Association between cardiovascular-kidney-metabolic syndrome and risks of 14 age-related health outcomes in primary prevention older population

Abstract

Background

The American Heart Association (AHA) recently introduced a new clinical entity; the cardiovascular-kidney-metabolic syndrome (CKMS), to promote a multi-disciplinary approach for chronic disease management. This study aims to investigate the relationship between CKMS and major adverse outcomes in older populations in primary care.

Methods

This study utilized data from 18,367 community-dwelling individuals aged ≥ 65, free from prior cardiovascular disease (CVD). Participants were classified into four CKMS stages (stage 0-no CKMS risk factor to stage 3-high CKMS risk) at baseline based on the AHA definition. The association with 14 health outcomes was analysed using multivariable cause-specific hazard models. Additional stratifications were performed by social disadvantage, inflammation levels, and number of CKMS components within each stage to refine risk staging.

Results

Over a median follow-up of 8.6 years, the prevalence of CKMS stages 0 to 3 was 2.8%, 8.2%, 52.6%, and 36.0%, respectively. Compared to stage 0, stage 3 associated with all-cause mortality (HR 1.31, 95%CI 1.01–1.69), CVD mortality (3.11, 1.46–6.60), incident total CVD events (2.78, 1.86–4.15), myocardial infarction (2.37, 1.21–4.62), stroke (2.75, 1.50–5.04), heart failure hospitalization (4.56, 1.45–14.37), atrial fibrillation (2.63, 1.61–4.29), cancer (1.31, 1.04–1.64), depression (1.20, 1.04–1.39), and physical disability (1.92, 1.20–3.08). Social disadvantage, high inflammation, and CKMS component count further amplified these associations. No linear trend was observed for cognitive outcomes and non-CVD cause-specific mortality.

Conclusion

This study is the first to establish relationships between CKMS and multiple age-related outcomes in older adults, providing critical insights for developing holistic approaches to primary care in the aged.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s11357-025-02058-x.

Introduction

Cardiovascular-kidney-metabolic syndrome (CKMS) is a novel disease classification introduced by the American Heart Association (AHA) in 2023 defining an intricate interconnection between obesity, diabetes, chronic kidney disease (CKD), and cardiovascular disease (CVD) [1]. The recognition of the shared risk factors, pathophysiological mechanisms, and complications between these diseeases has provided a refreshed insight in leveraging a holistic approach for chronic disease management. The prevalence of CKMS in the general population is high. An estimated 90% of U.S. adults met the criteria for CKMS stage 1 (only excess or dysfunctional adiposity) or more advanced stages between 2011–2020, of which 15% were at stage 3 (subclinical CVD) or 4 (clinical CVD events) [2]. Older adults exhibit a disproportionate burden. Almost all (99%) are affected and more than half (55%) met at least stage 3 [2]. CKMS was responsible for 29% of total deaths (approximately a million deaths per year) in the U.S. in 2022 [3, 4].

The clustering of CVD, CKD, and metabolic disorders significantly increases the likelihood of not just CVD and CKD progression but other adverse outcomes, including hospitalisation, disability, and premature death, particularly in older population due to their advancing age, presence of multimorbidity and polypharmacy [5–7]. Recent studies found that a stepwise relationship between CKMS stages and all-cause mortality [8, 9]. However, whether the severity of CKMS are associated with a broader spectrum of age-related diseases has been largely under-studied. As older people are most affected and are at the highest risk of morbidities and mortality, it is crucial to investigate whether CKMS is useful in sorting out older people who had no prior CVD events and other major disability or morbidities are at heightened risk of major age-related health outcomes thus beneficial from a more aggressive preventive treatment. For instance, the co-existing CVD and CKD may have an additive effect on increasing the risk of cancer, dementia, depression, and physical disability, due to shared risk factors and pathophysiological pathways, including chronic inflammation, metabolic dysregulation, oxidative stress and endothelial dysfunction.

In this study we investigated the relationship between stages of CKMS with both CVD and non-CVD health outcomes in healthy older adults. Since most participants were presumed to be at least in stage 1, thereby labelled as having CKMS, we also examined key risk enhancers for further risk stratification.

Methods

Data source and study population

We used data from participants recruited in the ASPREE (ASPirin in Reducing Events in the Elderly, NCT01038583) [10–13] clinical trial and its observational follow-on study, ASPREE-XT. Briefly, ASPREE was a multi-centre, bi-national, double-blinded, randomised clinical trial with the primary aim to evaluate the efficacy of 100 mg daily aspirin versus matching placebo to prolong disability-free survival. Between March 2010 and December 2014, 19,114 community-dwelling individuals ≥ 70 years of age (≥ 65 for U.S. minorities) were recruited, with the majority residing in Australia (87.4%) and others in the U.S. (12.6%). Participants had no history of CVD events, atrial fibrillation [14], diagnosed dementia, or independence-limiting physical disability at trial entry. Other exclusion criteria included a serious illness likely to cause death within the next five years (e.g., certain aggressive cancers such as pancreatic or brain cancer, end-stage CKD); systolic blood pressure (SBP) ≥ 180 mm Hg or diastolic blood pressure (DBP) ≥ 105 mm Hg; a modified Mini-Mental State Examination (3MS) score ≤ 77/100. The intervention phase of the ASPREE trial ended in June 2017. ASPREE-XT is an ongoing post-trial observational study following those (> 80%) who consented to be followed for a further five years [15]. For the present analysis, we used data from both ASPREE and ASPREE-XT (data cut at 2021) with a maximum follow-up of 11 years. The present analysis included all participants with sufficient data to ascertain their CKMS stage, as well as data on key covariates. A total of 4.5% participants missing urinary albumin-to-creatinine data were assumed not to have albuminuria, and 37% participants missing high sensitivity(hs)-Troponin I and NT-pro NBP (only measured in Australian participants) were assumed not having subclinical heart failure (HF). No imputation was performed as these data were not missing at random.

Study exposure

We defined CKMS presence and its stage (ASPREE participants were free of CVD events thus stage 4 was omitted here) according to the AHA statement as below [1]. Details can be found in Supplementary Methods.

Baseline data measurement

At baseline visit, participants were asked to fast overnight, and their blood samples were collected in a local clinic or pathology centre the following day for fasting lipid and blood glucose measurements. Urine samples were collected for the measurement of urinary albumin and creatinine. The eGFR was calculated using the CKD-Epidemiology Collaboration 2021 Equation. Baseline characteristics on demographics, self-reported medical history and family history, and socioeconomic factors were collected through questionnaires. Height, weight and waist circumference, and blood pressure were measured at the baseline visit. Current medication information was gathered either from primary care physician records by research staff, through review of medications brought in to study visits, or through self-reports of current medication use.

Study endpoints

Study endpoints of interest including:

1. CVD outcomes [13, 16]: incident total CVD events (defined as a composite of myocardial infarction [MI], stroke, and hospitalisation for HF [HHF]), MI, stroke (all types), and HHF. Atrial fibrillation [14] were only included, noting this outcome was recorded during ASPREE in-trial period;

2. Non-CVD outcomes: all-cause and cause-specific mortality [10] (CVD, cancer, non-CVD/non-cancer), dementia and cognitive impairment without no dementia (CIND) [17, 18], cancer [19], depressive symptoms (CES-D 10 score ≥ 8/30) [20], and persistent physical disability [12].

All outcomes were adjudicated by expert panels, except for depression were ascertained via CESD-10 questionaires, persistent physical disability via LIFE disability questionnaires and CIND via cognitive tests.

Statistical analysis

Analyses were performed to compare the incidence of outcome across CKMS stages. Potential confounding factors contributing to an individual's susceptibility to CKMS and study outcomes were identified from existing literature and opinions from clinicians. These included baseline age, sex, race/ethnicity, country, smoking status (never/former/current), alcohol consumption, ASPREE randomised treatment, family history of CVD, personal history of cancer (treated), baseline 3MS score, Fried frailty status, and the baseline Centre for Epidemiologic Studies Depression Scale (CES-D) 10-item questionnaire score.

Cause-specific hazards models were employed for outcome analysis (Cox regression for all-cause mortality) to account for competing risk of death by treating competing events as censored, in which baseline hazards were estimated using restricted cubic splines with 4 knots placed at the 5th, 35th, 65th, and 95th percentiles. Adjusted hazard ratios (HRs) with 95% confidence intervals (CIs) were obtained from these models for each outcome. P-values for linear trend between CKMS stages were cacluated. Cumulative incidence curves were plotted using the cumulative incidence function to account for competing mortality. Population attributable fractions (PAF) with 95% CIs were calculated using Miettinen’s formula (PAF = P_e (HR-1)/HR where P_e is the prevalence of the CKMS of our cohort) to measure the risk of each outcome attributable to the presence of CKMS stage from 1 to 3 vs 0 following the survival model [21]. A PAF of 100% means the exposure is entirely responsible for the outcome of interest in the population.

As most participants fall under the CKMS stages 2 and 3, further sub-stratification was conducted to refine risk staging. Participants in stage 2 were categorised into three subgroups: 1) meet 1 criterion; 2) meet 2 criteria; or 3) meet 3 or more criteria, based on the 5 criteria defining stage 2. Stage 3 participants were stratified into having a) either high CVD or high CKD risk, or b) both.

Further risk stratification was further conducted by socioeconomic status, and background inflammation status (measured by baseline hs-CRP of 2.0 mmol/L). Socioeconomic status was measured using the Index of Relative Socio-economic Advantage and Disadvantage (IRSAD) [22]. Cause-specific HRs were estimated for each sub-stratum using the same models as previously.

Subgroup analyses were performed for sex, BMI categories (normal weight vs. over-weight/obese, and FBG (normoglycaemia vs. hyperglycaemia), with the latter two using stage 1 as the reference group.

Sensitivity analysis

In the main analysis, we applied the PREVENT™ equation to all participants, including those aged 80 years or older, among whom the equation was not validated. This was based on our previous data demonstrating excellent PREVENT performance in the 80 + year age group [23]. We conducted a sensitivity analysis restricting the analysis to participants aged 79 years and under.

All analyses were performed using Stata 17.0 (StataCorp, College Station, Texas). Forest plots were made using R with the forestplot command (R Foundation for Statistical Computing), version 1.1.456. All statistical tests were 2-sided, with p values < 0.05 considered to indicate statistical significance.

Results

Baseline characteristics

After excluding 99 participants with a BMI < 18.5 kg/m² and 648 with insufficient data to ascertain CKM stage (FBG: 209, eGFR: 242, lipids: 165, BMI: 10, waist circumference: 12), 18,367 participants were included in the final cohort. The median (IQR) age was 74.0 (71.6–77.7) years, and 56.6% (n = 10,392) were female. The prevalence across CKMS stages 0 to 3 was 2.8%, 8.2%, 52.6%, and 36.0%, respectively. Participants in stage 3 were older and were more likely to be White compared to those in less advanced stages. Sex distribution was similar between stage 0 and stage 3. Participants in more advanced stages also had lower education levels and higher frailty status. Among participants in stage 2, hypertension was most prevalent (88.3%), followed by metabolic syndrome (41.9%), hypertriglyceridemia (34.5%) moderate- to high-risk CKD (24.9%), and diabetes (12.8%). Among stage 2 participants, 41.7% met the criteria for one item, 26.1% for two items, and 32.2% for three or more items. Among participants at stage 3, 9% had high or very high CKD risk, 16.2% had a predicted 10-year ASCVD risk score of ≥ 20%, and 89.1% had subclinical HF (elevated NT-pro BNP and/or hs-troponin I); only 1.7% (n = 111) had both CKD and subclinical CVD (subclinical ASCVD and/or subclinical HF). (Supplementary Table 1).

Main outcomes

The overall median (IQR) follow-up was 8.6 (7.4–10.1) years. AF had a median (IQR) follow-up of 4.5 (3.3–5.6) years. Cumulative incidence plots for each individual study outcome by CKM stage are presented in Supplementary Fig. 1. For CVD outcomes, higher CKMS stages were associated with increased risk of incident total CVD events, MI, and stroke; For non-CVD outcomes, higher CKMS stages were associated with increased risk of all-cause mortality, CVD mortality, cancer, depression, and persistent physical disability (all p-values for linear trend < 0.05). Compared with CKMS stage 0 (participants with no CKM risk factors), the HR for stage 3 was 2.78 (1.86–4.15) for total CVD events, 2.37 (1.21–4.62) for MI, 2.75 (1.50–5.04) for stroke, 1.31 (1.01–1.69) for all-cause mortality, 3.11 (1.46–6.60) for CVD mortality, 1.31 (1.04–1.64) for cancer, 1.20 (1.04–1.39) for depression, and 1.92 (1.20–3.08) for persistent physical disability. (Table 1).

Fig. 1.

Population attributable fractions of each soutcome attributable to each CKMS stage. Legend: Population attributable fractions (PAFs) for each study endpoint was calculated by using CKM stage 0 as an unexposed group and more advanced stage as an exposure group. The PAFs were highest for CVD outcomes including AF (all over 50%) attributable to CKM stage 3. Abbreviations: AF, atrial fibrillation; CIND: cognitive impairment no dementia; CKMS, cardiovascular kidney metabolic syndrome; CVD, cardiovascular disease; HHF, hospitalisation for heart failure; MI, myocardial infarction

Table 1.

Association between CKMS stages and risk of multiple health outcomes among initially healthy older adults

	Stage 0(n=509, 2.8%)	Stage 1(n=1508, 8.2%)	Stage 2(n=9655, 52.6%)	Stage 3(n=6614, 36.0%)	P for trend	PAF
Total CVD events
Event (Incidence rate)	25 (5.2)	99 (8.3)	820 (10.9)	921 (18.2)	-	54.2% (32.5% to 68.9%)
Unadjusted HR (95% CI)	1.00 (Ref)	1.60 (1.03-2.49)	2.12 (1.42-3.16)	3.56 (2.39-5.30)	<0.001
Adjusted HR (95% CI)	1.00 (Ref)	1.65 (1.07-2.56)	1.99 (1.33-2.96)	2.78 (1.86-4.15)	<0.001
Myocardial infarction
Event (Incidence rate)	9 (1.9)	38 (3.1)	306 (4.0)	291 (5.6)	-	50.7% (5.7% to 74.2%)
Unadjusted HR (95% CI)	1.00 (Ref)	1.69 (0.82-3.50)	2.16 (1.12-4.20)	3.03 (1.56-5.88)	<0.001
Adjusted HR (95% CI)	1.00 (Ref)	1.65 (0.80-3.41)	1.94 (1.00-3.77)	2.37 (1.21-4.62)	<0.001
Stroke
Event (Incidence rate)	11 (2.3)	45 (3.7)	346 (4.5)	389 (7.5)	-	53.8% (16.5% to 74.4%)
Unadjusted HR (95% CI)	1.00 (Ref)	1.65 (0.85-3.18)	2.01 (1.10-3.66)	3.35 (1.84-6.10)	<0.001
Adjusted HR (95% CI)	1.00 (Ref)	1.73 (0.90-3.35)	1.97 (1.08-3.60)	2.75 (1.50-5.04)	<0.001
Hospitalisation for HF
Event (Incidence rate)	3 (0.6)	7 (0.6)	138 (1.8)	211 (4.0)	-	67.9% (0.4% to 89.7%)
Unadjusted HR (95% CI)	1.00 (Ref)	0.94 (0.24-3.65)	2.95 (0.94-9.27)	6.74 (2.15-21.05)	<0.001
Adjusted HR (95% CI)	1.00 (Ref)	0.97 (0.25-3.77)	2.72 (0.86-8.55)	4.56 (1.45-14.37)	<0.001
AF (in-trial period only) a
Event (Incidence rate)	17 (6.5)	35 (5.4)	379 (9.3)	522 (19.5)	-	42.1% (7.3% to 63.9%)
Unadjusted HR (95% CI)	1.00 (Ref)	0.84 (0.47-1.51)	1.45 (0.89-2.36)	3.08 (1.90-4.99)	<0.001
Adjusted HR (95% CI)	1.00 (Ref)	0.86 (0.48-1.53)	1.43 (0.88-2.33)	2.63 (1.61-4.29)	<0.001
All-cause mortality
Event (Incidence rate)	63 (12.3)	161 (12.6)	1289 (15.8)	1373 (24.7)	-	17.7% (-4.7% to 35.2%)
Unadjusted HR (95% CI)	1.00 (Ref)	1.03 (0.77-1.38)	1.30 (1.01-1.67)	2.05 (1.59-2.64)	<0.001
Adjusted HR (95%CI)	1.00 (Ref)	1.09 (0.81-1.46)	1.19 (0.92-1.53)	1.31 (1.01-1.69)	0.002
CVD mortality
Event (Incidence rate)	7 (1.4)	34 (2.7)	272 (3.3)	382 (6.9)	-	60.2% (17.2% to 80.9%)
Unadjusted HR (95% CI)	1.00 (Ref)	1.97 (0.87-4.45)	2.47 (1.17-5.24)	5.17 (2.45-10.93)	<0.001
Adjusted HR (95%CI)	1.00 (Ref)	2.09 (0.92-4.71)	2.24 (1.06-4.76)	3.11 (1.46-6.60)	<0.001
Cancer mortality
Event (Incidence rate)	28 (5.5)	67 (5.2)	585 (7.2)	511 (9.2)	-	13.5% (-24.5% to 39.9%)
Unadjusted HR (95% CI)	1.00 (Ref)	0.96 (0.62-1.49)	1.31 (0.90-1.92)	1.70 (1.16-2.48)	<0.001
Adjusted HR (95% CI)	1.00 (Ref)	0.99 (0.63-1.53)	1.20 (0.82-1.75)	1.15 (0.78-1.69)	0.46
Non-CVD/-cancer mortality
Event (Incidence rate)	28 (5.5)	60 (4.7)	432 (5.3)	480 (8.7)	-	-6.0% (-53.6% to 26.8%)
Unadjusted HR (95% CI)	1.00 (Ref)	0.87 (0.56-1.37)	0.98 (0.67-1.44)	1.63 (1.11-2.38)	<0.001
Adjusted HR (95% CI)	1.00 (Ref)	0.94 (0.60-1.47)	0.91 (0.62-1.33)	1.00 (0.68-1.47)	0.50
	Stage 0 (n=509)	Stage 1 (n=1508)	Stage 2 (n=9655)	Stage 3 (n=6614)	P for trend	PAF
Cancer
Event (Incidence rate)	82 (17.9)	242 (21.2)	1681 (23.4)	1306 (26.6)		20.1% (1.0% to 35.5%)
Unadjusted HR (95% CI)	1.00 (Ref)	1.18 (0.92-1.52)	1.31 (1.05-1.63)	1.49 (1.19-1.86)	<0.001
Adjusted HR (95% CI)	1.00 (Ref)	1.18 (0.92-1.51)	1.25 (1.00-1.56)	1.31 (1.04-1.64)	0.012
CIND b
Event (Incidence rate)	152 (37.2)	350 (34.5)	2356 (38.0)	1958 (46.0)		-3.4% (-21.2% to 11.8%)
Unadjusted HR (95% CI)	1.00 (Ref)	0.95 (0.78-1.15)	1.07 (0.91-1.26)	1.35 (1.15-1.59)	<0.001
Adjusted HR (95% CI)	1.00 (Ref)	0.94 (0.77-1.13)	0.96 (0.82-1.13)	1.00 (0.84-1.18)	0.37
Dementia
Event (Incidence rate)	48 (10.1)	86 (7.2)	527 (6.9)	507 (9.8)		-60.0% (-112.4% to -20.5%)
Unadjusted HR (95% CI)	1.00 (Ref)	0.71 (0.50-1.02)	0.69 (0.52-0.93)	0.99 (0.74-1.33)	<0.001
Adjusted HR (95% CI)	1.00 (Ref)	0.72 (0.50-1.02)	0.59 (0.44-0.79)	0.64 (0.47-0.87)	0.09
Depression c
Event (Incidence rate)	200 (56.4)	521 (60.0)	3436 (63.3)	2586 (72.3)		11.7% (-1.6% to 23.3%)
Unadjusted HR (95% CI)	1.00 (Ref)	1.07 (0.91-1.26)	1.12 (0.97-1.30)	1.28 (1.11-1.48)	<0.001
Adjusted HR (95% CI)	1.00 (Ref)	1.10 (0.93-1.29)	1.11 (0.96-1.28)	1.20 (1.04-1.39)	<0.001
Persistent physical disability
Event (Incidence rate)	18 (3.8)	56 (4.7)	554 (7.5)	601 (12.0)		42.2% (8.7% to 63.4%)
Unadjusted HR (95% CI)	1.00 (Ref)	1.27 (0.75-2.16)	2.01 (1.26-3.22)	3.29 (2.06-5.25)	<0.001
Adjusted HR (95% CI)	1.00 (Ref)	1.29 (0.76-2.19)	1.70 (1.06-2.72)	1.92 (1.20-3.08)	<0.001

^athe median (IQR) follow up was 4.5 (3.3-5.6) years in the analysis of atrial fibrillation (AF). ^bA total of 982 participants with baseline cognitive score<1.5 SD of the total population were excluded from this analysis. ^c1284 with baseline CESD score of 8 or higher were excluded. Abbreviation’s: CIND, cognitive impairment – no dementia; CKMS, cardiovascular-kidney-metabolic syndrome; CVD, cardiovascular disease; CI, confidence interval; HF, heart failure; HR, hazard ratio; PAF, population-attributable fraction

Participants at CKMS stages 2 and 3 exhibited higher risks of HHF and AF compared to those at stage 0 (HR for stage 2 and 3 versus 0: 2.72 and 4.56 for HHF; 1.43 and 2.63 for AF) (p for linear trend < 0.001). However, risks of these outcomes were not higher for stage 1 compared to stage 0. (Table 1).

No significant association was found between CKMS stage and CIND (HR for stages 1 to 3 compared to stage 0: 0.94, 0.96, and 1.00, respectively). Interestingly, participants at higher CKMS stages demonstrated a consistently lower risk of dementia compared to those at stage 0, despite a lack of significant linear trend (Table 1).

Regarding PAF, the presence of CKMS (stage 1—3) was associated with the greatest fraction of attributable cases of CVD outcomes (HHF: 67.9%, CVD mortality: 60%, total CVD: 54%, stroke: 54%, MI: 51%, AF: 42%). This was followed by persistent physical disability (42%), cancer (20%) and depression (12%).(Table 1) Furthermore, there was a stepwise increase in PAFs for these outcomes attributable to a higher CKMS stage. The PAF attributable to CKMS stage 3 was 79% for HHF, 64% for CVD mortality, 61% for total CVD event and AF, 56% for MI and stroke, 48% for persistent physical disability, 25% for cancer, and 17% for depression (Fig. 1).

In sex-stratified analysis, CKMS stage 3 appears to be associated with higher risks of all CVD-related outcomes in males compared with females, although the interaction between sex and CKMS stages was only significant for all-cause mortality, CVD mortality, and AF. CKMS stage was linked to an increased risk of depression and non-CVD/non-cancer mortality in females but not males (p for interaction = 0.02 and 0.05, respectively) (Supplementary Table 2). Subgroup analysis by BMI and FBG categories found a generally higher risk of CVD outcomes in participants who were overweight/obese or who had hyperglycaemia, despite the lack of significant interaction likely due to limited analytic power. (Supplementary Table 3, Supplementary Table 4).

Sub-stratified analysis by the number of criteria met within CKMS stage 2 and 3

Further risk stratification within stage 2 and 3 by the number of criteria met, found that those meeting more criteria exhibited a greater risk for all outcomes which were observed to have a stepwise relationship with CKMS stage in the main analyses (p for trend < 0.05). Compared with stage 0, the HRs for all-cause mortality progressively increased across the strata: stage 1 (HR 1.09), stage 2 (meeting 1 criterion: HR 1.12; meeting 2 criteria: HR 1.19; meeting 3 or more criteria: HR 1.23), stage 3 (high CVD or CKD risk: HR 1.29; both: HR 2.08). A similar pattern was observed for total CVD events, with HRs of 1.66, 1.62, 1.86, 2.34, 2.79, and 4.18, respectively, across the same strata. (Fig. 2).

Fig. 2.

Association between CKMS stages and incident risks of study outcomes of interest. Legend: Stage 2 encompasses five criteria: (1) Hypertriglyceridemia: TG ≥125mg/dL; (2) Hypertension: BP>140/90 mm/Hg or use anti-hypertensive medications; [31] Metabolic syndrome; (4) Diabetes: Self-report, use of glucose-lowering medications, or FBG ≥126 mg/dL; or (5) moderate-to-high risk CKD according to KDIGO criteria. In stage 3, high CVD risk refers to (1) high predicted 10-year CVD risk (PREVENT10 score ≥20%), and/or subclinical heart failure (HF) diagnosed by elevated cardiac biomarkers (N-terminal pro-B-type natriuretic peptide (NT-proBNP) ≥125pg/mL, hs- troponin I ≥10 ng/L for women and ≥12 ng/L for men. High CKD risk refers to stage G4 or G5 CKD or very high risk as per the KDIGO classification. Abbreviations: CIND: cognitive impairment no dementia; CKM, cardiovascular kidney metabolic syndrome; CVD, cardiovascular disease.

Sub-stratified analysis by social determinant of health

In general, participants who were more socially disadvantaged had a greater risk of the most outcomes including all-cause mortality and cause-specific mortality, total CVD events, MI, stroke (except for stage 3), cancer (expect for stage 3), depression, and persistent physical disability. (Fig. 3).

Fig. 3.

Association between CKMS stages and health outcomes by social determinants of health (SDOH) index among initially healthy older adults. Legend: Participants were classified into ‘less disadvantaged’ group and ‘more disadvantaged’ group based on the median value of the Index of Relative Socio-economic Advantage and Disadvantage (IRSAD) score. Abbreviations: AF, atrial fibrillation; CIND: cognitive impairment no dementia; CKM, cardiovascular kidney metabolic syndrome; CVD, cardiovascular disease; HHF, hospitalisation for heart failure; MI, myocardial infarction

Sub-stratified analysis by background inflammation levels

Participants with a high baseline inflammation status were at a greater risk of all CVD outcomes except for AF, all-cause and cause-specific mortality, cancer, depression and persistent physical disability, compared to those at the same CKMS stage with baseline low inflammation levels. (Fig. 4).

Fig. 4.

Association between CKMS stages and health outcomes by CRP levels (hs-CRP =2mg/L) among initially healthy older adults. Legend: Abbreviations: CIND: cognitive impairment no dementia; CKM, cardiovascular kidney metabolic syndrome; CVD, cardiovascular disease

Sensitivity analysis

Repeating the analyses excluding participants aged 80 and older, among whom the PREVENT™ equation was not validated, yielded similar results as our main analysis. (Supplementary Table 5).

Discussion

In this longitudinal cohort study of older adults who were free of CVD events, dementia, and dependence-limiting physical disability initially, we found a stepwise relationship between higher CKM stage and greater incidence of multiple adverse health outcomes including all CVD outcomes of interest, all-cause and CVD mortality, cancer, depression and persistent physical disability, in up to 11-years of follow-up. No linear trend was observed for CIND, dementia, and non-CVD cause-specific mortality. The association with CVD outcomes was more pronounced in males, whereas the association with depression and non-CVD/non-cancer mortality was more pronounced in females, likely attributable to the differing background risks of these outcomes between the sexes. The CKMS-related risk of CVD outcomes was also higher in those who were overweight/obese or had hyperglyceridemia compared to their counterparts. Furthermore, we found that the risk of CKMS-related health outcomes is greater in those who had baseline lower socio-economic status and greater inflammation levels.

Comparison with existing evidence

Only one study have been published investigating the relationship between CKMS syndrome and risk of adverse health outcomes in older populations exclusively. A US population-based study found that among adullts aged between 70 to 79 years, CKMS stages 3–4 was associated with 50% (95% CI 1.20–1.86) increased risk of dementia compared with CKMS stage 0–2 [24]. An UK Biobank study also found a higher risk of dmentia in participants in CKMS stages 2–4 compared with stage 0 [25]. Another UK Biobank study found a stepwise relationship between CKMS stages and depression/anxiety [26]. A China Health and Retirement Longitudinal Study (CHARLS) found that among middle-aged and older adults in China (median age: 60 years), CKMS stages 2–3 was associated with 16% increased risk of ADL disability compared with CKMS stage 0 [27]. Older adults are more vulnerable to both CKMS syndrome and age-related morbidities compared to their younger counterparts. In contrast to most previous studies that included a broad age range, mixed primary and secondary population, and focused on a single endpoint, our study specfially targeted older adults in a primary prevention setting. This population may particularly benefit from more intensive preventive interventions if CKMS staging proves effective in identifying those at heightened risk for major age-related health outcomes.

CVD, CKD, and metabolic diseases are highly prevalent in older adults and frequently intertwined with other age-related chronic conditions. Our study is the first to establish a relationship between the severity of the CKMS and a broad spectum of age-related comorbidities. The PAF data demonstrate CKMS's significant contribution to both CVD morbidity and mortality, as well as non-CVD outcomes. Specifically, CKMS stage 3 accounted for 56% to 79% of various CVD subtypes, 48% of persistent physical disability, 25% of cancer, 22% of all-cause mortality, and 17% of depression symptoms. These findings suggest that CKMS not only affects cardiovascular and kidney health, but also likely contributes to other diseases that share common risk factors or underlying pathophysiological mechanisms. Additionally, the presence of one chronic condition often accelerates the onset and progression of others, creating a complex cycle of health deterioration in older adults, marked by systemic inflammation, progressive multi-organ dysfunction, and disrupted environmental homeostasis, including alterations in hemodynamic and metabolic processes. We also found that individuals meeting more criteria within the same CKMS stage also had a higher risk of CKMS-related outcomes, highlighting that both the overall CKMS stage and the specific number of CKMS components within each stage should be considered to inform tailored and proactive interventions in daily care.

In our study, we report a consistently lower risk of dementia from stage 1 to 3 when compared to stage 0. Our findings regarding dementia outcomes align with previous ASPREE studies, where we found a lower dementia risk in individuals with overweight/obesity, large waist circumference [28], low HDL-c [29], or high triglycerides [30]. These results suggest that many metabolic or vascular factors may exert complex, paradoxical effects on late-life cognition. While adverse in midlife, some factors, such as higher BMI or lipid levels, may reflect preserved metabolic reserves in older adults. Another possible explanation is the survival advantage in dementia patients, as dementia typically occurs in the later stages of life may compete with other conditions more prevalent in the earlier stages of later life. Reverse causality also cannot be ruled out. Cognitive decline and neurodegenerative processes can precede dementia diagnosis by many years and may influence metabolic regulation, body composition, and cardiovascular function, thereby confounding the directionality of association. Furthermore, participants with lower baseline cognitive performance were excluded from ASPREE, resulting in a cohort with preserved cognition at enrolment, which could have attenuated any true association between CKMS severity and cognitive outcomes.

The management of CKMS requires an integrated approach addressing the shared cardiovascular, kidney, and metabolic pathways highlighted in the recent AHA Presidential Advisory. [1] Multidomain interventions incorporating both lifestyle modification and pharmacological management targeting the underlying risk factors are likely to be most effective. Agents including renin-angiotensin system inhibitors, statins, SGLT2 inhibitors, and GLP-1 receptor agonists have demonstrated pleiotropic benefits on multiple CKM components. In older adults, personalised CKMS management should ideally occur within a multidisciplinary primary care framework that integrates cardiovascular and metabolic risk monitoring, renal function assessment, and mental health support. The recent presidential advisory from the AHA CKMS Health Scientific Advisory Group also highlighted the importance of addressing the impact of social determinants of health on the burden of CKMS and the need to incorporate it into CKMS care [1]. As expected, we found the multifaceted association with CKMS was much more pronounced in individuals who are more socio-economically disadvantaged, highlighting the importance of addressing equity in care. Other risk enhancers (e.g., high inflammation status, depression, sleep disorders, sex-specific risk enhancers such as premature menopause, family history of diabetes and kidney failure) were also highlighted by AHA. In our study, a risk enhancement by high hs-CRP levels was observed.

Strengths and limitations

The main strengths of this study lie in the high-quality, prospective data derived from community-dwelling older adults enrolled in the ASPREE trial and its extended observational phase, ASPREE-XT. Data of ASPREE participants were rigorously and systematically collected at baseline and through annual/semi-annual follow-up visits by well trained staff, encompasses a comprehensive range of information that is crucial for robust analysis. With a median follow-up period of 8.6 years, the study captures the occurrence of multiple outcomes of interest, and the sex distribution among participants is well-balanced. All study endpoints were pre-defined in the protocol, and most were adjudicated by clinicians with relevant expertise.

Major limitations include the lack of data on echocardiography, coronary angiography, and cardiac CT, which would lead to a proportion of stage 3 participants misclassified into lower stages. This misclassification would likely bias our HRs and PAFs toward the null, suggesting that the true disease burden of advanced CKMS is likely underestimated. When calculating the PAF, we relied on the prevalence of CKMS from our study cohort. However, this may not accurately reflect the true prevalence in the broader target population due to selection bias inherent in using trial data. Notably, only one published study to date has reported CKMS prevalence in older adults, but its sample size is half that of our cohort, and it lacks cardiac biomarker data, which likely leads to an underestimation of the proportion of individuals in CKMS stage 3 [2]. Our study is observational, exploring the relationship between CKMS stage and the risk of other diseases. It remains unclear whether managing CKMS can reduce the risk of these diseases. Additionally, since the majority of ASPREE participants were white, the findings may not be fully applicable to other ethnic groups, limiting the generalizability of our results.

Conclusion

Our study underscores the complex, multifaceted nature of CKMS and its potential to influence a broad spectrum of age-related comorbidities. Future research is essential to evaluate the effectiveness of pharmacological interventions in managing CKMS and mitigating its impact on various health outcomes, not only in older adults but also across different age groups.

Supplementary Information

Below is the link to the electronic supplementary material.

Authors’ contributions

Conceptualisation and study design: All authors; Formal analysis: Zhou Z, Rory Wolfe; Methodology: Zhou Z, Ryan J; Supervision: Ryan J; Writing – original draft: Zhou Z; Writing – review & editing: All authors.

Funding

None.

Data availability

Anonymised data not published within this article will be made available by request from any qualified investigator. Requests for data access will be via the ASPREE Principal Investigators with details for applications provided through https://aspree.org/aus/for-researchers/ or https://aspree.org/usa/for-researchers/.

Declarations

Ethics approval

This post-hoc analysis is covered under existing ethics approvals for secondary analysis of ASPREE and ASPREE-XT study approved by the Monash University Human Research Ethics Committee (ID 25829).

Competing interest

Tonkin AM. reported receiving research support or honoraria from Amgen, Merck, Novartis and Pfizer, as well as materials in the ASPREE trial from Bayer. Nelson MR. reported receiving speaker fees from Medtronic, honoraria from Sanofi and Amgen as well as Bayer for materials in ASPREE. Zoungas S. has received NHMRC and Australian Heart Foundation research funding as the principal investigator of the STAREE trial; and has received payment to the institution (Monash University) from Eli Lilly Australia, Boehringer-Ingelheim, Merck Sharp & Dohme Australia, AstraZeneca, Novo Nordisk, Sanofi, and Servier for consultancy work outside the submitted work. Shah RC. is a member of the steering committee of the PREVENTABLE clinical trial. Others declare that they have no conflict of interests.