作者：段菲1，闵春燕2，孙小卫2，肖江琴2

单位：

1.832000新疆维吾尔自治区石河子市，石河子大学医学院　2.830000新疆维吾尔自治区乌鲁木齐市，新疆维吾尔自治区人民医院护理部

单位（英文）：

1.School of Nursing, Shihezi University, Shihezi 832000, China2.Department of Nursing Administration, People's Hospital of Xinjiang Uygur Autonomous Region, Urumqi 830000, China

关键词：

慢性阻塞性肺疾病；衰弱；影响因素分析；列线图

关键词（英文）：

Chronic obstructive pulmonary disease; Frailty; Root cause analysis; Nomogram

中图分类号：

R 563.9

DOI：

10.12114/j.issn.1008-5971.2022.00.241

基金项目：

新疆维吾尔自治区自然科学基金资助项目（2022D01C111）

摘要：

目的　探讨老年慢性阻塞性肺疾病（COPD）患者发生衰弱的影响因素，并构建其风险预测列线图模型。方法　采用便利抽样法，选取2021年11月至2022年4月在新疆维吾尔自治区人民医院呼吸内科住院的老年COPD患者339例作为研究对象。由经过统一培训的3名护理硕士研究生完成问卷调查，包括衰弱表型（FP）、一般资料调查表、匹兹堡睡眠质量指数量表（PSQI）、微型营养评定简表（MNA-SF）及简版老年抑郁量表（GDS-15）。本研究共发放问卷350份，回收有效问卷339份，有效回收率为96.9%。并记录患者肺功能GOLD分级、呼吸困难发生情况及实验室检查指标〔血红蛋白（Hb）、白介素6（IL-6）、肌钙蛋白T（TnT）、降钙素原（PCT）、白蛋白（Alb）及脑钠肽（BNP）〕。老年COPD患者发生衰弱的影响因素分析采用多因素Logistic回归分析，采用R 3.5.3软件中的rms程序包构建老年COPD患者发生衰弱的风险预测列线图模型，并绘制ROC曲线以评估该列线图模型对老年COPD患者发生衰弱的预测价值。结果　339例老年COPD患者中衰弱、衰弱前期、正常者分别为114例（33.6%）、117例（34.5%）、108例（31.9%）。有无衰弱患者年龄、文化程度、婚姻状况、病程、吸烟率、多重用药者占比、PSQI评分、MNA-SF评分、GDS-15评分、呼吸困难发生率、肺功能GOLD分级、Hb、IL-6、TnT、BNP比较，差异有统计学意义（P ＜0.05）。多因素Logistic回归分析结果显示，婚姻状况〔OR =5.941，95%CI （1.161，30.400）〕、多重用药〔OR =14.070，95%CI（2.162，91.561）〕、PSQI评分〔OR =1.364，95%CI （1.142，1.629）〕、MNA-SF评分〔OR =0.283，95%CI （0.166，0.481）〕、GDS-15评分〔OR =2.806，95%CI （1.919，4.103）〕、Hb〔OR =1.069，95%CI （1.021，1.119）〕是老年COPD患者发生衰弱的独立影响因素（P ＜0.05）。基于多因素Logistic回归分析筛选的影响因素构建老年COPD患者发生衰弱的风险预测列线图模型。ROC曲线分析结果显示，该列线图模型预测老年COPD患者发生衰弱的AUC为0.974〔95%CI （0.961，0.987）〕，最佳截断值为89.060分，其灵敏度为0.974、特异度为0.880。结论　老年COPD患者衰弱发生率为33.6%。婚姻状况、多重用药、PSQI评分、MNA-SF评分、GDS-15评分、Hb是老年COPD患者发生衰弱的独立影响因素，而基于上述影响因素构建的列线图模型对老年COPD患者发生衰弱具有较高的预测价值。

英文摘要：

Objective　To explore the influencing factors of frailty in elderly patients with chronic obstructive pulmonarydisease (COPD) , and to construct a risk prediction nomograph model. Methods　Using the convenient sampling method, 339elderly COPD patients hospitalized in the Respiratory Medicine of People's Hospital of Xinjiang Uygur Autonomous Region fromNovember 2021 to April 2022 were selected as the research objects. The questionnaire survey was completed by three masternursing students after unified training, including Frailty Phenotype (FP) , General Information Questionnaire, Pittsburgh SleepQuality Index (PSQI) , Short-Form Mini Nutritional Assessment (MNA-SF) and 15-Item Geriatric Depression Scale (GDS-15) .A total of 350 questionnaires were distributed in this study, and 339 valid questionnaires were recovered, the effective responserate was 96.9%. The pulmonary function GOLD grade, incidence of dyspnea and laboratory examination indexes [hemoglobin(Hb) , interleukin 6 (IL-6) , troponin T (TnT) , procalcitonin (PCT) , albumin (Alb) and brain natriuretic peptide (BNP) ]were recorded. The influencing factors of frailty in elderly COPD patients were analyzed by multivariate Logistic regressionanalysis, the nomograph model for predicting the risk of frailty in elderly COPD patients was constructed by the rms package in R3.5.3 software, and the ROC curve was drawn to evaluate the predictive value of the nomograph model for frailty in elderly patientswith COPD. Results　Among the 339 elderly patients with COPD, 114 cases (33.6%) , 117 cases (34.5%) and 108 cases (31.9%)were frailty, pre-frailty and normal, respectively. There was significant difference in age, educational level, marital status, courseof disease, smoking rate, proportion of multiple drug users, PSQI score, MNA-SF score, GDS-15 score, incidence of dyspnea,pulmonary function GOLD grade, Hb, IL-6, TnT, BNP in patients with and without frailty (P < 0.05) . Multivariate Logisticregression analysis showed that marital status [OR =5.941, 95%CI (1.161, 30.400) ] , multiple drugs use [OR =14.070, 95%CI(2.162, 91.561) ] , PSQI score [OR =1.364, 95%CI (1.142, 1.629) ] , MNA-SF score [OR =0.283, 95%CI (0.166, 0.481) ] , GDS-15score [OR =2.806, 95%CI (1.919, 4.103) ] , Hb [OR =1.069, 95%CI (1.021, 1.119) ] were independent influencing factors of frailtyin elderly patients with COPD (P < 0.05) . Based on the influencing factors screened by multivariate Logistic regression analysis,a nomograph model for predicting the risk of frailty in elderly patients with COPD was constructed. The ROC curve analysis resultsshowed that the AUC of the nomogram model for predicting frailty in elderly patients with COPD was 0.974 [95%CI (0.961, 0.987) ] ,and the best cut-off value was 89.060 points, the sensitivity was 0.974, and the specificity was 0.880. Conclusion　The incidenceof frailty in elderly patients with COPD is 33.6%. Marital status, multiple drugs use, PSQI score, MNA-SF score, GDS-15 scoreand Hb are independent influencing factors of frailty in elderly patients with COPD, and the nomogram model based on the aboveinfluencing factors has high predictive value for frailty in elderly patients with COPD.

参考文献：