Association of dietary index of gut microbiota with cardiovascular disease risk: new evidence from NHANES 2007–2018

Abstract

Background

The dietary index of gut microbiota (DI-GM) is a newly proposed index for assessing dietary quality, and studies on its association with cardiovascular disease (CVD) are limited. This study aimed to investigate the association between DI-GM and the prevalence of CVD.

Methods

We utilized data from the National Health and Nutrition Examination Survey (NHANES). Logistic regression analyses were performed to examine the association between DI-GM and CVD. Smoothed curve fitting was employed to explore potential nonlinear relationships. Additionally, subgroup analyses were conducted to assess the stability of the results.

Results

The study included 22,590 participants, of whom 20,216 had no CVD and 2,374 had CVD. After adjusting for all covariates, the DI-GM score was significantly negatively associated with CVD risk, with a 4% reduction in CVD risk for each unit increase in DI-GM score (OR = 0.96, 95% CI: 0.94–0.99, P = 0.015). Notably, the highest DI-GM score group (6–12) had a 13% lower risk of CVD compared to the lowest DI-GM score group (0–3) (OR = 0.87, 95% CI: 0.76-1.00, P = 0.048).

Conclusion

The research results indicate that a higher DI-GM score protects against CVD, providing crucial empirical support for dietary intervention strategies based on gut microbiota modulation.

Clinical trial number

Not applicable.

Introduction

Cardiovascular disease (CVD) represents the leading cause of morbidity and mortality globally [1, 2], which significantly impacts public health and healthcare systems. The aging population is experiencing a dramatic increase in CVD prevalence and incidence [3], placing further strain on resources and highlighting the urgent need for preventative strategies and novel therapeutic targets. While traditional risk factors such as hypertension, dyslipidemia, and diabetes are well-established contributors to CVD [4, 5], unhealthy dietary patterns also play a significant role. Specifically, diets high in processed foods and animal fats, combined with low intake of plant-based foods, can disrupt the balance of the gut microbiota. This dysbiosis compromises intestinal barrier integrity, leading to increased intestinal inflammation and translocation of bacterial endotoxins, such as lipopolysaccharide (LPS), into the bloodstream. Subsequently, systemic low-grade inflammation ensues. The resultant inflammatory milieu activates immune cells, promoting atherogenesis. Concurrently, altered gut microbial metabolism produces detrimental metabolites, such as trimethylamine N-oxide (TMAO), which directly impair vascular endothelial function. Furthermore, neurohumoral pathways modulated by gut microbiota dysbiosis influence cardiovascular activity. These interconnected processes culminate in an increased risk of cardiovascular disease, manifested by vascular dysfunction, hypertension, and thrombosis [6, 7].

The intricate relationship between dietary patterns and gut microbiota is increasingly recognized as a crucial determinant of overall health, including cardiovascular well-being [8, 9]. While specific dietary components, such as fiber [10], fermented foods [11], and diverse plant-based food [12], have demonstrated the capacity to positively modulate gut microbial composition and function, traditional dietary indices, including the Healthy Eating Index (HEI) and the Mediterranean Diet Score (MDS), despite their established value in assessing diet quality, have shown inconsistent associations with specific markers of gut microbial diversity and richness [13, 14]. For example, Bowyer et al. [15], utilising data from the TwinsUK cohort, compared the ability of HEI, MDS, and other indices to explain inter-individual variations in gut microbiota. They observed that while the HEI exhibited superior performance in capturing overall microbial community variance, the associations of these indices with specific gut bacterial taxa were not consistently aligned with expectations. Furthermore, Del Chierico et al. [16] postulated that MDS should correlate well with gut microbial composition, given its established links with health. However, the observed associations of MDS with health parameters and the gut microbiota were surprisingly weak. In contrast, the HEI showed a stronger association with the gut microbiota. This suggests that traditional dietary indices may not fully capture the complex interplay between diet and gut microbial ecosystems. Moreover, the generalizability of these indices across diverse populations is limited. For instance, the HEI has been shown to perform less effectively than the MDS in older populations [17]. These findings underscore that existing dietary indices may not adequately capture the relevant dietary factors associated with alterations in gut microbial composition, thus highlighting the need for more targeted dietary assessment tools.

Novel dietary indices are emerging to address the limitations of current approaches and provide a more nuanced understanding of the complex interactions between diet and gut microbiota. The dietary index of the gut microbiome (DI-GM) was designed to quantify dietary intake patterns related to the composition and function of the gut microbiota. Such a tool could be handy for unraveling the intricate relationship between diet and shedding light on the specific functions of gut microbiota. Although associations between the DI-GM index and various health outcomes have been explored, investigations into its relationship with CVD risk remain relatively scarce. Moreover, given that dietary interventions are more straightforward to implement, more cost-effective, and generally better tolerated by individuals, we therefore aim to investigate the association between DI-GM scores and CVD risk. This investigation seeks to provide further insights into the role of diet and gut microbiome modulation in influencing CVD. Ultimately, it may help identify potential dietary interventions for preventing and managing CVD.

Study population

The data for this study were sourced from the National Health and Nutrition Examination Survey (NHANES), a nationally representative study that evaluated the nutritional status and general health of adults and children in the US. The NCHS Institutional Review Board authorised NHANES, all procedures complied with applicable regulations, and each participant gave written informed permission [18]. We performed a preliminary analysis of the data of 59,864 participants from 2007 to 2018. After meticulously excluding 3,963participants due to missing DI-GM data, 25,416 participants missing CVD data, 286 participants missing body mass index (BMI) data, 3160 participants with missing alcohol use data, and 4449 participants with missing poverty income ratio data, we ultimately included 22,590 participants in this study. The study procedure is illustrated in Fig. 1.

Fig. 1

Flowchart showing the selection of the studied population

Definition of CVD

A diagnosis of CVD was confirmed by self-report collected during a structured interview. That is, whether a healthcare professional had diagnosed them with congestive heart failure, myocardial infarction, angina, or coronary artery disease. An affirmative response to any of these inquiries identified the participant as having CVD.

Definition of DI-GM

DI-GM is a literature-based dietary assessment tool designed to quantify the impact of specific foods or nutrients on the gut microbiota. This index incorporates 14 components, categorised as either beneficial (e.g., fermented dairy products, chickpeas, whole grains) or detrimental (e.g., red meat, processed meats, refined grains) based on their potential effects on gut microbial diversity, short-chain fatty acid (SCFA) production, and the Firmicutes-to-Bacteroidetes ratio. The DI-GM scoring method utilises dietary data from the NHANES survey. It employs a weighted scoring system based on whether an individual’s intake meets or exceeds sex-specific medians, resulting in a cumulative score ranging from 0 to 13 [19, 20].

Covariates

To control for potential confounding, we adjusted for gender, age, race, educational level, marital status, physical exercise, medium movement, smoking and alcohol use, poverty income ratio, BMI, total cholesterol, high-density lipoprotein, chronic kidney disease, hypertension, and alcohol use. Diabetes status was self-reported and professionally verified. Further details on covariate definitions are available in the NHANES documentation.

Statistical analysis

Data were analysed using Empower Stats v2.0 and R v3.4.3. Participant differences were assessed using descriptive statistics; continuous variables (mean ± standard deviation [SD] or standard error [SE]) were compared using t-tests, whereas categorical variables (classified as proportions) were analysed using chi-square tests. Multivariate logistic regression models examining the independent association of DI-GM with CVD risk reported odds ratios (ORs) with 95% confidence intervals (CIs) to quantify the strength of the effect. To assess the robustness of the observed associations and potential sex, age, BMI, hypertension, and diabetes-specific impact, we performed stratified analyses and interaction analyses to explore the association of DI-GM with CVD in populations with different characteristics. In addition, we investigated the nonlinear association of DI-GM with CVD using smoothed curve fitting. Statistical significance was defined as P < 0.05.

Results

Description of participants

Table 1 presents the baseline characteristics of the 22,590 NHANES participants, categorised by the presence or absence of cardiovascular disease. The demographic composition of the participants was as follows: 15.19% Mexican American, 10.06% other Hispanic, 44.26% non-Hispanic White, 20.00% non-Hispanic Black, and 10.49% from other racial groups. Among a range of variables, including gender, age, race, educational level, marital status, diabetes, physical exercise, medium movement, smoking, and alcohol use, hypertension, chronic kidney disease, PIR (poverty income ratio), High-density lipoprotein (HDL), TC (total cholesterol), DI-GM, poverty income ratio, BMI, total cholesterol, high-density lipoprotein, chronic kidney disease, hypertension, there were significant differences between the two groups (P < 0.001).

Table 1 Characteristics of NHANES participants, 2007–2018

Correlation between DI-GM and CVD

Logistic regression analysis (Table 2) showed a significant positive correlation between DI-GM and CVD with or without adjustment for covariates.

Table 2 Association between DI-GM and CVD, NHANES 2007–2018

Association of DI-GM with CVD

Multivariate analysis (Model III) revealed that the highest tertile of DI-GM was associated with a 4% lower CVD risk compared to the lowest tertile (OR = 0.96, 95% CI: 0.94–0.99, P = 0.015).Notably, Trend analysis showed a 3% reduction in CVD risk for each additional tertile group of DI-GM scores (OR = 0.97, 95% CI:0.93-1.00; P for trend = 0.045).

Smoothed curve fitting

We also visualised and analysed the results by smoothing curve fitting. As shown in Fig. 2, after adjusting for all covariates, there was a trend toward decreasing CVD with increasing DI-GM score (overall P < 0.001).

Fig. 2

Correlation of DI-GM with Cardiovascular Disease

Subgroup analysis

Subgroup analyses were performed to assess the stability of the relationship between DI-GM and CVD. As shown in Fig. 3, no significant interactions were found between gender, age, BMI, diabetes mellitus, and hypertension in the subgroup analyses (P > 0.05 for interaction).

Fig. 3

Correlation of DI-GM with Cardiovascular Disease

Figure 3. Subgroup analysis between DI-GM and CVD

Discussion

The present study observed a negative association between DI-GM scores and CVD risk, suggesting that a dietary pattern that promotes healthy gut microbiota, such as plant foods, whole grains, and fermented foods, may have protective effects on cardiovascular health. This finding is consistent with the association of gut flora dysbiosis with increased CVD risk in previous studies [21, 22]. In addition, our findings further support the notion that modulating gut flora through dietary interventions may improve cardiovascular health [23, 24]. Specifically, nutritional components covered by the DI-GM Index, such as fibre, polyphenols, and dietary precursors of SCFA, have been shown to have the ability to modulate the structure and function of the gut flora, which in turn influences host metabolic and immune responses [25, 26]. These alterations may reduce the risk of CVD through various pathways, including lowering systemic inflammation, improving lipid metabolism, and regulating blood pressure [27]. While our findings suggest an association between DI-GM and CVD risk, the underlying mechanism requires further investigation.

CVD is characterised by complex pathophysiological mechanisms, and intestinal dysfunction and its association with systemic inflammation are becoming increasingly prominent. Deng et al. demonstrated that patients with congestive heart failure (CHF) suffer from significant abnormalities of intestinal structure and function, which contribute to the progression of CVD through multiple mechanisms [28]. Specifically, impaired intestinal barrier functionthe, commonly referred to as “leaky gut,” leads to the translocation of LPS into the bloodstream, triggering systemic inflammation and exacerbating vascular injury. Meanwhile, impaired intestinal microcirculation impairs intestinal barrier function and affects nutrient absorption. In addition, Wang et al. [29]. Showed that patients with CHF often suffer from intestinal dysbiosis, which is characterised by changes in the abundance of specific flora, such as an increase in Gram-negative bacteria and a decrease in SCFA-producing bacteria [30]. This dysbiosis affects CVD in several ways. Firstly, it exacerbates a “leaky gut,” activates immune cells, and promotes vascular inflammation and atherosclerosis. Secondly, the gut flora is involved in the metabolism of dietary components (e.g., choline and carnitine), producing trimethylamine (TMA), which is oxidised in the liver to trimethylamine oxide (TMAO), which promotes thrombosis and increases the risk of CVD [31]. Furthermore, gut flora may indirectly affecting influence host metabolism by affecting bile acid metabolism and SCFA production, thereby impacting cardiovascular health [30]. It is worth noting that the association between gut microbiota and CVD is not a simple linear causality but relatively complexly regulated by various factors such as host genetics, immune status, and lifestyle [32, 33]. Although studies targeting specific flora and metabolites have provided important clues, the particular mechanisms of gut flora in the development of CVD still need to be explored in depth to develop more effective CVD prevention and treatment strategies based on intestinal microecological regulation.

Hou et al. demonstrating that the dynamic balance of intestinal flora is intricately regulated by both environmental and host factors [34]. Environmental factors such as dietary patterns, geographic migration, and antibiotic use significantly affect the composition of the flora. In contrast, host phenotypes such as BMI, metabolic indicators, and disease states are also associated with dysbiosis. Genetic factors and family cohabitation also have an impact. Among these factors, diet is a key modifiable factor [35]. To quantify dietary patterns and their health effects, researchers have but developed a variety of dietary assessment tools, including the HEI and the MDS, which are widely used. However, these traditional dietary indices do not specifically assess the effects of diet on the gut microbiota. Their correlations with indicators of the diversity and abundance of the intestinal flora have been inconsistent [15, 36]. DI-GM, as a novel dietary index, offers several key advantages. Firstly, it is constructed based on evidence from the literature and focuses on a wide range of indicators, including gut microbiota diversity, SCFA production, and specific bacterial changes. Secondly, it is more targeted by including particular foods rather than food groups. Additionally, it incorporates beneficial components, such as fermented dairy and chickpeas, and distinguishes between unfavourable components. Furthermore, studies have confirmed that DI-GM correlates with biomarkers of gut microbiota diversity and is comparable, providing a reasonable basis for dietary assessment [36].

This study observed a negative correlation between DI-GM scores and CVD risk, a finding highlighting the role of diet as a potentially modifiable factor in regulating gut health and cardiovascular health. Several beneficial elements in the DI-GM, such as green tea and coffee, have been investigated and shown to have cardiovascular protective effects. Specifically, moderate coffee and green tea intake reduces the risk of CVD events such as coronary heart disease, heart failure, and stroke and positively affects metabolic syndrome [37, 38]. In addition, fermented foods are an essential component of DI-GM, and several studies have shown that fermented foods can potentially benefit cardiometabolic health by modulating gut flora, reducing inflammation and oxidative stress, and enhancing gut barrier function [39]. The mechanisms underlying these benefits may involve bioactive compounds produced during fermentation. These compounds can activate the Nrf2 pathway, which exerts cytoprotective effects and attenuates chronic inflammatory responses associated with obesity, atherosclerosis, and others [40]. Therefore, in-depth studies incorporating gut microbiome data are essential to comprehensively assess the practical value of DI-GM in the prevention and management of cardiovascular diseases. Future studies should focus on exploring the interactions between different food components in DI-GM and specific gut microbiota, as well as the effects of the resulting metabolites on the cardiovascular system, to provide the scientific basis for developing more precise and individualised dietary interventions.

Strengths and limitations

Strengths of this study include the large and representative sample size, which allowed subgroup analyses by sex, age, body mass index, blood pressure, and blood glucose to assess the robustness of DI-GM to CVD. However, the cross-sectional study design limited the ability to infer a causal relationship between DI-GM and CVD. In addition, residual confounders may remain despite full covariate adjustment.

Conclusion

Our findings suggest that DI-GM is significantly associated with increased CVD. These results provide a scientific basis for nutritional intervention strategies targeting CVD.