摘要
目的采用竞争风险模型探索影响慢性阻塞性肺疾病(chronic obstructive pulmonary disease,COPD)营养不良的因素。方法回顾性分析梧州市红十字会医院呼吸与危重症医学科2021年1月至2022年12月收治的271例COPD患者的临床资料。通过竞争风险模型评估营养不良的影响因素和发生风险,同时与Kaplan-Meier法(K-M)和Cox回归模型进行比较。根据多因素分析结果绘制列线图预测患者营养不良发生率,根据受试者工作特征曲线下面积(AUC)评价模型预测能力。结果中位随访334天,期间87例(32.10%)发生营养不良。Cox回归和竞争风险单因素分析显示,总蛋白、前白蛋白、血红蛋白、球蛋白、白蛋白/球蛋白比值、红细胞计数、白细胞计数、血氯和血钾与营养不良的发生相关。多因素竞争风险分析显示,总蛋白<63.7g/L、前白蛋白<17.4mg/dL和白蛋白/球蛋白比值<1.5是COPD患者发生营养不良的高危因素。K-M计算营养不良累积发生率(cumulative incidence function,CIF)为33.92%,竞争风险分析得到的CIF为32.10%。竞争风险模型的AUC为85.3%。结论在充分考虑竞争风险的情况下,总蛋白、前白蛋白、白蛋白/球蛋白比值越低,COPD患者发生营养不良的风险越高。与K-M和Cox回归相比,竞争风险模型能够更准确地评估COPD患者发生营养不良的风险因素和发生率,同时具有良好的预测价值,以此为依据的列线图有助于临床医生评估风险高的患者并制定个体化干预措施。
Objective To explore the effect of malnutrition on chronic obstructive pulmonary disease(COPD)by competitive risk model.Methods The clinical data of 271 patients with COPD admitted to Department of Respiratory and Critical Care Medicine,Wuzhou Red Cross Hospital from January 2021 to December 2022 were retrospectively analyzed.The competitive risk model was used to assess the influencing factors and risk of malnutrition,and the results were compared with the Kaplan-Meier method(K-M)and Cox regression model.According to the results of multivariate analysis,a nomogram was used to predict the incidence of malnutrition,and the predictive ability of the model was evaluated according to the area under the receiver operating characteristic curve(AUC).Results During a median follow-up of 334 days,87 patients(32.10%)developed malnutrition.Cox regression and competitive risk univariate analysis showed that total protein,prealbumin,hemoglobin,globulin,albumin/globulin ratio,red blood cell count,white blood cell count blood chlorine and potassium were associated with malnutrition.Multivariate competitive risk analysis showed that total protein<63.7 g/L,prealbumin<17.4 mg/dL and albumin/globulin ratio<1.5 were high risk factors for malnutrition in COPD patients.The cumulative incidence function(CIF)calculated by K-M was 33.92%,and the CIF obtained by competitive risk analysis was 32.10%.The AUC of the competitive risk model was 85.3%.Conclusion With full consideration of the risk of competition,the lower the total protein,prealbumin and albumin/globulin ratio,the higher the risk of malnutrition in COPD patients.Compared with K-M and Cox regression,the competitive risk model can more accurately assess the risk factors and incidence of malnutrition in COPD patients,and has good predictive value.The nomogram based on this can help clinicians assess patients with high risk and develop individualized interventions.
作者
李意云
蒋樟健
杨杰华
郭敏
杨超华
LI Yiyun;JIANG Zhangjian;YANG Jiehua;GUO Min;YANG Chaohua(Department of Respiratory and Critical Care Medicine,Wuzhou Red Cross Hospital,Wuzhou,Guangxi 543002,China)
出处
《临床肺科杂志》
2024年第2期201-206,211,共7页
Journal of Clinical Pulmonary Medicine
基金
梧州市科技计划项目(No.202002124)。
关键词
慢性阻塞性肺疾病
营养不良
竞争风险模型
列线图
chronic obstructive pulmonary disease
malnutrition
competitive risk model
nomogram
