老年肺癌患者根治术后营养不良风险预测列线图模型的构建与验证

Construction and validation of a nomogram model for predicting malnutrition risk in older patients with lung cancer after radical surgery

中文关键词: 肺肿瘤    营养不良    影响因素    列线图    预测模型

英文关键词: lung neoplasmsmalnutritioninfluencing factornomogramsprediction model

中文摘要:

      目的 分析老年肺癌患者根治术后营养不良的危险因素，构建列线图预测模型。方法 回顾性选取2021年6月至2024年9月简阳市人民医院收治的290例老年肺癌患者，并将其按7∶3随机分为建模集（n=203）和验证集（n=87）。依据术后3个月是否发生营养不良将建模集患者分为营养不良组（n=64）和营养正常组（n=139）。采用最小绝对收缩和选择算子（LASSO）回归联合多因素Logistic回归筛选老年肺癌患者根治术后营养不良的独立影响因素，构建列线图预测模型并评估其性能。结果 290例老年肺癌患者，年龄61～75岁，平均年龄（66.74±6.10）岁，其中男性166例、女性124例，术后营养不良发生率为32.07%（93/290）。LASSO回归结合Logistic回归分析显示，TNM分期Ⅲ期、营养风险筛查2002（NRS2002）评分升高、中性粒细胞/淋巴细胞比值（NLR）升高及抑郁自评量表（SDS）评分升高均为老年肺癌患者术后营养不良的独立危险因素，预后营养指数（PNI）升高是术后营养不良的保护因素（P<0.05）。受试者操作特征曲线显示，建模集和验证集中列线图模型预测患者发生营养不良的曲线下面积分别为0.885（95%CI:0.827～0.941）、0.823（95%CI:0.765～0.873）。Hosmer-Lemeshow检验结果显示，模型拟合度良好（建模集χ²=2.425，验证集χ²=1.863，均P> 0.05）。结论 TNM分期、NLR、PNI、NRS2002评分、SDS评分均为老年肺癌患者根治术后营养不良的相关影响因素，基于此构建的列线图模型预测效能良好。

英文摘要:

      Objective To analyze the risk factors for postoperative malnutrition in older patients with lung cancer after radical surgery and construct a nomogram prediction model.Methods A total of 290 older patients with lung cancer admitted to People's Hospital of Jianyang City from June,2021 to September,2024 were retrospectively selected and randomly divided into modeling set(n=203)and validation set(n=87)according to a ratio of 7∶3.Based on whether malnutrition occurred within 3 months after the surgery,the patients in the modeling set were divided into malnutrition group(n=64)and normal nutrition group(n=139).Least absolute shrinkage and selection operator(LASSO)regression combined with multivariate Logistic regression were used to screen the independent influencing factors for malnutrition in older patients with lung cancer after radical resection.A nomogram prediction model was constructed and its performance was evaluated.Results The ages of 290 older patients with lung cancer ranged from 61 to 75 years old,with an average age of(66.74±6.10)years.Among them,166 were male and 124 were female.The incidence of postoperative malnutrition was 32.07%(93/290).LASSO regression combined with Logistic regression analysis revealed that TNM stage Ⅲ,elevated nutrition risk screening 2002(NRS2002)score,elevated neutrophil/lymphocyte ratio(NLR),and elevated self-rating depression scale(SDS)score were all independent risk factors for postoperative malnutrition in older adults with lung cancer.An elevated prognostic nutritional index(PNI)was a protective factor for postoperative malnutrition(P＜0.05).The receiver operator characteristic curve showed that the area under the curve for the nomogram model in predicting the occurrence of malnutrition in the modeling set and the validation set was 0.885(95%CI:0.827-0.941)and 0.823(95%CI:0.765-0.873),respectively.Hosmer-Lemeshow test results showed that the model fitted well(modeling setx2=2.425,validation setx2=1.863,both P＞0.05).Conclusion TNM stage,NLR,PNI,NRS2002 score and SDS score are related factors for malnutrition in older adults with lung cancer after radical surgery,and the nomogram model based on these factors has a good predictive performance.