analyze the relationship between GLIM-defined malnutrition and the OS of patients with GC.LASSO and multivariate Cox regressions were performed to analyze independent prognostic predictors associated with survival.These factors were incorporated to develop a nomogram for individualized survival prediction.Harrell’s concordance index(C-index),receiver operating characteristic(ROC)curve,calibration curve and decision curve analysis(DCA)were used to verify the performance of the model.Results:Based on the GLIM criteria,31.4%of the patients were diagnosed as malnourished,including 20.8%moderate malnutrition and 10.6%severe malnutrition.GLIM criteria had a better agreement with PG-SGA(K=0.691,P<0.001).According to the Kaplan-Meier curves,malnutrition(both moderate and severe)was associated with shorter OS compared with no malnutrition(median OS,45.2 months versus OS not reached;P=0.0025).Cox regression analysis showed that age≥70,TNM stage I,TNM stage II,moderate malnutrition and severe malnutrition were independent risk predictors of overall survival.In contrast,radical resection was independent protective factors for survival(P<0.001).The GLIM nomogram showed good performance in predicting 5 years survival of GC patients and the decision curve analysis demonstrated that the nomogram was clinically useful.Conclusion:The GLIM criteria could not only diagnose malnutrition but could also predict the overall survival of gastric cancer patients after gastrectomy.The nomogram developed based on the GLIM criteria and other three independent risk predictors can provide a more accurate prediction of the prognosis for survival of gastric patients.