Association between cardiovascular-kidney-metabolic syndrome, lifestyle, and all-cause and cause-specific mortality: a prospective cohort study

Abstract

<h3>Background</h3><p>Cardiovascular-kidney-metabolic (CKM) syndrome is reported to be associated with increased all-cause and CVD-specific mortality. However, the association between CKM and cancer-specific mortality, as well as the modifying or joint effects of healthy lifestyle on mortality remain unclear.</p><h3>Methods</h3><p>In this prospective cohort study, we enrolled participants aged 37–73 years from the UK Biobank. Individuals with missing data on CKM status, lifestyle factors or potential confounders at baseline (between March 2006 and July 2010) were excluded. Followed-up was conducted until November 30, 2022. CKM stages (0-4) were defined per the American Heart Association criteria. A healthy lifestyle score (adequate physical activity, no current smoking, healthy sleep, and healthy diet) was categorized into unfavourable (0–1), intermediate (2–3), and favourable (4) groups. Outcomes included all-cause and cause-specific mortality (due to cardiovascular disease [CVD], cancer, and other causes). Cox proportional hazards models and Fine–Gray proportional subdistribution hazards models were used to assess associations with all-cause and cause-specific mortality, respectively.</p><h3>Findings</h3><p>A total of 319,291 participants were included. Over a median follow-up of 13.7 years (interquartile ranges 13.0–14.3), 27,267 deaths occurred (5558 CVDs [20.4%], 13,566 cancers [49.8%], 8143 other-causes [29.8%]). Compared with CKM Stage 0, Stages 2–4 were associated with progressively higher risks of all-cause mortality (hazard ratios [HRs] (95% confidence intervals [CIs]): Stage 2: 1.21 [1.12–1.30]; Stage 3: 1.54 [1.43–1.66]; Stage 4: 2.30 [2.13–2.49]), CVD-specific mortality (Stage 2: 2.38 [1.74–3.24]; Stage 3: 4.46 [3.23–6.14]; Stage 4: 10.40 [7.61–14.21]), and cancer-specific mortality (Stage 2: 1.15 [1.03–1.28]; Stage 3: 1.26 [1.12–1.41]; Stage 4: 1.32 [1.17–1.48]). In addition, Stages 3 and 4 were positively associated with other-cause mortality (Stage 3: 1.49 [1.29–1.72]; Stage 4: 2.08 [1.81–2.38]) (all <em>P</em>-trend <0.0001). Additionally, these associations were more pronounced in adults aged <60 years compared to those ≥60 years (<em>P</em>-interaction <0.0001). Significant CKM-lifestyle interactions were found for associations with all-cause (<em>P</em>-interaction = 0.021), cancer-specific (<em>P</em>-interaction = 0.021), and other-cause mortality (<em>P</em>-interaction = 0.0031), but not for CVD-specific mortality (<em>P</em>-interaction = 0.33). A favourable or intermediate lifestyle was associated with reduced all-cause and cause-specific mortality across all CKM stages, with substantial benefits observed for non-smoking, adequate physical activity, and healthy sleep duration. For all-cause mortality, HRs (95% CIs) for a favourable vs. unfavourable lifestyle were: Stage 0: 0.51 (0.38–0.68); Stage 1: 0.39 (0.28–0.55); Stage 2: 0.60 (0.55–0.66); Stage 3: 0.57 (0.52–0.62); Stage 4: 0.53 (0.47–0.59). For CVD-specific mortality, corresponding HRs (95% CIs) across CKM Stages were 0.56 (0.14–2.30), 0.59 (0.18–1.90), 0.60 (0.46–0.79), 0.74 (0.63–0.87), and 0.65 (0.53–0.81), respectively. For cancer-specific mortality, the values were 0.61 (0.44–0.85), 0.45 (0.28–0.73), 0.67 (0.59–0.76), 0.54 (0.47–0.61), and 0.61 (0.49–0.77), respectively. For other-cause mortality, the values were 0.35 (0.17–0.72), 0.24 (0.07–0.82), 0.50 (0.41–0.63), 0.59 (0.50–0.69), and 0.46 (0.36–0.58), respectively.</p><h3>Interpretation</h3><p>Participants at CKM stages 2–4 demonstrated a graded increase in the risks of all-cause, CVD-specific, and cancer-specific mortality, particularly among younger adults. Having a healthy lifestyle can mitigate these risks, highlighting the importance of lifestyle intervention, especially through non-smoking, adequate physical activity, and healthy sleep duration. As both CKM stage and lifestyle were assessed solely at baseline and the study is observational in nature, the findings may be subject to unmeasured temporal variability and residual confounding, which should be considered when interpretating the results.</p>