Yan Zhaoqi, Pu Xiangyi, Cai Yongyuan, Chang Xing, Liu Zhiming, Liu Ruxiu
Guang'anmen Hospital of China Academy of Chinese Medical Sciences, Beijing, China.
Biogerontology. 2025 Jan 20;26(1):40. doi: 10.1007/s10522-025-10185-3.
The American Heart Association's (AHA) Life's Essential 8 (LE8) metrics provide a framework for assessing cardiovascular health (CVH). This study evaluates the relationship between CVH levels from LE8 and mortality risk, considering biological aging's role. Using data from the NHANES non-CVD adult population, CVH scores were categorized as low (< 50), moderate (50-79), and high (≥ 80) per AHA guidelines. Cox regression model assessed the impact of CVH levels on all-cause and cardiovascular mortality, while examining four aging indicators as mediators. RCS explored the relationships between CVH scores and mortality risk. The model's performance was evaluated using nine machine learning algorithms, with SHAP analysis on the best model to determine CVH score components' importance. Cox regression showed that all-cause mortality rates decreased by 35% for moderate and 54% for high CVH groups compared to low CVH. The high CVH group had a 59% lower cardiovascular mortality rate. Each unit increase in CVH score reduced all-cause and cardiovascular mortality to 0.98 times. RCS analysis revealed a nonlinear trend between CVH scores and mortality risk. Biological aging indicators significantly mediated the CVH-mortality relationship, with PhenoAge (21.57%) and KDM-Age (20.33%) showing the largest effects. The XGBoost model outperformed others, with SHAP analysis ranking CVH components: physical activity, nicotine, blood pressure, BMI, lipids, healthy eating index, blood glucose, and sleep. Higher CVH levels correlate with reduced all-cause and cardiovascular mortality risk, with biological aging mediating these effects. Adhering to AHA's LE8 metrics is recommended to enhance life expectancy in the non-CVD population.
美国心脏协会（AHA）的生命基本八大指标（LE8）为评估心血管健康（CVH）提供了一个框架。本研究在考虑生物衰老作用的情况下，评估了LE8所衡量的CVH水平与死亡风险之间的关系。利用美国国家健康与营养检查调查（NHANES）非心血管疾病成年人群的数据，根据AHA指南，CVH得分被分为低（<50）、中（50 - 79）和高（≥80）三类。Cox回归模型评估了CVH水平对全因死亡率和心血管死亡率的影响，同时将四个衰老指标作为中介因素进行考察。限制立方样条（RCS）分析探讨了CVH得分与死亡风险之间的关系。使用九种机器学习算法评估了模型的性能，并对最佳模型进行SHAP分析，以确定CVH得分各组成部分的重要性。Cox回归显示，与低CVH组相比，中CVH组的全因死亡率降低了35%，高CVH组降低了54%。高CVH组的心血管死亡率降低了59%。CVH得分每增加一个单位，全因死亡率和心血管死亡率降低至0.98倍。RCS分析揭示了CVH得分与死亡风险之间的非线性趋势。生物衰老指标显著介导了CVH与死亡率之间的关系，其中表观遗传衰老时钟（PhenoAge，21.57%）和端粒长度相关衰老时钟（KDM - Age，20.33%）的中介作用最大。XGBoost模型的表现优于其他模型，SHAP分析对CVH各组成部分的排名为：身体活动、尼古丁、血压、体重指数（BMI）、血脂、健康饮食指数、血糖和睡眠。较高的CVH水平与降低全因死亡率和心血管死亡率风险相关，生物衰老介导了这些影响。建议遵循AHA的LE8指标，以提高非心血管疾病人群的预期寿命。
