背景和目的

高血压作为心血管疾病（Cardiovascular disease，CVD）发病与过早死亡的关键危险因素，其患病率在全球和我国范围内均持续攀升，并逐渐呈现年轻化趋势。相比于老年高血压患者，年轻患者的知晓率、治疗率和控制率均偏低，并且未来可能受到不良血压水平的长期累积影响，易产生较高的CVD风险和严重器官损害，给个人和社会带来沉重的医疗负担。

高血压发病是遗传与环境因素共同作用的结果，对于年轻个体，遗传因素可能是诱发高血压的重要原因。尽管已发现一些与早发高血压相关的遗传变异位点，但不同研究人群中的结果并不一致。目前，基于欧美人群构建的血压多基因遗传风险评分（Polygenic risk score，PRS）对我国人群血压水平的预测能力欠佳，遗传因素对我国人群高血压发病年龄的影响尚未得到量化。除遗传因素外，可改变的危险因素也在高血压防控中发挥关键作用。2022年，美国心脏协会（American Heart Association，AHA）提出了“生命8要素（Life’s Essential 8，LE8）”的定义，包括8项健康行为和代谢因素指标，用于指导人群心血管健康（Cardiovascular health，CVH）的管理。既往研究提示遗传风险和生活方式对CVD风险可能存在交互作用，但这种交互作用在高血压发病中的表现仍不明确，特别是在早发和晚发高血压间是否存在差异尚需进一步探索。另一方面，冠心病（Coronary heart disease，CHD）和脑卒中作为CVD的主要亚型，其疾病负担正持续增加。高血压被认为是CHD和脑卒中的重要危险因素，但高血压发病年龄对CHD和脑卒中风险的影响存在争议，同时良好的CVH可以在多大程度上降低不同发病年龄高血压人群的CHD和脑卒中风险尚不清楚，需要进一步评估。

基于以上背景，本研究利用中国前瞻性队列构建血压PRS，量化血压遗传风险与高血压发病年龄、早发和晚发高血压风险的关联，评估CVH与血压遗传风险的联合及交互作用。随后，进一步探究不同高血压发病年龄与CVD、CHD和脑卒中发病风险的关联，并探讨良好的CVH在不同发病年龄的高血压人群中对CVD、CHD和脑卒中发病的保护作用，为早发高血压高危人群的识别和管理以及CVD的精准防控提供科学证据。

对象和方法

本研究的对象来源于中国动脉粥样硬化性心血管疾病风险预测项目（The Prediction for Atherosclerotic Cardiovascular Disease Risk in China，China-PAR）具有遗传信息的三个子队列，包括中国心血管健康多中心合作研究（International Collaborative Study of Cardiovascular Disease in Asia，InterASIA）、中国心血管病流行病学多中心协作研究（China Multi-Center Collaborative Study of Cardiovascular Epidemiology，ChinaMUCA）1998和中国代谢综合征社区干预研究暨中国家庭健康研究（Community Intervention of Metabolic Syndrome in China and Chinese Family Health Study，CIMIC）。ChinaMUCA 1998、InterASIA及CIMIC子队列分别于1998、2000-2001及2007-2008年进行基线调查，ChinaMUCA 1998和InterASIA也于2007-2008年进行了第一次随访，随后三个子队列一同于2012-2015年以及2018-2021年进行了两次随访。

研究通过问卷调查收集研究对象的人口学特征、生活方式、疾病史、用药史和家族史等信息，通过体格检查收集身高、体重和血压信息，并采集空腹外周血样本用于生化检测及基因分型。基于东亚人群大型全基因组关联研究（Genome-wide association study，GWAS）构建收缩压（Systolic blood pressure，SBP）、舒张压（Diastolic blood pressure，DBP）、脉压（Pulse pressure，PP）和平均动脉压（Mean arterial pressure，MAP）4个PRS，用于反映血压的遗传风险，共纳入354个遗传位点。遗传风险按PRS的五分位数分为低（<20%）、中（20-80%）和高（≥80%）PRS组。研究对象的CVH水平参考AHA推荐的LE8评分系统衡量，按<50、50-80和≥80分划分为低、中和高CVH组。本研究涉及的结局包括高血压和发病年龄以及CVD、CHD和脑卒中发病，主要通过面对面调查收集，辅以电子病历、诊断证明和死亡证明核查。高血压定义为SBP≥140 mmHg，和/或DBP≥90 mmHg，和/或最近两周内服用过降压药物。高血压的发病时间依据问卷调查收集的首次诊断时间、开始服用降压药时间以及调查中血压测量日期三者中最早的时间来确定，发病年龄通过发病时间减去出生时间确定。根据发病年龄，研究还定义了早发（<55岁发病）和晚发（≥55岁发病）高血压。CVD发病定义为首次发生非致死性急性心肌梗死、不稳定性心绞痛、脑卒中、心衰和CVD死亡。CHD包括非致死性急性心肌梗死、不稳定性心绞痛及CHD死亡，脑卒中包括非致死性和致死性的缺血性和出血性脑卒中。

首先，研究采用以年龄为尺度的分层Cox比例风险模型，按<45、45-55、55-65和≥65岁发病年龄段进行分组，计算不同年龄组内血压遗传风险和基线CVH水平与高血压发病的风险比（Hazard ratio，HR）及95%置信区间（Confidence interval，CI）。通过在模型中加入乘积项后采用似然比检验，分别评估遗传风险和CVH与年龄组间的乘法交互作用。采用限制性平均生存时间（Restricted mean survival time，RMST）模型直观反映不同遗传风险和CVH组间发病年龄的差异。随后，探索不同遗传风险分层下CVH水平对早发和晚发高血压的影响，并将人群按遗传风险和CVH水平联合分为9组分析联合作用，评估遗传风险和CVH的乘法和加法交互作用。进一步地，为探究高血压发病年龄对CVD发病风险的影响，将基线前和随访中发生高血压的人群与非高血压人群按照年龄、性别、所属子队列、南北方地区以及随访时长进行1：1随机匹配分析。按发病/匹配年龄（对于高血压人群为发病年龄，对于非高血压人群为匹配年龄）<45、45-55和≥55岁进行分组，在各发病/匹配年龄组中，采用条件Cox比例风险模型，以观察时间为尺度计算高血压与CVD、CHD和脑卒中发病关联的HR及95%CI，通过加入乘积项检验发病/匹配年龄与高血压间的乘法交互作用。最后，在不同发病年龄的高血压人群中，探究发病/匹配后最近一次的CVH水平与CVD、CHD和脑卒中发病风险的关联。

结果

血压遗传风险及CVH水平与高血压发病年龄的关联：研究共纳入23,475名研究对象，平均随访14.4年后共观察到9,817例高血压发病。较高的血压（SBP、DBP、PP和MAP）PRS与高血压发病年龄提前密切相关，表现为发病年龄越小，PRS与高血压发病关联的效应值越大。以SBP的PRS为例，相比于低PRS组，高PRS组在<45、45-55、55-65和≥65岁发生高血压的HR（95%CI）分别为1.875（1.378，2.550）、1.671（1.474，1.895）、1.339（1.204，1.489）和1.178（1.056，1.314），早发高血压（HR：1.694；95%CI：1.508，1.903）的风险大于晚发高血压（HR：1.259；95%CI：1.167，1.358），PRS与年龄组存在显著乘法交互作用（P_交互<0.001）。RMST模型显示，中和高PRS组的发病年龄比低PRS组平均提前2.008和4.397年。良好的CVH与高血压发病年龄推迟相关，其对早发高血压的保护作用也较晚发高血压更明显。相比于低CVH组，高CVH组的早发和晚发高血压风险分别降低52.9%（HR：0.471；95%CI：0.388，0.570）和20.4%（HR：0.796；95%CI：0.692，0.915）（P_交互<0.001），高CVH组的高血压发病年龄比低CVH组平均推迟2.021年。

按PRS分层分析显示，较低的CVH水平对早发高血压的影响在高遗传风险组中更大。在低、中和高PRS人群中，以高CVH组为参照时，低CVH组早发高血压的HR（95%CI）分别为1.490（0.860，2.581）、2.061（1.615，2.630）和2.893（1.973，4.241），而晚发高血压的HR（95%CI）均相对较小，分别为1.347（1.003，1.809）、1.329（1.111，1.591）和1.422（1.048，1.929）。联合效应分析显示，与低PRS和高CVH组相比，高PRS和低CVH组的高血压发病风险最大，其中早发高血压的HR（95%CI）为4.277（2.958，6.183），而晚发高血压的HR（95%CI）为1.986（1.475，2.673），高PRS与低CVH对早发高血压存在显著的协同作用（协同作用指数：1.982；95%CI：1.027，3.824）。DBP、PP和MAP的PRS结果类似。

高血压发病年龄与CVD、CHD和脑卒中发病风险的关联及CVH的作用：高血压和非高血压人群各29,015例经随机匹配后被纳入分析，经过平均12.0年的随访，高血压人群中分别观察到4,209、1,111和3,046例CVD、CHD和脑卒中发病，非高血压人群中分别观察到2,294、620和1,519例。结果显示，高血压与CVD、CHD和脑卒中发病风险的关联随高血压发病年龄提前而增强。以CVD为例，发病年龄在<45、45-55和≥55岁的高血压人群中CVD发病的HR（95%CI）分别为2.269（1.838，2.802）   、1.980（1.710，2.303）      和1.760（1.578，1.963），发病/匹配年龄与高血压间存在显著的乘法交互作用（P_交互<0.001）。高CVH水平与高血压人群CVD发病风险显著降低相关，高血压发病年龄在<45、45-55、≥55岁的患者中相应的HR（95%CI）分别为0.633（0.494，0.810）、0.699（0.582，0.840）和0.722（0.642，0.813）。CHD和脑卒中的结果与CVD类似。

结论

本研究利用中国前瞻性队列分析发现，血压遗传风险与高血压发病年龄提前相关，其对早发高血压的影响较晚发高血压更大，且较早的高血压发病与随后CVD的发病风险升高显著相关。良好的CVH水平不仅与高血压的发病推迟相关，可以抵消一部分遗传风险带来的危害，还可能降低早发高血压人群未来发生CVD的风险。研究结果量化了遗传风险和CVH水平对高血压发病年龄，以及高血压发病年龄在随后CVD发生过程中的重要作用，揭示了早期遗传风险评估和开展CVH管理的重要性，为早发高血压高危人群的精准识别及管理提供了证据支持。

Background and Objectives

Hypertension, as a key risk factor for cardiovascular disease (CVD) and premature death, has been experiencing a continuous rise in prevalence and is gradually showing a younger-onset trend both globally and in China. Compared with elderly hypertensive patients, younger patients have lower awareness, treatment, and control rates. Affected by long-term accumulation of poor blood pressure (BP) level, they are likely to experience higher risk of CVD and severe organ damage, imposing heavy disease burden on individuals and society.

The development of hypertension results from a combination of genetic and environmental factors, with genetics potentially playing a significant role in young individuals. Although several genetic variants associated with early-onset hypertension have been identified, the loci are inconsistent across different study populations. Moreover, most polygenic risk scores (PRSs) for BP developed from western populations demonstrate poor predictive ability in the Chinese population. The impact of genetic factors on the onset age of hypertension in Chinese has not been quantified yet. In addition to genetic factors, modifiable risk factors also play a crucial role in the prevention and control of hypertension. The American Heart Association proposed the definition and algorithm of Life’s Essential 8 (LE8) in 2022, including eight indicators of health behaviors and metabolic factors, to guide cardiovascular health (CVH) management for people. Previous studies suggested that genetic risk and lifestyle might have interaction effect on CVD risk. However, the manifestation of this interaction in the onset of hypertension remains unclear, particularly whether it differs between early-onset and late-onset hypertension requires further exploration. Furthermore, coronary heart disease (CHD) and stroke are the main subtypes of CVD, with disease burden continuing to rise. Although hypertension is recognized as a significant risk factor for CHD and stroke, the impacts of hypertension onset age on CHD and stroke risks remain controversial. Additionally, whether ideal CVH can mitigate CHD and stroke risks among hypertensive individuals with different onset ages remains unclear and requires further exploration.

Using a prospective cohort from China, we constructed BP-related PRSs to quantify the association between genetic risk for BP and the onset age of hypertension, as well as the risks of early and late-onset hypertension. We also assessed the combined and interactive impacts of CVH and genetic risk for BP on early and late-onset hypertension. Subsequently, we further explored the association of onset age of hypertension with CVD, CHD and stroke risks, and investigated the protective impact of ideal CVH against these risks among hypertensive individuals with varying hypertension onset ages. This research would provide scientific evidence for the identification and management of high-risk populations with early-onset hypertension, as well as for the precision prevention and control of CVD.

Subjects and Methods

This study was based on three prospective sub-cohorts with genetic information in the Prediction for Atherosclerotic Cardiovascular Disease Risk in China (China-PAR) project, including the International Collaborative Study of Cardiovascular Disease in Asia (InterASIA), the China Multi-Center Collaborative Study of Cardiovascular Epidemiology 1998 (ChinaMUCA 1998), and the Community Intervention of Metabolic Syndrome in China and Chinese Family Health Study (CIMIC). Baseline investigations of the ChinaMUCA 1998, InterASIA, and CIMIC were conducted in 1998, 2000-2001, and 2007-2008, respectively. The first follow-ups of ChinaMUCA 1998 and InterASIA were also conducted in 2007-2008, and subsequent follow-up visits of three sub-cohorts were investigated in 2012-2015 and 2018-2021.

Data on demographic characteristics, lifestyles, disease history, medication history, and family history were collected through questionnaires. Physical examinations were preformed to gather information on height, weight, and BP level, and fasting peripheral blood samples were collected for biochemical testing and genotyping. Based on the large-scale genome-wide association study (GWAS) of East Asian populations, four BP-related PRSs were constructed to assess the genetic risk for BP, including PRS for systolic BP (SBP) and diastolic BP (DBP), pulse pressure (PP) and mean arterial pressure (MAP) containing 354 genetic loci with genome-wide significant associations（5×10^-8）. Each PRS was categorized into low (<20%), medium (20-80%), and high (≥80%) genetic risk groups by the quintile. CVH level was assessed by the AHA’s LE8 scoring system, and participants were categorized into low (<50 points), medium (50-80 points), and high (≥80 points) CVH groups. The outcomes including hypertension incidence and onset age, as well as CVD, CHD and stroke incidence, were mainly collected through face-to-face interviews and verified against electronic medical records, diagnostic certificates, and death certificates. Hypertension was defined as SBP≥140 mmHg, and/or DBP≥90 mmHg, and/or taking antihypertensive drugs within the past two weeks. The onset time of hypertension was determined based on the earliest date among three indicators: the first diagnosis date collected via questionnaire survey, the initiation date of antihypertensive medication, and the BP measurement date of the investigation. Onset age was calculated by subtracting the date of birth from the determined onset time. Early-onset hypertension was defined as onset at age <55 years, and late-onset was defined as onset at age ≥55 years. CVD incidence was defined as the first occurrence of non-fatal acute myocardial infarction, unstable angina, stroke, heart failure, and CVD death. CHD included non-fatal acute myocardial infarction, unstable angina, and CHD death. Stroke included non-fatal and fatal ischemic stroke and hemorrhagic stroke.

Firstly, the study used stratified Cox proportional hazards model with age as timescale and grouped participants by onset age of <45, 45-55, 55-65, and ≥65 years old, to calculate the hazard ratio (HR) and 95% confidence interval (CI) for the association of genetic risk for BP and baseline CVH level with hypertension risk in different onset age groups. By adding product terms to the models and performing likelihood ratio tests, we separately evaluated the multiplicative interactions between genetic risk or CVH, and age groups. Meanwhile, the restricted mean survival time (RMST) model was used to directly reflect differences in onset age of hypertension across different genetic risk and CVH groups. Subsequently, participants were stratified by different genetic risk levels to explore the impacts of CVH level on early-onset and late-onset hypertension. Then, participants were further categorized into nine groups based on the combination of genetic risk and CVH levels to analyze their joint effects, and the multiplicative interaction and additive interaction between genetic risk and CVH were evaluated. Furthermore, to explore the impact of onset age of hypertension on CVD risk, a 1:1 random matching analysis was performed for hypertensive and non-hypertensive individuals observed at baseline and during follow-up, based on age, sex, subcohort, southern/northern regions and follow-up duration. The participants were grouped by onset/matching age (onset age of hypertension for hypertensive individuals and matching age for non-hypertensive individuals) into <45, 45-55, and ≥55 years. Within each onset/matching age group, conditional Cox proportional hazards model was used to estimate HRs and 95% CIs for the association between hypertension and CVD, CHD and stroke risks, with follow-up time as the scale. Multiplicative interactions between onset/matching age and hypertension were tested by adding product terms to the model. Finally, the associations of CVH level with CVD, CHD and stroke risks were investigated among hypertensive individuals with different hypertension onset ages.

Results

The associations of genetic risk and CVH level with the onset age of hypertension: A total of 23,475 participants were included, and there were 9,817 cases of hypertension observed after an average follow-up of 14.3 years. Higher genetic risks for SBP, DBP, PP and MAP were significantly associated with earlier onset of hypertension, with the associations being stronger at younger onset age. Taking the PRS for SBP as an example, compared with the low PRS group, the HRs (95%CIs) for hypertension onset at age <45, 45-55, 55-65, and ≥65 years were 1.875 (1.378, 2.550), 1.671 (1.474, 1.895), 1.339 (1.204, 1.489) and 1.178 (1.056, 1.314) in high PRS group, respectively. The risk of early-onset hypertension (HR: 1.694; 95% CI: 1.508, 1.093) was greater than that of late-onset hypertension (HR: 1.259; 95% CI: 1.167, 1.358), and a significant multiplicative interaction was observed between PRS and age groups (P_interaction <0.001). RMST model indicated that the medium and high PRS groups experiencing average onset ages of 2.008 and 4.397 years earlier than the low PRS group, respectively. Ideal CVH was associated with a delayed onset age of hypertension, exhibiting stronger protective impact for early-onset than late-onset hypertension. Compared with the low CVH group, the risks of early-onset and late-onset hypertension in high CVH group were reduced by 52.9% (HR: 0.471; 95% CI: 0.388, 0.570) and 20.4% (HR: 0.796; 95% CI: 0.692, 0.915), respectively (P_interaction <0.001). The average onset age of hypertension in the high CVH group was delayed by 2.021 years compared with the low CVH group.

Stratified analysis by PRS showed that the impact of lower CVH level on early-onset hypertension was more pronounced in the high genetic risk group. Compared with high CVH group, the HRs (95% CIs) for early-onset hypertension in the low CVH group were 1.490 (0.860, 2.581), 2.061 (1.615, 2.630) and 2.893 (1.973, 4.241) among individuals with low, medium, and high PRS, respectively, while the corresponding HRs (95% CIs) for late-onset hypertension were relatively lower, at 1.347 (1.003, 1.809), 1.329 (1.111, 1.591) and 1.422 (1.048, 1.929), respectively. The joint effect analyses showed that the high PRS and low CVH group had the highest risk of hypertension, with HRs (95%CIs) of 4.277 (2.958, 6.183) for early-onset hypertension, and 1.986 (1.475, 2.673) for late-onset hypertension, demonstrating significant synergistic effect between high genetic risk and low CVH on early-onset hypertension (synergy index: 1.982；95%CI: 1.027, 3.824). Similar results were also observed for genetic risks for DBP, PP and MAP.

The association of onset age of hypertension with CVD, CHD and stroke incidence risks and the impact of CVH: A total of 29,015 hypertensive individuals and 29,015 non-hypertensive individuals were included after random matching. Over an average follow-up of 12.0 years, 4,209, 1,111, and 3,046 incident cases of CVD, CHD, and stroke were observed in the hypertensive group, respectively, while 2,294, 620, and 1,519 cases were observed in the non-hypertensive group, respectively. The results indicated that the associations between hypertension and the risks of CVD, CHD and stroke were found to increase with earlier onset age of hypertension. Taking the outcome of CVD as an example, compared with non-hypertensive individuals, the HRs (95%CIs) of CVD incidence in hypertensive individuals with onset ages of <45, 45-55, and ≥55 years were 2.269 (1.838, 2.802), 1.980 (1.710, 2.303) and 1.760 (1.578, 1.963), respectively, demonstrating a significant multiplicative interaction between onset/matching age and hypertension (P_interaction <0.001). Higher CVH level was found to be associated with lower CVD risk among hypertensive patients. In patients with <45, 45-55, and ≥55 years of onset age, the HRs (95%CIs) of CVD risk in high CVH group was 0.633 (0.494, 0.810), 0.699 (0.582, 0.840) and 0.722 (0.642, 0.813), respectively, compared with low CVH group. The results of CHD and stroke were similar to those of CVD.

Conclusions

This study used data from a Chinese prospective cohort and found significant associations between blood pressure genetic risk and earlier onset age of hypertension, with a more pronounced effect on early-onset compared to late-onset hypertension. Earlier onset of hypertension is further significantly associated with increased risk of subsequent CVD. Ideal CVH level can not only partially counteract the harmful impact caused by genetic risk and delay the onset of hypertension, but also reduce future CVD risk among early-onset hypertensive patients. The findings quantify the critical roles of genetic risk and CVH in the development of early-onset hypertension and subsequent CVD events, highlighting the importance of early genetic risk assessment and CVH management, thereby providing evidence for the precise identification and management of high-risk groups for early-onset hypertension.