2026 年第 10 期 第 21 卷

基于中性粒细胞与淋巴细胞比值和衰弱表型联合多维度危险因素构建老年衰弱前期人群向衰弱进展的预测模型

Construction of a prediction model for the progression from pre-frailty to frailty in the elderly based on neutrophil to lymphocyte ratio and frailty phenotype combined with multidimensional risk factors

作者：黎慧姜萍曹媛单梓梅

英文作者：Li Hui Jiang Ping Cao Yuan Shan Zimei

单位：新疆维吾尔自治区人民医院医疗保健中心，乌鲁木齐830000

英文单位：Health Care Center People′s Hospital of Xinjiang Uygur Autonomous Region Urumqi 830000 China

关键词：衰弱；衰弱前期；老年综合征；衰弱表型；中性粒细胞与淋巴细胞比值

英文关键词：Frailty;Pre-frailty;Geriatricsyndrome;Frailtyphenotype;Neutrophiltolymphocyteratio

• 摘要：

• 目的  基于中性粒细胞与淋巴细胞比值（NLR）和衰弱表型联合多维度危险因素构建老年衰弱前期人群向衰弱进展的预测模型，为临床早期筛查高风险人群提供科学依据。方法 采用前瞻性队列设计。选取2022年1月至2023年11月于新疆维吾尔自治区人民医院干部保健病区筛查确诊的692例老年衰弱前期患者作为研究对象。收集研究对象基线资料，计算NLR，采用Fried衰弱表型量表评估衰弱与否。所有研究对象进行为期24个月随访，以进展为衰弱作为结局事件，根据是否发生结局事件将研究对象分为进展组与非进展组。采用多因素Logistic回归方法筛选老年衰弱前期患者向衰弱进展的危险因素，基于危险因素构建列线图预测模型，采用Bootstrap法（1 000次重抽样）进行内部验证，以受试者工作特征曲线下面积评估模型区分度，通过校准曲线及Hosmer-Lemeshow检验评估校准度，决策曲线分析验证临床实用性。结果 2组患者年龄、性别、基础病种类≥2种、服药种类≥2种、从不运动、营养不良、睡眠障碍、抑郁比例比较，差异均有统计学意义（均P＜0.05）。进展组NLR［（3.2±0.6）比（2.1±0.5）］、衰弱表型≥3分比例［68.0%（87/128）比36.2%（204/564）］均高于未进展组（t=22.062，P＜0.001； χ2=21.588，P＜0.001）。多因素Logistic回归分析表明NLR、衰弱表型≥3分、基础病种类≥2种、从不运动、抑郁均是老年衰弱前期向衰弱进展的危险因素（均P＜0.05）。基于回归构建的列线图预测模型具有良好的预测效能、校准度和临床实用性。结论 基于NLR和衰弱表型联合多维度危险因素构建的列线图模型能有效预测老年衰弱前期向衰弱进展的风险，有助于早期识别高危人群并进行针对性干预。

• Objective To construct a prediction model for the progression from pre-frailty to frailty in the elderly based on neutrophil to lymphocyte ratio (NLR) and frailty phenotype combined with multidimensional risk factors, and to provide scientific evidence for early clinical screening of high-risk populations. Methods A prospective cohort design was adopted. A total of 692 elderly patients diagnosed with pre-frailty through screening in the Cadre Health Care Ward, People′s Hospital of Xinjiang Uygur Autonomous Region from January 2022 to November 2023 were enrolled as research subjects. Baseline data were collected and NLR was calculated. The Fried frailty phenotype scale was used to assess frailty status. All subjects were followed up for 24 months, with progression to frailty taken as the endpoint event. Participants were divided into progression group and non-progression group according to the occurrence of the endpoint event. Multivariate Logistic regression was used to screen risk factors for the progression from pre-frailty to frailty in the elderly. A nomogram prediction model was established based on the screened risk factors. Internal validation was performed using the Bootstrap method with 1 000 resamples. The area under the receiver operating characteristic curve was used to evaluate the discrimination of the model. Calibration curve and Hosmer-Lemeshow test were adopted to assess calibration degree, and decision curve analysis was used to verify clinical applicability. Results There were statistically significant differences in age, gender, two or more underlying diseases, two or more types of medications, physical inactivity, malnutrition, sleep disorders and depression ratio between the two groups (all P＜0.05). The NLR level ［(3.2±0.6) vs (2.1±0.5)］ and the proportion of patients with frailty phenotype score ≥3 ［68.0%(87/128) vs 36.2%(204/564)］ in the progression group were higher than those in the non-progression group (t=22.062, P＜0.001; χ2=21.588, P＜0.001). Multivariate Logistic regression analysis showed that NLR, frailty phenotype score ≥3, two or more underlying diseases, physical inactivity and depression were independent risk factors for the progression from pre-frailty to frailty in the elderly (all P＜0.05). The nomogram prediction model constructed based on regression analysis exhibited good predictive efficiency, calibration degree and clinical practicability. Conclusion The nomogram model established on the basis of NLR, frailty phenotype and multidimensional risk factors can effectively predict the risk of progression from pre-frailty to frailty in the elderly, which is conducive to early identification of high-risk populations and targeted intervention.