The aging population and rising prevalence of unhealthy lifestyles have increased the global burden of cardiovascular disease (CVD). Between 1990 and 2019, the number of CVD cases worldwide increased by almost two-fold, from 271 million to 523 million, while the annual number of deaths increased from 12.1 million to 18.6 million [1]. In China alone, CVD has become the major cause of death, accounting for 40% of deaths in 2010 [2].

Obesity is a major and modifiable CVD risk factor linked by the mechanisms of inflammation, insulin resistance, and abnormal lipid profiles [3]. Obesity has reached epidemic status in China and globally, posing serious public health and economic challenges [4]. The traditional marker for overall obesity is body mass index (BMI) [5, 6], although this does not consider fat distribution and variability in body composition, with the latter being perhaps more important for CVD risk estimation. A study of 36,656 subjects by Lee et al. [7] revealed the visceral-to-subcutaneous fat ratio was more closely associated with all-cause mortality than general obesity, as indicated by BMI. Other studies have also reported that abdominal fat distribution was a stronger predictor for CVD than general obesity [8, 9].

Thomas et al. [10] introduced the body roundness index (BRI) in 2013. This novel anthropometric index combines height and waist circumference measurements to estimate both visceral and overall body fat using mathematical modeling. Various studies have since demonstrated significant correlations between BRI and several metabolic conditions, including diabetes, hypertension, hyperuricemia, and metabolic syndrome, with several of the associations even surpassing those of conventional anthropometric measurements [11, 12, 13, 14]. A recent large-scale, cross-sectional analysis of 17,360 participants from a Chinese population confirmed significant correlations between BRI and a composite of cardiometabolic risk factors [15]. Despite the promising preliminary evidence of BRI as a possible indicator of cardiometabolic health, there is currently only limited longitudinal evidence on its relation with CVD risk in representative national samples. In addition, the dose-response relationship between BRI and CVD risk, as well as its stability in diverse subgroups, has yet to be explored. Using data from the China Health and Retirement Longitudinal Study (CHARLS), the aim of this analysis was to explore the relationship between BRI and CVD risk, examine how this relationship varies across demographic and clinical subgroups, and investigate potential non-linear dose-response patterns between BRI and CVD risk.

This analysis examined the association between BRI and CVD risk. A total of 1121 patients developed CVD over the course of the study. Kaplan-Meier analysis found significant inter-group differences according to BRI quartiles. This result strengthens the potential application of BRI as a simple and accessible tool for clinical CVD risk stratification, allowing clinicians to identify individuals who are at increased risk according to abdominal fat distribution. Adjusted Cox proportional hazards analysis revealed a significant association between BRI and CVD risk when BRI was modeled as a continuous variable (HR = 1.33, 95% CI 1.25–1.40, p 0.001). When BRI was modeled as a categorical variable, the HRs for CVD in the second, third and fourth BRI groups compared to the Q1 reference group were 2.12 (95% CI 1.72–2.75), 2.99 (95% CI 2.33–3.83) and 3.14 (95% CI 2.50–3.96) respectively. These findings demonstrate the utility of BRI as an efficient tool for the discrimination of individuals with different CVD risks, further emphasizing its utility in clinical applications for risk prediction. RCS models also demonstrated a non-linear association between elevated BRI and increased CVD risk. Moreover, in certain subgroups, elevated BRI was associated with a significantly higher incidence of CVD. Significant interactions were observed between BRI and gender, age, hypertension, diabetes, alcohol consumption, smoking status, and BMI. The ability of BRI to predict CVD risk in different subgroups further supports its utility in clinical practice. Additionally, its reliability and non-invasive nature provides a useful tool for identifying high-risk individuals in resource-limited environments where costly imaging is not feasible. However, BRI may be an indicator of underlying common risk factors, such as chronic inflammation, insulin resistance, and metabolic dysfunction, rather than being a direct causative agent of CVD. This is an important consideration when interpreting the predictive function of BRI, and emphasizes the need for further research to elucidate the mechanistic processes involved.

Obesity is a chronic metabolic disorder characterized by abnormal fat deposition and metabolic derangement. It is closely associated with several diseases, including CVD, diabetes, cancer, and metabolic syndrome [16, 17]. Although the pathophysiological basis for the link between visceral fat deposition and CVD risk is well-established, the current research highlights the utility of BRI as a simple, accessible, and non-invasive tool for estimating this risk. Compared to the expensive imaging methods of computed tomography (CT) and MRI, BRI is a cost-effective option for the early determination of high cardiovascular risk in clinical practice. BRI is a new anthropometric index for measuring abdominal fat deposition. It is assessed by measuring waist circumference and height, allowing more accurate assessment of visceral fat and body fat distribution. Excess deposition of visceral fat is a well-known and potent risk factor for CVD, with elevated BRI being associated with metabolic disturbances such as insulin resistance, diabetes, hypertension, dyslipidemia, and metabolic syndrome [18, 19, 20, 21, 22, 23]. For example, a prospective cohort study conducted in Southwest China and involving 9280 participants found that for each unit increase in BRI, the risk of hypertension increased by 17% after adjusting for confounding factors [11]. A study by Cai et al. [24] on patients with hypertension and obstructive sleep apnea also validated the predictive ability of BRI for CVD risk. A large cross-sectional study conducted by Li et al. [25] showed that BRI outperforms other indices in predicting CVD risk in the Chinese population.

In the present analysis, a positive correlation was found between increased BRI and CVD risk after controlling for appropriate covariates. The main strength of our study was its detailed exploration of how different covariates affect the association between BRI and CVD risk, further supporting the utility of BRI as an assessment tool for cardiovascular risk. Our results add to the clinical evidence of an association between BRI and CVD risk, emphasizing the potential utility of BRI as a simple yet accessible marker for the screening of high-risk groups. Several studies have demonstrated that elevated BRI is significantly associated with CVD, cardiovascular-related deaths, and overall mortality [26, 27, 28, 29]. Although the current study did not directly investigate the specific biological mechanisms involved, existing research suggests that excessive accumulation of abdominal fat may increase the CVD risk through inflammatory responses and metabolic disturbances. Previous studies have shown that abdominal fat accumulation may lead to chronic low-grade inflammation, thereby affecting insulin sensitivity, blood pressure regulation, and lipid metabolism, all of which increase the risk of CVD [30, 31, 32, 33, 34, 35, 36]. Furthermore, the accumulation of fat tissue can activate neurohormonal systems and promote oxidative stress responses, further exacerbating the cardiovascular burden [37, 38]. The focus of our study was to determine the association between BRI and CVD risk, rather than conducting a detailed investigation of the specific processes involved. Future research should aim to investigate these mechanisms in greater detail, thereby allowing the mechanism by which BRI affects cardiovascular health to be clarified.

This study has several strengths. First, it was based on a large-scale, nationally representative sample from the CHARLS, thereby reducing the population heterogeneity and enhancing the generalizability of the results. Second, a longitudinal study design was employed, allowing a more reliable inference of the relationship between BRI and CVD through time-to-event analysis. This distinguishes our study from most previous cross-sectional studies. Furthermore, RCS models revealed a non-linear dose-response relationship between BRI and CVD risk, adding to the novelty of our findings. We also performed subgroup analyses to evaluate the impact of factors such as gender, age, BMI, and comorbidities on the BRI-CVD relationship, allowing us to identify differential effects in various population subgroups and filling gaps in the current literature.

This study also has several limitations. Although we employed a longitudinal design, causality cannot be definitively established because the study was based on observational data, and hence residual confounding factors may still be present. For instance, unmeasured variables such as dietary habits, physical activity, lifestyle factors, and environmental influences could still potentially affect the risk of CVD. Additionally, while the follow-up period was adequate for preliminary risk assessment, it may be insufficient to fully capture the long-term cardiovascular outcomes. Lastly, BRI was measured only at baseline. Dynamic changes in fat levels over time were not considered and may impact the comprehensive assessment of long-term health effects. Another major limitation is that CVD incidence was determined by physician-diagnosed self-reports. This methodology is susceptible to recall bias and misclassification bias, since certain participants might over-report or under-report their health status due to memory lapses or subjective interpretation. Although the CHARLS dataset is generally reliable, the accuracy of CVD diagnosis is limited by the lack of verification through medical recordings or clinical visits. This potential source of bias should be taken into consideration when interpreting the results of our analysis. Furthermore, the confounding variables included in Model 3 of this study are closely related to the metabolic processes associated with BRI and may act as mediators influencing CVD (Fig. 3). Adjusting for these variables could lead to over-adjustment of the true relationship between BRI and CVD, thereby introducing bias and reducing the interpretability of the model. Future studies should carefully select adjustment variables to avoid over-adjustment and the confounding of mediating effects, ensuring the accuracy of causal inferences.

Fig. 3.

Directed acyclic graph.

These findings suggest that future research should have a longer-term, prospective study design. Moreover, it should incorporate repeated BRI measurements and adjustment for other behavioral and environmental variables in order to better elucidate the causality of BRI for CVD.

In summary, elevated BRI is associated with increased CVD risk. Measurement of BRI may be useful for identifying high-risk individuals and for informing CVD prevention and intervention strategies. Additional research is required to elucidate the specific mechanisms underlying the association between BRI and CVD, therefore allowing better prevention.

The raw data supporting the conclusions of this article are available from the corresponding author upon reasonable request.

YZS and JFL designed the research study. YZS, PS, SWC, YLL, and YYZ made major contributions to the data collection, analysis, and interpretation. All authors contributed to the editorial revisions of the manuscript, read and approved the final version. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

The study was carried out in accordance with the guidelines of the Declaration of Helsinki. This research made use of the dataset of the CHARLS, which was approved by the Institutional Review Board at Peking University (IRB00001052-11015). Written informed consent was obtained from all participants prior to their recruitment into the study. The complete CHARLS dataset is publicly accessible for academic research on the official CHARLS web page (http://charls.pku.edu.cn/en).

Medical Science Research Project of Hebei (20261340).

The authors declare no conflict of interest.