心血管–肾脏–代谢综合征0至3期人群中累积TCBI暴露与新发卒中的关联
Association of cumulative TCBI exposure with incident stroke in cardiovascular kidney metabolic syndrome stages 0 to 3
一句话CKM 0–3期中老年人中,较高累积TCBI及高水平两时间点TCBI模式与新发卒中相关。
研究问题
数据与样本
研究流程
- 1确定研究人群:纳入4475名年龄为45岁及以上的CKM 0–3期CHARLS参与者。
- 2刻画纵向暴露:根据2个时间点的TCBI评估累积暴露,并用K-means聚类划分两时间点TCBI模式。
- 3分析卒中关联:采用多变量logistic回归评估CumTCBI及TCBI模式与新发卒中的关联。
- 4评估非线性:采用限制性立方样条建模,探索暴露与结局的非线性关系及变化点。
- 5检验稳健性:开展敏感性分析,检查关联方向是否一致。
关键方法与工具
- CHARLS
- 提供研究人群数据
- K-means
- 划分两时间点TCBI模式
- multivariable logistic regression
- 估计调整后的卒中关联
- restricted cubic spline
- 分析非线性关联
设计思路
主要发现
- 充分调整后,CumTCBI每增加1单位,新发卒中的优势升高12%(OR = 1.12, 95% CI 1.07–1.17)。
- 高水平两时间点模式组相较于低水平模式组,新发卒中的OR为1.66(95% CI 1.32–2.08)。
- 暴露与新发卒中的关联呈非线性,探索性变化点接近27.83。
- 敏感性分析中,关联方向保持一致。
可借鉴之处
- 分析重复测量的代谢指标时,可借鉴本文同时构建累积暴露与聚类模式的做法,从累积水平和纵向特征两个角度研究临床结局。
- 除回归模型中的总体关联外,可结合限制性立方样条检验非线性,避免只用单一线性效应概括暴露与结局的关系。
- 可借鉴本文将敏感性分析与主要模型结合的设计,检查关联方向的稳健性,同时将流行病学关联与个体风险分层的临床效用明确区分。
数据与代码
局限
- 观察性研究只能支持关联,不能确定TCBI暴露与卒中之间的因果关系。
- TCBI仅在2个时间点测量,对时间变化的刻画限于这2次评估。
- 研究结果尚不能确立TCBI用于个体风险分层的临床效用。
摘要(中文翻译)
卒中仍造成沉重的全球健康负担,尤其是在心血管–肾脏–代谢(CKM)综合征人群中。单次测量可能无法充分反映随时间累积的代谢暴露。本研究评估了CKM 0–3期人群中累积TCBI暴露及两时间点TCBI模式与新发卒中的关联。本研究分析了4475名年龄为45岁及以上的CHARLS参与者,在2个时间点评估TCBI,通过K-means聚类划分纵向模式,并采用多变量logistic回归结合限制性立方样条建模,以评估不同CKM分期人群中TCBI的累积水平及时间变化与卒中发生的关系。较高的累积TCBI(CumTCBI)及归属于高水平两时间点模式组均与更高的新发卒中优势相关。在充分调整的模型中,CumTCBI每增加1单位,新发卒中的优势升高12%(OR = 1.12, 95% CI 1.07–1.17)。高水平模式组相较于低水平模式组的相应OR为1.66(95% CI 1.32–2.08)。该关联呈非线性,探索性变化点接近27.83,且在敏感性分析中关联方向保持一致。对于被归类为CKM 0–3期的中老年人,较高CumTCBI及高水平两时间点TCBI模式与新发卒中相关。这些发现代表流行病学关联,并不能确立TCBI用于个体风险分层的临床效用。
英文原文摘要
Stroke continues to impose a substantial global health burden, particularly among individuals with cardiovascular–kidney–metabolic (CKM) syndrome. Single measurements may not fully reflect metabolic exposure over time. This investigation assessed the associations of cumulative TCBI exposure and two-time-point TCBI patterns with incident stroke among individuals with CKM stages 0–3. This study sought to evaluate how cumulative and temporal changes in the TCBI relate to stroke occurrence across CKM stages by analyzing 4475 CHARLS participants aged 45 years or older, assessing TCBI at two occasions, categorizing longitudinal pattern via K-means clustering, and employing multivariable logistic regression combined with restricted cubic spline modeling. Higher cumulative TCBI (CumTCBI) and membership in the high-level two-time-point pattern group were associated with greater odds of incident stroke. In the fully adjusted model, each 1-unit increase in CumTCBI was associated with 12% higher odds (OR = 1.12, 95% CI 1.07–1.17). The corresponding OR for the high-level versus low-level pattern group was 1.66 (95% CI 1.32–2.08). The association was nonlinear, with an exploratory change point near 27.83, and remained directionally consistent in sensitivity analyses. Higher CumTCBI and high-level two-time-point TCBI patterns were associated with incident stroke among middle-aged and older adults classified as CKM stages 0–3. These findings represent epidemiological associations and do not establish the clinical utility of TCBI for individual risk stratification.
摘要来自 Crossref。翻译与解读由模型根据摘要生成,可能有误,不构成诊疗建议;数据以原文为准,版权归原出版方所有。