Muesli Intake May Protect Against Coronary Artery Disease

Background Diet is a key modifiable risk factor of coronary artery disease (CAD). However, the causal effects of specific dietary traits on CAD risk remain unclear. With the expansion of dietary data in population biobanks, Mendelian randomization (MR) could help enable the efficient estimation of causality in diet-disease associations. Objectives The primary goal was to test causality for 13 common dietary traits on CAD risk using a systematic 2-sample MR framework. A secondary goal was to identify plasma metabolites mediating diet-CAD associations suspected to be causal. Methods Cross-sectional genetic and dietary data on up to 420,531 UK Biobank and 184,305 CARDIoGRAMplusC4D individuals of European ancestry were used in 2-sample MR. The primary analysis used fixed effect inverse-variance weighted regression, while sensitivity analyses used weighted median estimation, MR-Egger regression, and MR-Pleiotropy Residual Sum and Outlier. Results Genetic variants serving as proxies for muesli intake were negatively associated with CAD risk (OR: 0.74; 95% CI: 0.65-0.84; P = 5.385 × 10−4). Sensitivity analyses using weighted median estimation supported this with a significant association in the same direction. Additionally, we identified higher plasma acetate levels as a potential mediator (OR: 0.03; 95% CI: 0.01-0.12; P = 1.15 × 10−4). Conclusions Muesli, a mixture of oats, seeds, nuts, dried fruit, and milk, may causally reduce CAD risk. Circulating levels of acetate, a gut microbiota-derived short-chain fatty acid, could be mediating its cardioprotective effects. These findings highlight the role of gut flora in cardiovascular health and help prioritize randomized trials on dietary interventions for CAD.

C oronary artery disease (CAD) is a leading cause of global death, accounting for 9.4 million deaths each year. 1 Diet is a modifiable lifestyle factor amenable to safe, early, and costeffective intervention. 2Previously, the INTERHEART study, a large, international, case-control study assessing the importance of risk factors for coronary artery disease worldwide, attributed 30% of CAD risk to poor dietary habits. 3Meanwhile, the COR-

DIOPREV study (Coronary Diet Intervention
With Olive Oil and Cardiovascular Prevention) 4 and the Lyon Diet Heart Study 5 highlighted the protective effects of a Mediterranean diet in secondary cardiovascular disease prevention.Dietary interventions have also demonstrated a 45% reduction in CAD-related mortality 6 and signs of reversing atherosclerosis. 7While past studies have examined certain foods, such as red meat, 8 fish, 9 and plant oils, 10 the full diversity of foods and beverages in the human diet remains to be tested for causal effects on CAD risk.
Causal relationships between diet and CAD remain unclear because observational cohort studies carry limitations preventing causal interpretations.
Depending on their design, some observational studies may be biased by residual confounders, which can encompass a range of lifestyle factors, such as smoking and physical activity, as well as environmental determinants like socioeconomic status, toxin exposure, and health care access.Reverse causation is another potential bias, where individuals alter their diet in response to their condition rather than the diet causing the condition.These biases may explain why several prominent observational findings were later contradicted by evidence from randomized controlled trials (RCTs). 11,12RCTs provide a critical approach to addressing some of these challenges, but they too are susceptible to error. 13Additionally, RCTs face ethical, financial, and methodological constraints, including stringent participant selection criteria.To ascertain causality in diet-CAD associations, an integrated approach combining multiple study designs is needed to "triangulate" 14 and consolidate the evidence from multiple approaches.
Mendelian randomization (MR) is a method employed in genetic epidemiology that leverages genetic variants as instrumental variables (IVs) to infer, under critical assumptions, causal links in observational data between modifiable exposures (such as risk factors) and outcomes of interest (disease). 15By utilizing germline variants, in which alleles are randomly allocated at conception prior to the onset of diseases like CAD, MR emulates the random assignment characteristic of RCTs.Historically, MR faced challenges with testing dietary traits due to limited genetic data on diet, resulting in a shortage of genetic instruments representing these traits.However, with the expansion of large-scale population biobanks like the UK Biobank (UKB), there is a proliferation of dietary data linked to both genetics and health records.
The unprecedented size and granularity of these data sets is enabling genome-wide association studies (GWAS) to identify potential genetic instruments for detailed dietary traits, despite the typically low genetic heritability of such variables. 16re, we perform 2-sample MR to estimate the causal effects of 13 dietary traits on CAD risk.Our study also included summary statistics from the CARDioGRAMplusC4D 1000 genomes-based GWAS, which included 60,801 CAD cases and 123,504 controls. 19We restricted our analyses to participants of European genetic ancestry to reduce heterogeneity.

METHODS
DIETARY TRAIT ASSESSMENT.UKB collected dietary intake information via 2 well-validated methods: 1) a touchscreen questionnaire at recruitment (2006-), 20 which asked 29 questions related to the average frequency of consumption of common foods and food groups over the past year; and 2) a more detailed, web-based dietary recall questionnaire (2011-), which documented self-reported intake of over 200 commonly consumed foods and beverages during the preceding 24 hours. 21Thus, self-reported muesli intake was documented by 2 traits: one trait (datafield 1468) was of nominal categorical data collected in the touchscreen questionnaire by a multiple-choice question: "What type of cereal do you mainly eat?" The answer choices were "Bran cereal (eg, All Bran, Branflakes)"; "Biscuit cereal (eg, Weetabix)"; "Oat cereal (eg, Ready Brek, porridge)"; "Muesli"; "Other (eg, Cornflakes, Frosties)"; "Do not know"; and "Prefer not to answer".A second trait (data-field 100800) was of ordinal categorical data collected in the 24-hour recall questionnaire by a multiple-choice question: "How many bowls of muesli?"This question was only asked to participants who reported consuming breakfast cereals the previous day.
Answer choices were "none," "1/2," "1," and "2þ."  22 The exact sample size of each dietary trait GWAS in Pan-UKB is included in Supplemental Table 1.Using the Genome-wide Complex Trait Analysis (GCTA) software package, 23 we performed a conditional and joint association analysis (COJO) 24 to select independently associated SNPs for each of the 13 traits of interest in the European ancestry subset of Pan-UKB, based on the following options: -cojo-slct-cojo-p 5 Â 10-8 -cojo-collinear 0.9 -cojo-wind 10,000 -diff-freq 0.2.A subset of Pan-UKB was used as the reference sample (sample size ¼ 10,000 individuals; 17,091,093 SNPs).The 13 SNPs instrumenting muesli intake (data-field 1468) are detailed in Supplemental Table 2.The number of SNPs instrumenting each trait is also included in Supplemental Table 1.
GENETIC INSTRUMENTS FOR CAD RISK.We used GWAS summary statistics from the Coronary Artery Disease Genome-wide Replication and Meta-analysis plus The Coronary Artery Disease Genetics (CARDIo-GRAMplusC4D) consortium.CARDioGRAMplusC4D (60,801 cases; 123,504 controls) combined data from 48 genetic studies to identify risk loci for CAD and myocardial infarction. 25We performed COJO analysis using GCTA software, using the same parameters and reference sample, as described above.3. Details of the metabolic profiling platform have been described. 26 identify genetic instruments for plasma metabolite levels, we conducted genome-wide association tests on each of 171 metabolic traits, using directly genotyped or imputed genetic data as predictors, while adjusting for age, sex, and principal components 1 to 10.We then used REGENIE, 27 a machine learning method for fitting a whole-genome regression model for quantitative and binary phenotypes that is computationally efficient.Using PLINK, 28 a free, open-source whole-genome association analysis toolset, we then extracted summary statistics for the 13 SNPS instrumenting muesli intake listed in Supplemental Table 2.
STATISTICAL ANALYSES.Analyses were performed using R 4.1.0. 29The primary analysis used fixed effect, inverse-variance weighted (IVW) regression. 30-Egger regression 31 and weighted median estimation 32 were also used in sensitivity analyses to detect signs of IV assumption violations, namely horizontal pleiotropy.The trait "smoking status: never" was included in this analysis as a control, and a Bonferroni-adjusted threshold for statistical significance was used for 13 dietary traits and 1 control trait: To detect IVs sensitive to horizontal pleiotropy, MR-Pleiotropy Residual Sum and Outlier (PRESSO) outlier tests 33 were conducted on 9 dietary traits that had significant MR-PRESSO global test results (P < 0.05) in the primary analysis.Between 1 and 2 SNPs were identified as outliers and removed for 6 of the 9 traits based on a corrected significance threshold (P < 0.05/n), where n was the number of SNPs instrumenting each trait.IVW tests were then rerun without the outlier IVs for these 6 traits.
No SNPs were removed for the remaining 3 traits that had nonsignificant MR-PRESSO global test results.
In addition to recording muesli intake as a nominal categorical trait (data-field 1468), UKB had documented the amount of muesli consumed as an ordinal categorical trait (data-field 100800).However, this ordinal trait did not have sufficient genetic association data to construct instruments in the primary IVW analysis.We thus used IVs for the nominal trait (datafield 1468) as exposure instruments and extracted association statistics with the ordinal trait (data-field 100800) for MR analysis, as described previously.
Statistical significance was defined conventionally for this follow-up analysis (P < 0.05).
Since diet can influence the plasma metabolome and metabolites can suggest biological mechanisms, we further tested whether the effects of muesli on CAD could be mediated by any of the 171 plasma metabolites measured in UKB.The IVs for muesli intake served as exposure instruments (Supplemental Table 2), and for each IV, association statistics with each metabolite were extracted.Each allele was then harmonized to ensure consistency between the effect and non-effect alleles.Subsequently, we performed MR using IVW, MR-Egger, and weighted median methods with a Bonferroni-corrected threshold for statistical significance (P < 0.05/171 ¼ 2.923 Â 10 À4 ).
DATA AVAILABILITY.Data generated in this study are included in this paper and in the Supplemental Tables.All analyses used publicly available data from UKB.
In sensitivity analyses, we used several MR methods robust to horizontal pleiotropy and/or Park et al
Although MR-PRESSO global tests detected outlier Mendelian Randomization on Muesli and Heart Disease SNPs for 6 traits, adjustment with MR-PRESSO outlier tests did not uncover significant associations.Full results, including the number of genetic instruments for each trait, are provided in Supplemental Table 1.
Additional MR analyses testing for mediation through muesli's ingredients suggested the potential role of dried fruits, nuts, and cholesterol-lowering milk in muesli's cardioprotective effects (Supplemental Table 4).Importantly, these data do not provide evidence for the absence of contributions from muesli's other ingredients.

VALIDATION OF MUESLI CAD RISK ASSOCIATION
USING ORDINAL CATEGORICAL DATA.UKB documented muesli intake using 2 different assessment methods, resulting in 2 different data sets on muesli: nominal categorical data (data-field 1468) and ordinal categorical data (data-field 100800).Our finding of an association between muesli and CAD risk was based on nominal data, but ordinal data can potentially provide a more graded measure of the exposure and capture subtle changes that could be masked when data are categorized without hierarchy.Thus, we also performed MR analyses between ordinal data on muesli intake and CAD risk.Supporting the primary finding, IVW regression estimated a significant negative association between the number of bowls of muesli consumed and CAD risk (OR: 0.29; 95% CI: 0.10-0.83;P ¼ 0.040; Figure 3).Estimates were not statistically significant in sensitivity analyses, but the direction of effect remained consistent (Supplemental Table 5).6).

DISCUSSION
We used 2-sample MR in UKB and CARDIo-GRAMplusC4D to test for independent causal associations between 13 dietary traits and CAD risk (Central Illustration).Our results suggested genetic evidence for a potential causal relationship between muesli intake and reduced CAD risk (Figure 2).This was supported using 2 different data sets on muesli intake obtained either by food-frequency questionnaires or 24-hour recall (Figure 3).Secondary MR analyses on plasma metabolite levels suggested this may be mediated by increased levels of acetate, a gut microbiota-derived short-chain fatty acid (SCFA).
These genetic data are consistent with previous observational findings of a negative association between muesli and heart disease. 34They also Mendelian Randomization on Muesli and Heart Disease substantiate previous studies associating acetate with antihypertensive 35 and immunoregulatory mechanisms 36 (Figure 4).Lastly, they agree with the current guidelines for primary prevention of cardiovascular disease, which emphasizes a diet rich in fruits, vegetables, and whole grains. 37esli is commonly defined as a mixture of oats, seeds, nuts, dried fruits, and milk.Many of these nutrient-dense ingredients have been linked to cardiovascular health, suggesting biological relevance. 38,39Oats are notably high in a soluble fiber called b-glucan, which has been shown to reduce LDL cholesterol levels and improve insulin sensitivity. 40,41Oats also contain antioxidant avenanthramides, which are not present in other cereal grains, and these can support endothelial function and reduce blood pressure. 42 inflammation, and LDL cholesterol. 43,44Dried fruits offer essential vitamins, minerals, fibers, and antioxidant flavonoids that can further promote cardiovascular health. 45,46Muesli's cardioprotective effects likely arise from a combination of mechanisms, but targeted analyses isolating key ingredients may be a useful direction for clinical translation.
Acetate is a SCFA primarily generated by colonic bacteria from dietary fiber fermentation.Increasingly, data suggest that gut microbiota-derived metabolites can modify cardiovascular health by mediating diet-microbiome-disease pathways. 47,48r example, dietary red meat increases systemic levels of trimethylamine-N-oxide (TMAO), a gut microbiota-generated metabolite with proatherogenic mechanisms, 49 by providing dietary precursors, increasing microbial TMAO production, and decreasing renal TMAO excretion. 50In this study, we found that higher plasma acetate levels, another gutderived metabolite, could link muesli consumption to reduced CAD risk (Figure 4).Previous studies have shown acetate can independently prevent the development of hypertension in mice and promote the colonic accumulation of anti-inflammatory regulatory T cells. 35,36One possibility is a potential protective muesli-acetate-CAD pathway, analogous to an atherogenic red meat-TMAO-CAD pathway.
The study had limitations.First, the validity of MR hinges on several core assumptions: namely, that genetic instruments (SNPs) are: 1) robustly associated with the exposure (muesli); 2) not confounded by common causes with the outcome (CAD); and 3) have effects on the outcome (CAD) solely through the exposure (muesli). 51However, the interpretation of genetic instruments for predominantly environmental traits, such as dietary intake, is complex.As ETHICAL APPROVAL.UKB received approval from the North West Multi Centre Research Ethics Committee as a Research Tissue Bank (11/NW/0382).This study was approved by UKB application number 16218 and adheres to the principles of the Declaration of Helsinki.STUDY DESIGN AND PARTICIPANTS.The study design is summarized in Figure 1.Following the Strengthening the Reporting of Observational Studies in Epidemiology using MR (STROBE-MR) guidelines, 17 we note that MR relies on 3 assumptions regarding IVs: 1) the genetic variant (eg, a single-nucleotide polymorphism [SNP]) is directly associated with the exposure (muesli intake); 2) the genetic variant is unrelated to confounders between the exposure and the outcome (CAD risk); and 3) the genetic variant has no effect on the outcome (CAD risk) other than through the exposure (muesli intake). 15A B B R E V I A T I O N S A N D A C R O N Y M S CAD = coronary artery disease GWAS = genome-wide association study/studies IV = instrumental variable IVW = inverse-variance weighted LDL = low-density lipoprotein MR = Mendelian randomization OR = odds ratio RCT = randomized controlled trial SCFA = short-chain fatty acid SNP = single-nucleotide polymorphism TMAO = trimethylamine-Noxide UKB = United Kingdom Biobank The authors attest they are in compliance with human studies committees and animal welfare regulations of the authors' institutions and Food and Drug Administration guidelines, including patient consent where appropriate.For more information, visit the Author Center.Manuscript received November 15, 2023; accepted December 12, 2023.Park et al J A C C : A D V A N C E S , V O L . 3 , N O . 4 Summary statistics were obtained on up to 429,531 unrelated, male and female individuals of European genetic ancestry in UKB, a longitudinal, populationbased cohort study with genetic and phenotypic data on over 500,000 participants aged 40 to 69 years at recruitment from across the United Kingdom during 2006 to 2010.
GENETIC INSTRUMENTS FOR DIETARY TRAITS.We identified 13 dietary traits (11 nominal categorical and 2 ordinal categorical traits) with sufficient numbers of IVs (ie, >3 associated variants) in Pan-UKB for the construction of robust genetic instruments.Pan-UKB had performed 16,131 GWAS analyses on 7,221 phenotypes in w420,531 UKB participants of European genetic ancestry.

FIGURE 1 Flowchart
FIGURE 1 Flowchart Summarizing the Study Design GENETIC INSTRUMENTS FOR PLASMA METABOLITE LEVELS.UKB recently released metabolomic data on 121,731 participants.Nightingale Health Ltd had assayed a panel of 249 metabolic biomarkers (168 in absolute values and 81 ratio measures) from plasma obtained at recruitment using nuclear magnetic resonance spectroscopy between June 2019 and April 2020.Most biomarkers were measured in concentration units (mmol/L), but exceptions are indicated in Supplemental Table

FIGURE 2
FIGURE 2 Causal Estimates of 13 Genetically Proxied Dietary Traits on CAD Risk

FIGURE 3 4 Mendelian
FIGURE 3 Causal Estimates of Muesli Intake Documented by 24-Hour Recall on CAD Risk

FIGURE 4
FIGURE 4 Causal Estimates of Muesli on CAD Risk Through Plasma Metabolites

4 Mendelian
Meanwhile, nuts and seeds are rich sources of omega (u)-3 fatty acids, dietary fiber, and antioxidants, such as lignans.They are also associated with reductions in blood pressure, CENTRAL ILLUSTRATION Mendelian Randomization on Muesli and Coronary Artery Disease Park JK, et al.JACC Adv.2024;3(4):100888.Two-sample Mendelian randomization was performed to estimate independent causal associations between 13 dietary traits and coronary artery disease (CAD) risk.Genetic variants serving as proxies for muesli intake were significantly associated with lower CAD risk.Shown is the reduction percentage in the odds of CAD among individuals who documented consuming muesli as a nominal categorical trait compared to those who did not.95% CI: 0.65-0.84and P ¼ 5.39 Â 10 À4 .Park et al J A C C : A D V A N C E S , V O L . 3 , N O . 4 , 2 0 2 Randomization on Muesli and Heart Disease A P R I L 2 0 2 4 : 1 0 0 8 8 8

MendelianTRANSLATIONAL OUTLOOK 2 : 4 Mendelian
Randomization on Muesli and Heart Disease PERSPECTIVES COMPETENCY IN MEDICAL KNOWLEDGE: Genetic data suggest that consuming muesli, a mixture of oats, nuts, seeds, and dried fruits, may causally protect against CAD risk, emphasizing the importance of a healthy, plantbased diet in cardiovascular health.COMPETENCY IN PATIENT CARE: Dietary behaviors are a lifelong determinant of CAD risk.Additional studies are needed, but dietary muesli could potentially be a safe and cost-effective component in diet-based interventions for CAD.TRANSLATIONAL OUTLOOK 1: Although MR is a powerful and useful method, whether muesli intake can truly cause a reduction in CAD risk must be validated by a triangulation of evidence from complementary study designs, such as animal studies and randomized trials.Gut microbiotaderived SCFAs, such as acetate, are increasingly associated with pivotal roles in cardiovascular health.A better understanding of how muesli could be cardioprotective through plasma acetate is needed to potentially harness those mechanisms for therapeutic effect.Park et al J A C C : A D V A N C E S , V O L . 3 , N O . 4 , 2 0 2 Randomization on Muesli and Heart Disease A P R I L 2 0 2 4 : 1 0 0 8 8 8 and egg intake, could not be tested due to the lack of robust genetic instruments for these traits.Dr Do is supported by the National Institute of General Medical Sciences of the National Institutes of Health (NIH) (R35-GM124836) and the National Heart, Lung, and Blood Institute of the NIH (R01-HL139865 and R01-HL155915); is a scientific cofounder, consultant, and equity holder (pending) for Pensieve Health; and is a consultant for Variant Bio, all unrelated to this work.Dr Rosenson has received grants from Amgen, Arrowhead, Lilly, Novartis, and Regeneron; consulting fees from Amgen, Arrowhead, Lilly, Novartis, and Regeneron; honoraria for non-promotional lectures from Amgen, Kowa, and Regeneron, royalties from Wolters Kluwer (UpToDate); and stock holdings in MediMergent, LLC, all unrelated to this work.Dr Verbanck is supported by the French National Research Agency (ANR) (ANR-21-CE45-0023-01), also unrelated to this work.All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
33Nevertheless, causal interpretations warrant caution until the 'triangulation' of supportive evidence from alternative study designs, namely RCTs.A second limitation of the study is that external validity could not be tested due to the absence of genetics-linked data on muesli outside UKB.Moreover, several major dietary traits, such as red meat ADDRESS FOR CORRESPONDENCE: Dr Ron Do, Icahn School of Medicine at Mount Sinai, Annenberg Building, Floor 18, Room 80A, 1468 Madison Avenue, New York, New York 10029, USA.E-mail: ron.do@ mssm.edu.