Aging represents a pivotal risk factor for cardiovascular disease, and the Life’s Essential 8 (LE-8) score, encompassing physical activity, diet, nicotine exposure, sleep health, BMI, blood glucose, blood lipids, and blood pressure, has been associated with age-related diseases. To elucidate the relationship between LE-8 scores and biological aging, a cross-sectional study was conducted utilizing the UK Biobank cohort. The LE-8 score, ranging from 0 to 100, adheres to the American Heart Association’s standards. Biological aging was evaluated using the frailty index, phenotypic age acceleration, retinal age gap, and brain age gap. Multiple linear regression analyses were performed to assess correlations. Among the 233,031 participants (mean age 56.0 years, 50.7% female), 43,497 (18.7%) had a high LE-8 score, 177,950 (76.3%) had a moderate score, and 11,584 (5.0%) had a low score. The average measurements for biological aging markers were: ocular aging clock 0.30 ± 4.55 years, phenotypic age acceleration − 4.06 ± 5.24 years, frailty index 0.13 ± 0.07 years, and brain age gap 0.64 ± 3.59 years. After adjusting for potential confounders, the LE-8 score exhibited significant correlations with the ocular aging clock (moderate vs. low LE-8: β: 0.25; 95% CI: 0.02–0.48; high vs. low LE-8: β: 0.68; 95% CI: 0.20–1.16; P trend = 0.001), phenotypic age acceleration (moderate vs. low LE-8: β: 0.27; 95% CI: 0.22–0.33; high vs. low LE-8: β: 1.47; 95% CI: 1.35–1.58; P trend < 0.001), and frailty index (moderate vs. low LE-8: β: 0.01; 95% CI: 0.01–0.01; high vs. low LE-8: β: 0.02; 95% CI: 0.02–0.02; P trend < 0.001). Furthermore, nicotine exposure and blood glucose scores were correlated with all four biological aging metrics. Overall, the LE-8 score and its components were inversely associated with biological aging., suggesting that maintaining an optimal LE-8 score may attenuate the aging process.
Research Article
Impact of cardiovascular health on biological aging: evidence from UK Biobank
https://doi.org/10.21203/rs.3.rs-4657843/v1
This work is licensed under a CC BY 4.0 License
You are reading this latest preprint version
The global community is witnessing a rise in life expectancy, along with the acceleration of population aging. As stated by the United Nations, the population aged 65 and above stood at 727 million in 2020 and is anticipated to surge to 1.6 billion by 2050, constituting over 16% of the global populace[1]. Nonetheless, the healthy life expectancy, defined as the duration of life free from disease, has not escalated in tandem with overall life expectancy. Consequently, the increasing incidence of cardiovascular events has imposed significant economic burdens stemming from treatment costs[2]. Therefore, promoting healthy aging, which entails maximizing the duration of disease-free life, remains a pressing priority. Nonetheless, the question remains unsettled regarding whether maintaining cardiovascular health can defer the aging process.
In 2010, the American Heart Association introduced Life’s Simple 7 (LE-7) as a means to evaluate and enhance cardiovascular health while also aiming to decrease disease incidence; the LE-7 comprises various metrics, including physical activity, diet quality, smoking, weight, blood sugar, cholesterol, and blood pressure[3]. In 2022, sleep health was incorporated into the framework, expanding LE-7 to Life’s Essential 8 (LE-8)[2]. Individuals with higher LE-8 scores are found to have lower risks of age-related diseases, such as dementia[4], diabetes[5], and cardiovascular disease[6], as well as a reduced mortality rate[7] and longer healthy life expectancy[8].
Aging constitutes an intricate biological process encompassing cells, tissues, and organs. A diverse array of biomarkers, spanning from physical attributes to molecular alterations, serves as indicators of biological aging[9]. The gerontology hypothesis posits that the deceleration of biological aging has the potential to forestall diseases and extend healthy life expectancy. Previous study has revealed a correlation between LE-8 and accelerated phenotypic and biological ages[10], albeit the indicators stemming from plasma biomarkers need to be more comprehensive in their assessment. There exists a dearth of comprehensive studies examining the correlation between LE-8 and other measures of biological age.
There are four widely used biological age indicators: frailty index, phenotypic age acceleration, retinal age gap, and brain age gap. The frailty index assesses disease susceptibility based on the accumulation of health deficits[11]. Phenotypic age acceleration, defined as the discrepancy between phenotypic and chronological age, is calculated using nine clinical biomarkers[12]. The retinal age gap[13] and brain age gap[14], derived from retinal images and brain MRI images, respectively, quantify the disparities between retinal/brain and chronological ages. These indicators are regarded as reliable indicators of aging outcomes.
This study uses data from the UK Biobank cohort to evaluate the correlation between LE-8 scores and biological age. The objectives of this study are to investigate whether maintaining cardiovascular health can decelerate aging and to identify the LE-8 components that exhibit the strongest correlation with biological age. This approach will ultimately enhance our comprehension of the underlying mechanism of aging.
Study Participants
The UK Biobank (UKB) is a prospective, observational, population-based cohort study encompassing over 500,000 UK residents aged 40 to 69, recruited from 2006 to 2010[15]. Participants consented to undergo continuous monitoring through the National Health Service (NHS). At baseline, participants completed electronic questionnaires encompassing sociodemographic, lifestyle factors, environmental exposures, medical histories, and cognitive function assessments. Furthermore, participants underwent physical and functional examinations, providing biological samples, including blood, urine, and saliva. In 2010, the study introduced eye examinations, inclusive of retinal photographs. Since 2015, more than 48,000 participants have undergone brain MRI imaging. Comprehensive details regarding the UK Biobank are available from other sources. This study was subject to review and approval by the National Health and Social Care Information Governance Council and the NHS North West Multi-Center Study Ethics Committee (11/NW/0382). This study adheres strictly to the Declaration of Helsinki, and informed consent was obtained from all participants.
Measurement of LE-8
Based on the American Heart Association guidelines, LE-8 comprises four health behaviors (diet, physical activity, nicotine exposure, sleep health) and four health factors (BMI, blood lipids, blood sugar, and blood pressure). Owing to the disparities in dietary measures between NHANES and UKB, we opted for the Healthy Diet Score[16] as an alternative to the AHA Diet Score. The Healthy Diet Score was based on the consumption of fruits, vegetables, whole grains, fish, dairy products, and vegetable oil, while also considering the reduction of refined grains, processed meat, unprocessed meat, and sugary drinks. Each dietary goal achieved was awarded 1 point to participants with a maximum total of 10 points. The scores of the LE-8 components ranged from 0 to 100. The composite cardiovascular health score was calculated as an unweighted average of all components, with a range of 0 to 100. Scores ranging from o 80 to 100 were considered ideal, scores between 50 and 79 were moderate, and scores between 0 and 49 were deemed poor (Table S1).
Retinal Photography and Retinal Age Gap
In 2010, retinal fundus and optical coherence tomography (OCT) imaging were conducted using the Topcon 3D OCT 1000 Mk2 (Topcon Corp). A non-scattered and non-stereoscopic 45-degree retinal fundus image centered on the optic disc and macula was captured. Among the 131,238 fundus images collected from 66,500 participants, 80,169 images from 46,969 participants fulfilled the criteria for image quality and were included in the study. The specifics of the image quality assessment are described in another study[13]. A total of 19,200 retinal fundus images from 11,052 healthy participants were used to train and validate the age prediction model. The evaluation of the deep learning (DL) model’s performance involved the Mean Absolute Error (MAE) and the correlation coefficient between the predicted retinal age and chronological age. The DL model accurately predicted chronological age with a correlation coefficient of 0.80 (P < 0.001) and an MAE of 3.55 years. The retinal age gap refers to the difference between retinal and chronological age, where a positive value signifies an "older" retina, and a negative value signifies a "younger" retina.
Phenotypic Age Acceleration and Frailty Index
Phenotypic age was determined using the following mathematical expressions[17]:
Phenotypic age acceleration is defined as the deviation between phenotypic and chronological ages, where positive values signify accelerated aging.
The frailty index, a metric of cumulative health deficits, was evaluated based on 49 self-reported items spanning various health domains. It was calculated as the proportion of deficits present relative to the total possible deficits for each participant, ranging from 0 to 1. A higher frailty index signifies a greater degree of biological aging.
Brain MRI Imaging and Measurement of Brain Age Gap
Brain MRI data were obtained using a 3.0T Siemens scanner (Skyra, VD13ASP4, Siemens Healthcare) under a publicly accessible agreement. The imaging protocol comprised six sequences: structural, diffusion, and functional imaging, with a scanning duration of 35 minutes per subject. High-resolution T1-weighted images were acquired using the MPRAGE sequence, achieving 1 mm isotropic resolution, while T2-weighted scanning utilized fluid attenuation inversion recovery contrast with a resolution of 1.05x1x1x1 mm. The Supplemental Methods section provides descriptions of the data processing procedures. We identified 25 volumetric features extracted from structural MRI (Table S2) as crucial predictors for estimating brain age[18].
Bayesian Ridge regression, known for its robust performance, was employed to estimate the brain age. These 25 brain imaging-derived phenotypes (IDPs) served as independent variables, while chronological age served as the dependent variable. To mitigate the effects of confounders, volumetric scaling was adjusted for gender, education level, height, and T1 head size normalization to standard space. Prior to brain age modeling, a linear regression model was implemented to eliminate the influence of head size on the selected features. The brain aging model was constructed using data from healthy participants with no underlying diseases at baseline. The dataset was randomly divided into training (80%) and test (20%) subsets. Model performance was assessed using the mean absolute error (MAE) and the coefficient of determination (R²). MAE quantifies the deviation between predicted brain age and chronological age, with higher values suggesting a more aged appearance. Conversely, R² indicates the proportion of variance in predicted brain age explained by the model. Using the trained model, we estimated the brain age of the non-healthy population. After eliminating age-dependent bias, brain age increment was calculated by subtracting chronological age from predicted brain age. Positive values indicate accelerated brain aging.
Assessment of Covariates
Covariates encompassed factors previously associated with LE-8 or biological aging: age (continuous), gender (female, male), race (white, other), education level (college or university degree, other), BMI (continuous), smoking history (never, former, current), alcohol consumption status (never, former, current), serum creatinine level (continuous, mg/dL), C-reactive protein level (continuous, mg/dL), longevity polygenic risk score (continuous), and the presence of the APOE4 gene (yes, no).
Statistical Analysis
Baseline characteristics were summarized via weighted means with standard error for continuous variables and sample size with weighted percentage for categorical variables. To investigate differences among LE-8 score groups (poor, moderate, ideal), chi-square tests were employed for categorical variables, while one-way ANOVA was utilized for continuous variables.
Multiple linear regression was conducted to investigate the relationship between LE-8 score and biological age metrics, including retinal age gap, phenotypic age acceleration, and brain age delta. Furthermore, the association between each LE-8 component (diet, physical activity, nicotine exposure, sleep health, BMI, blood lipids, blood glucose, and blood pressure) and biological age metrics was explored. Four regression models were constructed: Model 1 adjusted for age, gender, and ethnicity; Model 2 Furthermore adjusted for education level, BMI, smoking status, alcohol consumption status, serum creatinine, and C-reactive protein; Model 3 further incorporated the longevity polygenic risk score; and Model 4 further considered the APOE4 gene.
All statistical analyses were conducted using Stata (version 17, StataCorp, Texas, USA), and a two-sided p-value threshold of < 0.05 was set as the criterion for statistical significance.
Characteristics of the study participants
Among the 233,031 individuals assessed for LE-8 scores, 43,497 (18.7%) exhibited desirable scores, 177,950 (76.3%) demonstrated moderate scores, and 11,584 (5.0%) possessed very low scores. At baseline, the mean age of the cohort was 56.00 ± 8.15 years, with 118,063 participants (50.7%) being female and 220,881 (94.8%) belonging to the white ethnic group.
Table 1 summarizes the baseline characteristics of participants categorized according to their LE-8 scores. In comparison to individuals with optimal LE-8 scores, those with low scores exhibited a greater tendency to be older, male, socioeconomically disadvantaged, non-white, and current smokers. Furthermore, they often had shorter sleep durations, unhealthy dietary habits, and reduced physical activity. Notably, these individuals were more prone to have a history of diabetes or cardiovascular disease, requiring medications for hypertension, hyperglycemia, or hyperlipidemia, and exhibiting accelerated aging in terms of phenotypic age, brain age, and frailty index (P < 0.001).
| Overall | Life’s essential 8 score | P trend | |||
|---|---|---|---|---|---|
| Ideal | Intermediate | Poor | |||
| No. of participants, % | 236755 | 47221 (19.9%) | 177950 (75.2%) | 11584 (4.9%) | < 0.001 |
| Socio-demographic data | |||||
| Age, years | 55.98 ± 8.16 | 53.67 ± 8.37 | 56.57 ± 8.02 | 56.00 ± 7.78 | < 0.001 |
| Female, % | 120444 (50.9%) | 32608 (69.1%) | 84118 (47.3%) | 3718 (32.1%) | < 0.001 |
| White ethnicity, % | 224420 (94.8%) | 45117 (95.5%) | 168602 (94.7%) | 10701 (92.4%) | < 0.001 |
| Socio-economic status | |||||
| Townsend deprivation index | -1.54 ± 2.94 | -1.80 ± 2.78 | -1.54 ± 2.94 | -0.56 ± 3.34 | < 0.001 |
| College or university degree | 149815 (63.3%) | 24823 (52.6%) | 116009 (65.2%) | 8983 (77.5%) | < 0.001 |
| Annual household income | |||||
| <18000 | 38684 (16.3%) | 5670 (12.0%) | 30230 (17.0%) | 2784 (24.0%) | < 0.001 |
| 18000–30999 | 51883 (21.9%) | 9103 (19.3%) | 40159 (22.6%) | 2621 (22.6%) | < 0.001 |
| 31000–51999 | 56625 (23.9%) | 11781 (24.9%) | 42321 (23.8%) | 2523 (21.8%) | < 0.001 |
| 52000–100000 | 47011 (19.9%) | 11580 (24.5%) | 33789 (19.0%) | 1642 (14.2%) | < 0.001 |
| >100000 | 13639 (5.8%) | 3990 (8.4%) | 9289 (5.2%) | 360 (3.1%) | < 0.001 |
| Unknown | 28913 (12.2%) | 5097 (10.8%) | 22162 (12.5%) | 1654 (14.3%) | < 0.001 |
| Anthropologic data | |||||
| Systolic blood pressure, mmHg | 137.45 ± 18.48 | 126.49 ± 16.00 | 139.61 ± 17.87 | 149.03 ± 18.22 | < 0.001 |
| Diastolic blood pressure, mmHg | 82.04 ± 10.07 | 75.68 ± 8.18 | 83.19 ± 9.70 | 90.29 ± 10.09 | < 0.001 |
| Body mass index, kg/m2 | 26.90 ± 4.35 | 23.86 ± 2.73 | 27.39 ± 4.17 | 31.62 ± 5.08 | < 0.001 |
| Waist to hip ratio | 0.87 ± 0.09 | 0.81 ± 0.07 | 0.88 ± 0.08 | 0.94 ± 0.08 | < 0.001 |
| Lifestyle data | |||||
| Current smoker, % | 20689 (8.7%) | 360 (0.8%) | 15989 (9.0%) | 4340 (37.5%) | < 0.001 |
| Current drinker, % | 221906 (93.8%) | 44212 (93.6%) | 166896 (93.8%) | 10798 (93.3%) | < 0.001 |
| Sleep duration, hours | 7.15 ± 1.03 | 7.30 ± 0.78 | 7.14 ± 1.04 | 6.78 ± 1.51 | < 0.001 |
| Healthy diet score | 4.03 ± 1.40 | 4.97 ± 1.08 | 3.87 ± 1.35 | 2.67 ± 1.20 | < 0.001 |
| Physical activity, MET hours/week | 57.11 ± 49.69 | 63.63 ± 47.73 | 55.93 ± 50.01 | 44.04 ± 49.75 | < 0.001 |
| Laboratory results | |||||
| Glycated hemoglobin (HbA1c), % | 5.41 ± 0.55 | 5.23 ± 0.32 | 5.43 ± 0.54 | 5.90 ± 1.03 | < 0.001 |
| Total cholesterol, mmol/L | 5.71 ± 1.12 | 5.23 ± 0.91 | 5.80 ± 1.12 | 6.21 ± 1.20 | < 0.001 |
| Triglyceride, mmol/L | 1.68 ± 0.99 | 1.17 ± 0.57 | 1.77 ± 0.99 | 2.51 ± 1.41 | < 0.001 |
| High-density lipoprotein, mmol/L | 1.47 ± 0.38 | 1.61 ± 0.39 | 1.45 ± 0.37 | 1.28 ± 0.31 | < 0.001 |
| Low-density lipoprotein, mmol/L | 3.56 ± 0.85 | 3.12 ± 0.67 | 3.65 ± 0.85 | 4.01 ± 0.91 | < 0.001 |
| C-reactive protein, mg/L | 2.25 ± 3.85 | 1.53 ± 3.46 | 2.34 ± 3.86 | 3.76 ± 4.56 | < 0.001 |
| Serum creatinine, µmol/L | 72.74 ± 15.45 | 68.82 ± 13.09 | 73.61 ± 15.77 | 75.27 ± 16.68 | < 0.001 |
| eGFR, mg/min/1.73 m2 | 90.98 ± 12.85 | 93.21 ± 12.38 | 90.36 ± 12.87 | 91.49 ± 13.53 | < 0.001 |
| Microalbuminuria, mg/mL | 26.85 ± 102.94 | 19.38 ± 52.00 | 26.52 ± 100.98 | 43.74 ± 169.01 | < 0.001 |
| History of diseases | |||||
| Diabetes mellitus, % | 9353 (4.0%) | 263 (0.6%) | 7742 (4.4%) | 1348 (11.7%) | < 0.001 |
| Cardiovascular diseases, % | 9675 (4.1%) | 1290 (2.7%) | 7692 (4.3%) | 693 (6.0%) | < 0.001 |
| Medications | |||||
| Anti-hypertension drugs, % | 17656 (7.5%) | 2168 (4.6%) | 14493 (8.1%) | 995 (8.6%) | < 0.001 |
| Statins, % | 12471 (5.3%) | 1842 (3.9%) | 9909 (5.6%) | 720 (6.2%) | < 0.001 |
| Anti-diabetes drugs, % | 6446 (2.7%) | 194 (0.4%) | 5326 (3.0%) | 926 (8.0%) | < 0.001 |
| Biological age | |||||
| Deep learning based OAC, years | 0.30 ± 4.55 | 0.76 ± 4.72 | 0.15 ± 4.50 | 0.78 ± 4.42 | < 0.001 |
| Phenotypic age acceleration, years | -4.08 ± 5.24 | -5.68 ± 4.68 | -3.88 ± 5.16 | -0.73 ± 6.37 | < 0.001 |
| 49-item based frailty score | 0.13 ± 0.07 | 0.11 ± 0.06 | 0.13 ± 0.07 | 0.16 ± 0.08 | < 0.001 |
| Brain age delta, years | 0.48 ± 3.54 | 0.40 ± 3.45 | 0.46 ± 3.55 | 1.22 ± 3.57 | < 0.001 |
| *Data were presented as mean ± standard deviation (SD) or number (%). eGFR = estimated glomerular filtration rate; OAC = OculoAgeClock. | |||||
LE-8 and Biological Age Acceleration
In a linear regression model that accounted for age, sex, and ethnicity, individuals with moderate and poor LE-8 scores displayed significantly greater retinal age gap, phenotypic age acceleration, frailty index, and brain age gap compared to those with ideal LE-8 scores. Specifically, for the retinal age gap, the β coefficients were 0.34 (95% CI: 0.14–0.55) for moderate LE-8 scores and 0.97 (95% CI: 0.55–1.39) for poor LE-8 scores, with a statistically significant P trend (< 0.001). Similarly, for phenotypic age acceleration, the β coefficient was 1.45 (95% CI: 1.39–1.50) for moderate LE-8 scores and 4.36 (95% CI: 4.25–4.46) for poor LE-8 scores, with a significant P trend (< 0.001). With respect to the frailty index, the β coefficient was 0.02 (95% CI: 0.02–0.02) for moderate LE-8 scores and 0.05 (95% CI: 0.05–0.06) for poor LE-8 scores, also indicating a significant P trend (< 0.001). Lastly, for the brain age gap, the β coefficient was 0.21 (95% CI: 0.05–0.38) for moderate LE-8 scores and 1.07 (95% CI: 0.73–1.41) for poor LE-8 scores, with a P significant trend (< 0.001). After adjusting all covariates, LE-8 scores correlated significantly with the ocular aging clock, phenotypic age acceleration, and frailty index, as shown in Table 2.
| Coefficient (95%CI) | P trend | |||
|---|---|---|---|---|
| Ideal | Intermediate | Poor | ||
| Retinal age gap, years | ||||
| Model 1 * | 0 (ref) | 0.32 (0.12, 0.52) | 0.94 (0.53, 1.36) | < 0.001 |
| Model 2 † | 0 (ref) | 0.21 (-0.01, 0.43) | 0.65 (0.18, 1.12) | 0.002 |
| Model 3 ‡ | 0 (ref) | 0.23 (0.01, 0.45) | 0.68 (0.20, 1.15) | 0.001 |
| Model 4 § | 0 (ref) | 0.23 (0.01, 0.46) | 0.67 (0.19, 1.15) | < 0.001 |
| Phenotypic age acceleration, years | ||||
| Model 1 * | 0 (ref) | 1.44 (1.39, 1.49) | 4.35 (4.25, 4.46) | < 0.001 |
| Model 2 † | 0 (ref) | 0.28 (0.23, 0.34) | 1.47 (1.36, 1.58) | < 0.001 |
| Model 3 ‡ | 0 (ref) | 0.29 (0.24, 0.34) | 1.49 (1.38, 1.60) | < 0.001 |
| Model 4 § | 0 (ref) | 0.30 (0.24, 0.35) | 1.49 (1.39, 1.60) | < 0.001 |
| Frailty phenotype score | ||||
| Model 1 * | 0 (ref) | 0.02 (0.02, 0.02) | 0.05 (0.05, 0.06) | < 0.001 |
| Model 2 † | 0 (ref) | 0.01 (0.01, 0.01) | 0.02 (0.02, 0.02) | < 0.001 |
| Model 3 ‡ | 0 (ref) | 0.01 (0.01, 0.01) | 0.02 (0.02, 0.02) | < 0.001 |
| Model 4 § | 0 (ref) | 0.01 (0.01, 0.01) | 0.02 (0.02, 0.02) | < 0.001 |
| Brain age gap, years | ||||
| Model 1 * | 0 (ref) | 0.21 (0.05, 0.37) | 1.07 (0.73, 1.41) | < 0.001 |
| Model 2 † | 0 (ref) | -0.17 (-0.34, 0.01) | 0.17 (-0.20, 0.55) | 0.533 |
| Model 3 ‡ | 0 (ref) | -0.15 (-0.33, 0.02) | 0.19 (-0.19, 0.57) | 0.392 |
| Model 4 § | 0 (ref) | -0.16 (-0.33, 0.02) | 0.19 (-0.19, 0.57) | 0.410 |
| * Model 1: adjusted for age, gender, and ethnicity. | ||||
| † Model 2: adjusted for model 1 plus education level, Townsend deprivation index, body mass index, smoker, drinker, serum creatine, C-reactive protein. | ||||
| ‡ Model 3: adjusted for model 2 plus PRS of longevity. | ||||
| § Model 4: adjusted for model 3 plus apolipoprotein E4 (APOE4) variant. Abbreviations: OR = odds ratio; CI = confidence interval. | ||||
LE-8 Components and Biological Age Acceleration
The comprehensive linear regression analysis incorporating adjustment for various factors showed a negative correlation between the increased retinal age and physical activity. Specifically, the association yields β coefficients of -0.18 (95% CI: -0.45–0.09) for moderate physical activity scores and − 0.31 (95% CI: -0.52 - -0.11) for ideal physical activity levels, indicating a statistically significant P trend (< 0.001). Furthermore, nicotine exposure exhibited a similar correlation with a retinal age gap, with β coefficients of -0.08 (95% CI: -0.40–0.23) for moderate nicotine exposure scores and − 0.38 (95% CI: -0.68 - -0.08) for ideal nicotine exposure scores, yielding a P trend of 0.014. Furthermore, blood glucose levels were also associated with the retinal age gap, with β coefficients of -0.31 (95% CI: -0.61 - -0.01) for moderate blood glucose scores and − 0.62 (95% CI: -0.88 - -0.35) for ideal blood glucose scores, displaying a statistically significant P trend (< 0.001).
Accelerated brain age exhibited significant association with five components of the LE-8 score. For diet, β coefficients of 0.14 (95% CI: 0.00–0.28) for moderate diet scores and 0.31 (95% CI: 0.10–0.53) for ideal diet scores, with a significant P trend (= 0.009). Nicotine exposure was associated with accelerated brain age, β coefficients of -0.50 (95% CI: -0.77 - -0.22) for moderate nicotine exposure scores and − 0.98 (95% CI: -1.25 - -0.72) for ideal nicotine exposure scores, both indicating a significant P trend (< 0.001). For lipids, β coefficients of 0.07 (95% CI: -0.09–0.22) for moderate lipid scores and 0.27 (95% CI: 0.06–0.48) for ideal lipid scores, indicating a significant P trend < 0.001. Blood glucose levels were also associated with accelerated brain age, β coefficients of -0.60 (95% CI: -0.85 - -0.35) for moderate blood glucose scores and − 0.85 (95% CI: -1.06 - -0.63) for ideal blood glucose scores, with a significant P trend (< 0.001). Blood pressure was associated with accelerated brain age, β coefficients of -0.29 (95% CI: -0.45 - -0.13) for moderate blood pressure scores and − 0.37 (95% CI: -0.60 - -0.15) for ideal blood pressure scores, with a P trend < 0.001. Phenotypic age acceleration and frailty index exhibited a significant association with all components of LE-8, as outlined in Table 3.
| Poor | Intermediate | Ideal | P trend | |
|---|---|---|---|---|
| Retinal age gap, years | ||||
| Diet | 0 (Ref) | -0.16 (-0.34, 0.03) | 0.12 (-0.14, 0.38) | 0.821 |
| Physical activity | 0 (Ref) | -0.20 (-0.47, 0.07) | -0.31 (-0.52, -0.11) | < 0.001 |
| Nicotine exposure | 0 (Ref) | -0.09 (-0.41, 0.22) | -0.37 (-0.67, -0.07) | 0.017 |
| Sleep health | 0 (Ref) | 0.32 (-0.09, 0.72) | 0.14 (-0.24, 0.51) | 0.323 |
| Body mass index | 0 (Ref) | -0.04 (-0.37, 0.29) | 0.13 (-0.35, 0.61) | 0.437 |
| Blood lipids | 0 (Ref) | 0.10 (-0.10, 0.30) | 0.01 (-0.24, 0.26) | 0.379 |
| Blood glucose | 0 (Ref) | -0.31 (-0.61, -0.01) | -0.63 (-0.89, -0.37) | < 0.001 |
| Blood pressure | 0 (Ref) | 0.11 (-0.11, 0.33) | -0.10 (-0.39, 0.20) | 0.585 |
| Phenotypic age acceleration, years | ||||
| Diet | 0 (Ref) | -0.41 (-0.45, -0.37) | -0.66 (-0.72, -0.59) | < 0.001 |
| Physical activity | 0 (Ref) | -0.12 (-0.19, -0.06) | -0.14 (-0.18, -0.09) | < 0.001 |
| Nicotine exposure | 0 (Ref) | -1.88 (-1.95, -1.81) | -1.96 (-2.03, -1.89) | < 0.001 |
| Sleep health | 0 (Ref) | -0.42 (-0.51, -0.33) | -0.39 (-0.47, -0.30) | < 0.001 |
| Body mass index | 0 (Ref) | -0.21 (-0.28, -0.13) | 0.10 (-0.01, 0.21) | 0.039 |
| Blood lipids | 0 (Ref) | 0.32 (0.27, 0.36) | 0.89 (0.83, 0.95) | < 0.001 |
| Blood glucose | 0 (Ref) | -1.39 (-1.47, -1.32) | -3.25 (-3.32, -3.19) | < 0.001 |
| Blood pressure | 0 (Ref) | 0.07 (0.02, 0.12) | 0.13 (0.06, 0.20) | < 0.001 |
| Frailty phenotype score | ||||
| Diet | 0 (Ref) | -0.01 (-0.01, -0.01) | -0.01 (-0.01, -0.01) | < 0.001 |
| Physical activity | 0 (Ref) | -0.00 (-0.00, -0.00) | -0.01 (-0.01, -0.01) | < 0.001 |
| Nicotine exposure | 0 (Ref) | -0.01 (-0.01, -0.01) | -0.02 (-0.02, -0.02) | < 0.001 |
| Sleep health | 0 (Ref) | -0.02 (-0.02, -0.02) | -0.04 (-0.04, -0.04) | < 0.001 |
| Body mass index | 0 (Ref) | -0.01 (-0.01, -0.01) | -0.01 (-0.01, -0.00) | < 0.001 |
| Blood lipids | 0 (Ref) | -0.00 (-0.01, -0.00) | -0.00 (-0.00, -0.00) | < 0.001 |
| Blood glucose | 0 (Ref) | -0.02 (-0.02, -0.02) | -0.02 (-0.02, -0.02) | < 0.001 |
| Blood pressure | 0 (Ref) | 0.00 (0.00, 0.00) | 0.01 (0.01, 0.01) | < 0.001 |
| Brain age gap, years | ||||
| Diet | 0 (Ref) | 0.15 (0.01, 0.29) | 0.30 (0.09, 0.51) | 0.013 |
| Physical activity | 0 (Ref) | 0.00 (-0.20, 0.21) | 0.07 (-0.09, 0.23) | 0.381 |
| Nicotine exposure | 0 (Ref) | -0.48 (-0.76, -0.21) | -0.98 (-1.25, -0.71) | < 0.001 |
| Sleep health | 0 (Ref) | -0.42 (-0.74, -0.11) | -0.19 (-0.48, 0.10) | 0.469 |
| Body mass index | 0 (Ref) | -0.31 (-0.56, -0.05) | -0.23 (-0.60, 0.14) | 0.073 |
| Blood lipids | 0 (Ref) | 0.07 (-0.08, 0.23) | 0.25 (0.05, 0.46) | < 0.001 |
| Blood glucose | 0 (Ref) | -0.59 (-0.84, -0.35) | -0.84 (-1.06, -0.63) | < 0.001 |
| Blood pressure | 0 (Ref) | -0.29 (-0.45, -0.13) | -0.34 (-0.57, -0.12) | < 0.001 |
| * Adjusted for age, gender, and ethnicity, education level, Townsend deprivation index, body mass index, smoker, drinker, serum creatine, C-reactive protein, serum creatine, PRS of longevity and apolipoprotein E4 (APOE4) variant. | ||||
In the large prospective cohort of UK Biobank, our analysis revealed a notable inverse correlation between the LE-8 score, a novel cardiovascular health indicator, and multiple metrics of biological aging, specifically retinal age gap, phenotypic age acceleration, frailty index, and brain age gap. After adjusting for confounding variables, the association persisted for the retinal age gap, phenotypic age acceleration, and frailty index. Among the constituents of the LE-8 score, nicotine exposure and blood glucose levels displayed a significant association with all four indices of biological aging. This study offers compelling evidence from a large-scale biobank cohort, affirming the correlation between LE-8 scores and biological aging, and extends these findings to specific LE-8 components. Our results underscore the inherent connection between cardiovascular health and biological aging, implying the potential utility of cardiovascular health in managing biological aging and advancing public health.
Numerous prior investigations have examined the association between cardiovascular health indices and biological aging, employing diverse methodologies. For instance, a study encompassing 5,194 adults within the NHANES cohort (1999–2002) revealed a correlation between suboptimal LE-7 and shortened leukocyte telomeres, suggesting a linkage between cardiovascular health metrics and biological aging at the molecular and cellular levels among a youthful demographic (average age: 44 years)[19]. Furthermore, a separate investigation encompassing 2,170 postmenopausal women within the WHI cohort (1993–1998) demonstrated a positive correlation between optimal LE-7 scores and reduced apparent age acceleration, ascertained using the Horvath and Hannum methodologies, thereby confirming the correlation between cardiovascular health and lifespan extension at the epigenomic level[20]. Nonetheless, the data from these studies need to be updated, given the advancements in techniques for assessing biological aging, thus rendering updated analyses imperative.
In a comprehensive analysis of 2,474 participants from the Taiwan Biobank cohort (2016–2021), the study failed to detect a significant correlation between LE-7 scores and phenotypic age acceleration (EAA), assessed by the Horvath and Hannum methods. However, a negative dose-response relationship was observed with the PhenoEAA and GrimEAA metrics[21]. Notably, the studies neglected to investigate the potential correlation between specific LE-7 components and biological aging. Furthermore, they failed to incorporate an evaluation of sleep health, a crucial addition to the 2022 AHA revised cardiovascular health score, LE-8. Prior study has established a link between sleep health and LE-7 scores, as well as other cardiovascular health metrics[2].
Phenotypic age has been demonstrated to be superior to molecular-level aging measurements in predicting adverse health outcomes. A study encompassing 23,896 participants from the NHANES cohort (2005–2018) revealed that a 10-point increment in LE-8 score correlated with a 1.14-year decrement in phenotypic age, exhibiting significant negative associations across all components except lipids[22]. Another study involving 11,729 NHANES participants (2005–2010) identified a notable nonlinear relationship between LE-8 and biological aging, quantified using phenotypic age and KDM biological age[10]. Notably, as LE-8 scores increased, both phenotypic age acceleration and KDM biological age acceleration decreased, especially in relation to health factors. Nonetheless, these studies were limited in the number of aging indicators utilized, as both phenotypic age and KDM biological age were derived solely from clinical markers. Our study comprehensively elucidates the negative correlation between LE-8 and biological aging in a larger cohort, employing a multifaceted array of biological aging measures.
We observed a 1.47-year increase in phenotypic age acceleration among individuals with low LE-8 scores, in contrast to those with ideal LE-8 scores, aligning with previous studies. For instance, an NHANES study revealed a 5.27-year deceleration in phenotypic age acceleration among individuals with high LE-8 scores as opposed to those with low LE-8 scores[10]. The less significant association observed in the UKB cohort, compared to the NHANES cohort, could be attributed to the UKB’s predominantly white, healthy, and socioeconomically advantaged participants. In contrast, the NHANES cohort comprised a more diverse and socioeconomically heterogeneous population.
The retinal age gap measured by utilizing the deep learning method is associated with various age-related diseases, including Parkinson’s disease[23], metabolic syndrome[24], chronic renal failure[25], and cardiovascular disease[26]. A UKB study revealed that individuals with ideal LE-7 scores exhibited a 42% lower risk of retinal aging compared to those with poor LE-7 scores, and this association was also observed with factors such as smoking, BMI, blood pressure, and blood glucose[27]. By extending the definition of cardiovascular health to the latest LE-8, our study demonstrated that the individuals with low LE-8 scores exhibited an accelerated retinal aging clock of 0.68 years compared to those with ideal LE-8 scores.
Frailty is directly associated with mortality in the elderly population and serves as a critical indicator of accelerated aging. The Chinese Kadoorie Biobank, with over 500,000 participants, revealed that a 0.1 increment in the frailty index correlated with a 0.68-fold rise in all-cause mortality and a 0.89-fold increase in cardiovascular mortality[28]. This finding indicates a potential association between the frailty index and cardiovascular health. A prospective cohort study involving 1,745 individuals aged 65 and older without cardiovascular disease demonstrated that higher LE-7 scores were associated with a reduced risk of frailty[29]. Our study employed 49 indicators to calculate a continuous frailty index, enabling improved differentiation among adults with lower frailty levels. Our study builds upon these findings and reveals that individuals with poor LE-8 scores exhibited a 0.02 increment in the frailty index compared to those with ideal LE-8 scores.
Prior study has demonstrated that accelerated brain age is associated with a 13% heightened risk of dementia and exhibits a significant correlation with ischemic heart disease[14]. Elevated LE-7 scores are associated with a reduced risk of dementia[30]. Our study found a negative correlation between the LE-8 score and the brain age gap, following adjustments for age, sex, and ethnicity; however, this correlation was not significant in the fully adjusted model. Nonetheless, five LE-8 components, comprising diet, nicotine exposure, blood pressure, blood lipids, and blood glucose, exhibited significant associations with the brain age gap. This discrepancy may stem from the necessity for further refinement of the LE-8 score as an evaluation tool, particularly in its assessment of physical activity and sleep health. Furthermore, the current LE-8 score does not include mental health and social factors, which are emphasized in AHA guidelines as crucial determinants of cardiovascular health.
Our findings yield significant implications for decelerating aging processes and advancing public health. Firstly, LE-8 serves as a straightforward indicator for assessing biological aging. While current biological age clocks accurately forecast disease risk, their implementation on a population-wide scale is often intricate and financially burdensome. Conversely, LE-8 indicators are straightforward, cost-effective, and readily accessible, enabling individuals to compute their scores and gain insights into their health status. Secondly, the LE-8 score offers a framework for elucidating the correlation between biological age and disease risk. Elevated frailty index values are linked to a diverse array of chronic diseases, hospital admissions, and mortality. The LE-8 score provides insights into the shared pathophysiological mechanism between frailty and cardiovascular health, encompassing oxidative stress, inflammation, and endothelial dysfunction[31].
Maintaining cardiovascular health effectively contributes to the deceleration of biological aging. Two LE-8 components, nicotine exposure and blood glucose, exhibit a robust correlation with biological age. Lifestyle modifications aimed at these factors can potentially delay the aging process. Nicotine exposure accelerates aging through free radical damage and shortened telomere length[32]. Conversely, smoking cessation significantly mitigates the risk of cardiovascular disease, cognitive decline, lung disease, and cancer[33]. The maintenance of blood glucose homeostasis is intricately associated with aging and cardiovascular health[34], and existing hypoglycemic drugs such as metformin[35] and SGLT-2 inhibitors[36], exhibiting anti-aging properties. Nevertheless, it is imperative to address all LE-8 components, given that their collective impact on biological age surpasses the cumulative effect of individual factors. The attainment of optimal biological age necessitates the proactive management of risk factors and maintenance of high LE-8 levels across all indicators.
This study possesses numerous strengths. The updated LE-8 was employed to assess cardiovascular health, and a thorough analysis was conducted to elucidate the relationship between LE-8 components and biological aging, yielding comprehensive and targeted outcomes. The significant sample size and standardized data collection protocols of the UKB cohort strengthen the validity of this study. Confounding factors were comprehensively accounted for in the analysis. Nonetheless, this study also has certain limitations. The cross-sectional design of this study precludes causal inference regarding the relationship between LE-8 and biological aging, thereby necessitating the conduct of longitudinal studies. The predominantly healthy, white, and socioeconomically advantaged nature of the UKB cohort participants may limit the generalizability of the study findings. The subjective measurement of specific LE-8 components, including physical activity, diet, and nicotine exposure, via self-reported questionnaires poses a potential risk of introducing recall biases. Finally, brain age estimation relied solely on brain volume features, suggesting that the inclusion of additional imaging modalities may further enhance the performance of the model.
This study establishes a negative correlation between the LE-8 score and biological aging, suggesting a close association between cardiovascular health and biological age, as well as adverse health outcomes. Optimizing LE-8 scores can facilitate the monitoring and managing of biological age, leading to improvements in cardiovascular health. Among the LE-8 components, nicotine exposure and blood glucose levels displayed the most significant negative correlations with all four indicators of biological aging. Consequently, sustaining high LE-8 scores, primarily via smoking cessation and blood sugar regulation, could potentially delay the aging process. Future longitudinal studies are imperative to establish a causal relationship between the LE-8 score and biological aging. Furthermore, it is necessary to explore whether lifestyle and behavioral interventions designed to enhance LE-8 scores can substantially modify biological age, ultimately contributing to the delay of aging.
DR = diabetic retinopathy; ETDRS = Early Treatment Diabetic Retinopathy Study; UKB = UK Biobank; GDES = Guangzhou Diabetic Eye Study; STROBE = strengthening the Reporting of Observational Studies in Epidemiology; NMR = nuclear magnetic resonance; HbA1c = hemoglobin A1c; OCT-A = optical coherence tomography-angiography; SCP = superficial capillary plexus; DCP = deep capillary plexus; RPC = radial peripapillary capillary; VD = vessel density; BMI = body mass index; SBP = systolic blood pressure; HDL-c = high-density lipoprotein cholesterol; TC = total cholesterol; NRI = net reclassification improvement; IDI = integrated discrimination improvement; TCA = tricarboxylic acid.
Ethics approval
The study was approved by the Northwest Multicenter Study Ethics Committee (11/NW/0382) and the Ethics Committee of Zhongshan Ophthalmic Center (2017KYPJ094).
Competing interest statement
No potential conflicts of interest relevant to this article were reported.
Author contributions
Study concept and design: WW, WH; Acquisition, analyses, or interpretation: All authors; Drafting of the manuscript: MY, SY, WW; Critical revision of the manuscript for important intellectual content: All authors; Statistical analyses: WW, SY, MY; Obtained funding: WW; Administrative, technical, or material support: All authors; Study supervision: WW, WH.
Acknowledgements
This study was funded by the National Natural Science Foundation of China (82371086, W.W.).
Data availability
All data used in this study are available from UKB via data access procedures (http://www.ukbiobank.ac.uk). Permission to use the UK Biobank Resource was obtained via material transfer agreement as part of Application 62525 and 105658. The data from the GDES are available upon reasonable request to the corresponding author.
- Economic UNDo, Affairs S. World Social Report 2023. United Nations; 2023.
- Lloyd-Jones DM, Allen NB, Anderson CAM, Black T, Brewer LC, Foraker RE, Grandner MA, Lavretsky H, Perak AM, Sharma G et al. Life's Essential 8: Updating and Enhancing the American Heart Association's Construct of Cardiovascular Health: A Presidential Advisory From the American Heart Association. Circulation. 2022; 146(5):e18-e43.10.1161/cir.0000000000001078.
- Lloyd-Jones DM, Hong Y, Labarthe D, Mozaffarian D, Appel LJ, Van Horn L, Greenlund K, Daniels S, Nichol G, Tomaselli GF et al. Defining and setting national goals for cardiovascular health promotion and disease reduction: the American Heart Association's strategic Impact Goal through 2020 and beyond. Circulation. 2010; 121(4):586-613.10.1161/CIRCULATIONAHA.109.192703.
- Zhou R, Chen HW, Li FR, Zhong Q, Huang YN, Wu XB. Life's Essential 8 Cardiovascular Health and Dementia Risk, Cognition, and Neuroimaging Markers of Brain Health. J Am Med Dir Assoc. 2023; 24(11):1791-1797.10.1016/j.jamda.2023.05.023.
- Tian X, Chen S, Xu Q, Xia X, Zhang Y, Zhang X, Wang P, Wu S, Wang A. Longitudinal cardiovascular health measured by life's essential 8 metrics with incident diabetes: A 13-year prospective cohort study. Diabetes Metab Res Rev. 2023:e3757.10.1002/dmrr.3757.
- Li X, Ma H, Wang X, Feng H, Qi L. Life's Essential 8, Genetic Susceptibility, and Incident Cardiovascular Disease: A Prospective Study. Arterioscler Thromb Vasc Biol. 2023; 43(7):1324-1333.10.1161/atvbaha.123.319290.
- Sun J, Li Y, Zhao M, Yu X, Zhang C, Magnussen CG, Xi B. Association of the American Heart Association's new Life's Essential 8 with all-cause and cardiovascular disease-specific mortality: prospective cohort study. BMC Med. 2023; 21(1):116.10.1186/s12916-023-02824-8.
- Wang X, Ma H, Li X, Heianza Y, Manson JE, Franco OH, Qi L. Association of Cardiovascular Health With Life Expectancy Free of Cardiovascular Disease, Diabetes, Cancer, and Dementia in UK Adults. JAMA Intern Med. 2023; 183(4):340-349.10.1001/jamainternmed.2023.0015.
- Hamczyk MR, Nevado RM, Barettino A, Fuster V, Andrés V. Biological Versus Chronological Aging: JACC Focus Seminar. J Am Coll Cardiol. 2020; 75(8):919-930.10.1016/j.jacc.2019.11.062.
- Zhang R, Wu M, Zhang W, Liu X, Pu J, Wei T, Zhu Z, Tang Z, Wei N, Liu B et al. Association between life's essential 8 and biological ageing among US adults. J Transl Med. 2023; 21(1):622.10.1186/s12967-023-04495-8.
- Kim S, Myers L, Wyckoff J, Cherry KE, Jazwinski SM. The frailty index outperforms DNA methylation age and its derivatives as an indicator of biological age. Geroscience. 2017; 39(1):83-92.10.1007/s11357-017-9960-3.
- Levine ME, Lu AT, Quach A, Chen BH, Assimes TL, Bandinelli S, Hou L, Baccarelli AA, Stewart JD, Li Y et al. An epigenetic biomarker of aging for lifespan and healthspan. Aging (Albany NY). 2018; 10(4):573-591.10.18632/aging.101414.
- Zhu Z, Shi D, Guankai P, Tan Z, Shang X, Hu W, Liao H, Zhang X, Huang Y, Yu H et al. Retinal age gap as a predictive biomarker for mortality risk. Br J Ophthalmol. 2023; 107(4):547-554.10.1136/bjophthalmol-2021-319807.
- Rauseo E, Salih A, Raisi-Estabragh Z, Aung N, Khanderia N, Slabaugh GG, Marshall CR, Neubauer S, Radeva P, Galazzo IB et al. Ischemic Heart Disease and Vascular Risk Factors Are Associated With Accelerated Brain Aging. JACC Cardiovasc Imaging. 2023; 16(7):905-915.10.1016/j.jcmg.2023.01.016.
- Sudlow C, Gallacher J, Allen N, Beral V, Burton P, Danesh J, Downey P, Elliott P, Green J, Landray M et al. UK biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med. 2015; 12(3):e1001779.10.1371/journal.pmed.1001779.
- Said MA, Verweij N, van der Harst P. Associations of Combined Genetic and Lifestyle Risks With Incident Cardiovascular Disease and Diabetes in the UK Biobank Study. JAMA Cardiol. 2018; 3(8):693-702.10.1001/jamacardio.2018.1717.
- Yang G, Cao X, Li X, Zhang J, Ma C, Zhang N, Lu Q, Crimmins EM, Gill TM, Chen X et al. Association of Unhealthy Lifestyle and Childhood Adversity With Acceleration of Aging Among UK Biobank Participants. JAMA Netw Open. 2022; 5(9):e2230690.10.1001/jamanetworkopen.2022.30690.
- Xifra-Porxas A, Ghosh A, Mitsis GD, Boudrias MH. Estimating brain age from structural MRI and MEG data: Insights from dimensionality reduction techniques. Neuroimage. 2021; 231:117822.10.1016/j.neuroimage.2021.117822.
- Gebreab SY, Manna ZG, Khan RJ, Riestra P, Xu R, Davis SK. Less Than Ideal Cardiovascular Health Is Associated With Shorter Leukocyte Telomere Length: The National Health and Nutrition Examination Surveys, 1999–2002. J Am Heart Assoc. 2017; 6(2).10.1161/jaha.116.004105.
- Pottinger TD, Khan SS, Zheng Y, Zhang W, Tindle HA, Allison M, Wells G, Shadyab AH, Nassir R, Martin LW et al. Association of cardiovascular health and epigenetic age acceleration. Clin Epigenetics. 2021; 13(1):42.10.1186/s13148-021-01028-2.
- Lo YH, Lin WY. Cardiovascular health and four epigenetic clocks. Clin Epigenetics. 2022; 14(1):73.10.1186/s13148-022-01295-7.
- Liu W, Wang J, Wang M, Hou H, Ding X, Ma L, Liu M. Oxidative Stress Factors Mediate the Association Between Life's Essential 8 and Accelerated Phenotypic Aging: NHANES 2005–2018. J Gerontol A Biol Sci Med Sci. 2023.10.1093/gerona/glad240.
- Hu W, Wang W, Wang Y, Chen Y, Shang X, Liao H, Huang Y, Bulloch G, Zhang S, Kiburg K et al. Retinal age gap as a predictive biomarker of future risk of Parkinson's disease. Age Ageing. 2022; 51(3).10.1093/ageing/afac062.
- Zhu Z, Liu D, Chen R, Hu W, Liao H, Kiburg K, Ha J, He S, Shang X, Huang Y et al. The Association of Retinal age gap with metabolic syndrome and inflammation. J Diabetes. 2023; 15(3):237-245.10.1111/1753-0407.13364.
- Zhang S, Chen R, Wang Y, Hu W, Kiburg KV, Zhang J, Yang X, Yu H, He M, Wang W et al. Association of Retinal Age Gap and Risk of Kidney Failure: A UK Biobank Study. Am J Kidney Dis. 2023; 81(5):537–544.e531.10.1053/j.ajkd.2022.09.018.
- Zhu Z, Chen Y, Wang W, Wang Y, Hu W, Shang X, Liao H, Shi D, Huang Y, Ha J et al. Association of Retinal Age Gap With Arterial Stiffness and Incident Cardiovascular Disease. Stroke. 2022; 53(11):3320-3328.10.1161/strokeaha.122.038809.
- Chen R, Xu J, Shang X, Bulloch G, He M, Wang W, Zhu Z. Association between cardiovascular health metrics and retinal ageing. Geroscience. 2023; 45(3):1511-1521.10.1007/s11357-023-00743-3.
- Fan J, Yu C, Guo Y, Bian Z, Sun Z, Yang L, Chen Y, Du H, Li Z, Lei Y et al. Frailty index and all-cause and cause-specific mortality in Chinese adults: a prospective cohort study. Lancet Public Health. 2020; 5(12):e650-e660.10.1016/s2468-2667(20)30113-4.
- Graciani A, García-Esquinas E, López-García E, Banegas JR, Rodríguez-Artalejo F. Ideal Cardiovascular Health and Risk of Frailty in Older Adults. Circ Cardiovasc Qual Outcomes. 2016; 9(3):239-245.10.1161/circoutcomes.115.002294.
- Wu J, Xiong Y, Xia X, Orsini N, Qiu C, Kivipelto M, Rizzuto D, Wang R. Can dementia risk be reduced by following the American Heart Association's Life's Simple 7? A systematic review and dose-response meta-analysis. Ageing Res Rev. 2023; 83:101788.10.1016/j.arr.2022.101788.
- Dugan B, Conway J, Duggal NA. Inflammaging as a target for healthy ageing. Age Ageing. 2023; 52(2).10.1093/ageing/afac328.
- Verde Z, Reinoso-Barbero L, Chicharro L, Garatachea N, Resano P, Sánchez-Hernández I, Rodríguez González-Moro JM, Bandrés F, Santiago C, Gómez-Gallego F. Effects of cigarette smoking and nicotine metabolite ratio on leukocyte telomere length. Environ Res. 2015; 140:488-494.10.1016/j.envres.2015.05.008.
- Kivipelto M, Mangialasche F, Ngandu T. Lifestyle interventions to prevent cognitive impairment, dementia and Alzheimer disease. Nat Rev Neurol. 2018; 14(11):653-666.10.1038/s41582-018-0070-3.
- Chia CW, Egan JM, Ferrucci L. Age-Related Changes in Glucose Metabolism, Hyperglycemia, and Cardiovascular Risk. Circ Res. 2018; 123(7):886-904.10.1161/circresaha.118.312806.
- Kulkarni AS, Aleksic S, Berger DM, Sierra F, Kuchel GA, Barzilai N. Geroscience-guided repurposing of FDA-approved drugs to target aging: A proposed process and prioritization. Aging Cell. 2022; 21(4):e13596.10.1111/acel.13596.
- Soares RN, Ramirez-Perez FI, Cabral-Amador FJ, Morales-Quinones M, Foote CA, Ghiarone T, Sharma N, Power G, Smith JA, Rector RS et al. SGLT2 inhibition attenuates arterial dysfunction and decreases vascular F-actin content and expression of proteins associated with oxidative stress in aged mice. Geroscience. 2022; 44(3):1657-1675.10.1007/s11357-022-00563-x.