Advertisement

Intake of legumes and cardiovascular disease: A systematic review and dose–response meta-analysis

  • Vânia Mendes
    Affiliations
    Department of Hygiene, Epidemiology and Medical Statistics, School of Medicine, National and Kapodistrian University of Athens, Athens, Greece
    Search for articles by this author
  • Aikaterini Niforou
    Affiliations
    Department of Hygiene, Epidemiology and Medical Statistics, School of Medicine, National and Kapodistrian University of Athens, Athens, Greece
    Search for articles by this author
  • Maria I. Kasdagli
    Affiliations
    Department of Hygiene, Epidemiology and Medical Statistics, School of Medicine, National and Kapodistrian University of Athens, Athens, Greece
    Search for articles by this author
  • Ermolaos Ververis
    Affiliations
    Department of Hygiene, Epidemiology and Medical Statistics, School of Medicine, National and Kapodistrian University of Athens, Athens, Greece

    European Food Safety Authority, Parma, Italy
    Search for articles by this author
  • Androniki Naska
    Correspondence
    Corresponding author. Dept. of Hygiene, Epidemiology and Medical Statistics, School of Medicine, National and Kapodistrian University of Athens, 75 Mikras Asias Str, Athens 11527, Greece.
    Affiliations
    Department of Hygiene, Epidemiology and Medical Statistics, School of Medicine, National and Kapodistrian University of Athens, Athens, Greece
    Search for articles by this author
Published:October 20, 2022DOI:https://doi.org/10.1016/j.numecd.2022.10.006

      Highlights

      • High legume intake was inversely associated with CVD and CHD, but not with stroke.
      • Regular weekly intakes of legumes are associated with a decrease in CHD risk.
      • The benefit for CHD appears to level off at intakes higher than 400 g/week.
      • Further research is needed to understand the role of legumes in stroke subtypes.

      Abstract

      Aims

      To summarize the evidence on the association between the intake of legumes and the risk of cardiovascular disease (CVD) overall, coronary heart disease (CHD) and stroke, and to identify optimal intake levels for reduced disease risk through a systematic review and dose–response meta-analysis.

      Data synthesis

      We have systematically searched PubMed, Scopus and Web of Science up to March, 2022 for the retrieval of intervention and observational studies (PROSPERO Reg. number: CRD42021247565). Pooled relative risks (RRs) comparing extreme categories of intake were computed using random-effects models. One-stage dose–response meta-analyses were also performed using random-effects models. 22 831 articles were screened resulting in 26 eligible observational studies (21 prospective cohort and 5 case–control studies). When comparing extreme categories of intake, the consumption of legumes was inversely associated with CVD (n = 25: RR = 0.94; 95%CI:0.89,0.99) and CHD (n = 16: RR = 0.90; 95%CI:0.85,0.96), but not with stroke (n = 9: RR = 1.00; 95%CI:0.93,1.08). We further found evidence for an inverse dose–response association with CHD, increasing in magnitude up to an intake of 400 g/week, after which the benefit seems to level-off.

      Conclusions

      The intake of legumes was associated with a reduced risk of CVD and CHD, but not with stroke, among individuals with the highest consumption levels. An intake level of 400 g/week seemed to provide the optimal cardiovascular benefit. Further research is needed to better understand the role of legumes in stroke subtypes.

      Keywords

      Acronyms:

      BMI (Body Mass Index), CI (Confidence interval), CVD (Cardiovascular disease), CHD (Coronary heart disease), FFQ (Food frequency questionnaire), HR (Hazard Ratio), LDL (Low-density lipoprotein), MRI (Magnetic resonance imaging), OR (Odds ratio), PRISMA (Preferred reporting items for systematic reviews and meta-Analyses), PROSPERO (International prospective register of systematic reviews), ROBINS-I (Risk of bias in non-randomised studies of interventions), RCT (Randomized controlled trial), RR (Relative risk), T2D (Type 2 diabetes)

      1. Introduction

      Legumes are defined as the succulent seeds and pods of the botanical family Leguminosae or Fabaceae, including a wide variety of species such as beans, peas, chickpeas, lentils, broad beans, soya beans and lupins [
      • Joint FAO/WHO Food Standards Programme C.A.C.
      Codex classification of foods and animal feeds.
      ], that can be consumed as whole pods, their fresh shelled products or in the form of dried mature seeds (also commonly known as pulses). There has been an increasing recognition on the potential of legumes as environmentally sustainable plant protein sources, with several benefits for human health [
      • Mullins A.P.
      • Arjmandi B.H.
      Health benefits of plant-based nutrition: focus on beans in cardiometabolic diseases.
      ]. In addition to their high protein content, legumes are sources of fibre, essential minerals such as magnesium, potassium, iron and zinc, B vitamins and other bioactive compounds. Moreover, legumes are naturally low in saturated fat and provide low glycaemic-index carbohydrates [
      • Rebello C.J.
      • Greenway F.L.
      • Finley J.W.
      A review of the nutritional value of legumes and their effects on obesity and its related co-morbidities.
      ]. Previous research has consistently linked legumes' consumption with important health benefits, such as reduced incidence of cardiovascular disease (CVD), diabetes, overweight/obesity [
      • Viguiliouk E.
      • Blanco S.
      • Kendall C.
      • Sievenpiper J.
      Can pulses play a role in improving cardiometabolic health? Evidence from systematic reviews and meta-analyses.
      ,
      • Ferreira H.
      • Vasconcelos M.
      • Gil A.M.
      • Pinto E.
      Benefits of pulse consumption on metabolism and health: a systematic review of randomized controlled trials.
      ] and some types of cancer [
      • Nchanji E.B.
      • Ageyo O.C.
      Do common beans (Phaseolus vulgaris L.) promote good health in humans? A systematic review and meta-analysis of clinical and randomized controlled trials.
      ]. Accordingly, there is a growing body of evidence addressing the potential beneficial impact of legumes in cardiovascular health and cardiometabolic factors, highlighting the advantage of integrating legumes in the habitual diet [
      • Viguiliouk E.
      • Blanco S.
      • Kendall C.
      • Sievenpiper J.
      Can pulses play a role in improving cardiometabolic health? Evidence from systematic reviews and meta-analyses.
      ,
      • Ferreira H.
      • Vasconcelos M.
      • Gil A.M.
      • Pinto E.
      Benefits of pulse consumption on metabolism and health: a systematic review of randomized controlled trials.
      ]. For these reasons, legumes feature in several healthy eating indices [
      • Trichopoulou A.
      • Costacou T.
      • Bamia C.
      • Trichopoulos D.
      Adherence to a Mediterranean diet and survival in a Greek population.
      ,
      • Fung T.T.
      • Chiuve S.E.
      • Rexrode K.M.
      • Logroscino G.
      • Hu F.B.
      Adherence to a DASH-style diet and risk of coronary heart disease and stroke in women.
      ,
      • Krebs-Smith S.M.
      • Pannucci R.E.
      • Subar A.F.
      • Kirkpatrick S.I.
      • Lerman J.L.
      • Tooze J.A.
      • et al.
      Update of the healthy eating index: HEI-2015.
      ] and dietary guidelines of Scientific Bodies and Medical Societies [
      • World Cancer Research Fund/American Institute for Cancer Research
      Continuous update Project expert report 2018.
      ,
      • Mach F.
      • Baigent C.
      • Catapano A.L.
      • Koskinas K.C.
      • Casula M.
      • Badimon L.
      • et al.
      2019 ESC/EAS Guidelines for the management of dyslipidaemias: lipid modification to reduce cardiovascular risk: the Task Force for the management of dyslipidaemias of the European Society of Cardiology (ESC) and European Atherosclerosis Society (EAS).
      ,
      • Lichtenstein A.H.
      • Appel L.J.
      • Vadiveloo M.
      • Hu F.B.
      • Kris-Etherton P.M.
      • Rebholz C.M.
      • et al.
      2021 Dietary guidance to improve cardiovascular health: a scientific statement from the American heart association.
      ]. Globally, CVD is the leading cause of mortality and a major contributor to the loss of years of healthy life due to disability [
      • Roth G.A.
      • Mensah G.A.
      • Johnson C.O.
      • Addolorato G.
      • Ammirati E.
      • Baddour L.M.
      • et al.
      Global burden of cardiovascular diseases and risk factors, 1990–2019: update from the GBD 2019 study.
      ]. There is consistent evidence that diet can favourably modify major CVD risk factors, such as dyslipidaemia, diabetes mellitus, hypertension and obesity and further decrease CVD incidence [
      • Roth G.A.
      • Mensah G.A.
      • Johnson C.O.
      • Addolorato G.
      • Ammirati E.
      • Baddour L.M.
      • et al.
      Global burden of cardiovascular diseases and risk factors, 1990–2019: update from the GBD 2019 study.
      ,
      • Collaborators G.B.D.D.
      Health effects of dietary risks in 195 countries, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017.
      ]. Legumes, as part of a balanced healthy diet, have the potential to protect against the development of CVD through several etiological pathways [
      • Viguiliouk E.
      • Blanco S.
      • Kendall C.
      • Sievenpiper J.
      Can pulses play a role in improving cardiometabolic health? Evidence from systematic reviews and meta-analyses.
      ,
      • Ferreira H.
      • Vasconcelos M.
      • Gil A.M.
      • Pinto E.
      Benefits of pulse consumption on metabolism and health: a systematic review of randomized controlled trials.
      ].
      Previous studies have attempted to summarize the evidence addressing the association between the consumption of legumes and cardiovascular outcomes, indicating a potential beneficial pattern for higher intake levels [
      • Bechthold A.
      • Boeing H.
      • Schwedhelm C.
      • Hoffmann G.
      • Knüppel S.
      • Iqbal K.
      • et al.
      Food groups and risk of coronary heart disease, stroke and heart failure: a systematic review and dose-response meta-analysis of prospective studies.
      ,
      • Grosso G.
      • Marventano S.
      • Yang J.
      • Micek A.
      • Pajak A.
      • Scalfi L.
      • et al.
      A comprehensive meta-analysis on evidence of Mediterranean diet and cardiovascular disease: are individual components equal?.
      ,
      • Marventano S.
      • Izquierdo Pulido M.
      • Sánchez-González C.
      • Godos J.
      • Speciani A.
      • Galvano F.
      • et al.
      Legume consumption and CVD risk: a systematic review and meta-analysis.
      ,
      • Becerra-Tomás N.
      • Papandreou C.
      • Salas-Salvadó J.
      Legume consumption and cardiometabolic health.
      ,
      • Viguiliouk E.
      • Glenn A.J.
      • Nishi S.K.
      • Chiavaroli L.
      • Seider M.
      • Khan T.
      • et al.
      Associations between dietary pulses alone or with other legumes and cardiometabolic disease outcomes: an umbrella review and updated systematic review and meta-analysis of prospective cohort studies.
      ]. However, previous meta-analyses have not differentiated between non-soy and soy legumes [
      • Bechthold A.
      • Boeing H.
      • Schwedhelm C.
      • Hoffmann G.
      • Knüppel S.
      • Iqbal K.
      • et al.
      Food groups and risk of coronary heart disease, stroke and heart failure: a systematic review and dose-response meta-analysis of prospective studies.
      ,
      • Grosso G.
      • Marventano S.
      • Yang J.
      • Micek A.
      • Pajak A.
      • Scalfi L.
      • et al.
      A comprehensive meta-analysis on evidence of Mediterranean diet and cardiovascular disease: are individual components equal?.
      ,
      • Marventano S.
      • Izquierdo Pulido M.
      • Sánchez-González C.
      • Godos J.
      • Speciani A.
      • Galvano F.
      • et al.
      Legume consumption and CVD risk: a systematic review and meta-analysis.
      ], which differ in their nutritional composition [
      • Messina M.J.
      Legumes and soybeans: overview of their nutritional profiles and health effects.
      ]; have provided results for combined incidence and mortality data [
      • Grosso G.
      • Marventano S.
      • Yang J.
      • Micek A.
      • Pajak A.
      • Scalfi L.
      • et al.
      A comprehensive meta-analysis on evidence of Mediterranean diet and cardiovascular disease: are individual components equal?.
      ,
      • Marventano S.
      • Izquierdo Pulido M.
      • Sánchez-González C.
      • Godos J.
      • Speciani A.
      • Galvano F.
      • et al.
      Legume consumption and CVD risk: a systematic review and meta-analysis.
      ]; have not assessed dose–response associations or have used methods that cannot be applied in studies with only two categories of exposure [
      • Bechthold A.
      • Boeing H.
      • Schwedhelm C.
      • Hoffmann G.
      • Knüppel S.
      • Iqbal K.
      • et al.
      Food groups and risk of coronary heart disease, stroke and heart failure: a systematic review and dose-response meta-analysis of prospective studies.
      ,
      • Grosso G.
      • Marventano S.
      • Yang J.
      • Micek A.
      • Pajak A.
      • Scalfi L.
      • et al.
      A comprehensive meta-analysis on evidence of Mediterranean diet and cardiovascular disease: are individual components equal?.
      ,
      • Marventano S.
      • Izquierdo Pulido M.
      • Sánchez-González C.
      • Godos J.
      • Speciani A.
      • Galvano F.
      • et al.
      Legume consumption and CVD risk: a systematic review and meta-analysis.
      ,
      • Becerra-Tomás N.
      • Papandreou C.
      • Salas-Salvadó J.
      Legume consumption and cardiometabolic health.
      ,
      • Viguiliouk E.
      • Glenn A.J.
      • Nishi S.K.
      • Chiavaroli L.
      • Seider M.
      • Khan T.
      • et al.
      Associations between dietary pulses alone or with other legumes and cardiometabolic disease outcomes: an umbrella review and updated systematic review and meta-analysis of prospective cohort studies.
      ].
      We therefore performed a systematic review and dose–response meta-analysis to summarize the evidence on the association between the intake of legumes and the risk of CVD overall, coronary heart disease (CHD) and stroke, and to identify optimal intake levels for a reduced disease risk.

      2. Methods

      This systematic review and meta-analysis was prepared in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines [
      • Page M.J.
      • McKenzie J.E.
      • Bossuyt P.M.
      • Boutron I.
      • Hoffmann T.C.
      • Mulrow C.D.
      • et al.
      The PRISMA 2020 statement: an updated guideline for reporting systematic reviews.
      ]. The study protocol was registered in PROSPERO – International prospective register of systematic reviews (Reg. number CRD42021247565) [
      • Mendes V.
      • Niforou A.
      • Ververis E.
      • Naska A.
      Intake of legumes and cardiovascular disease: a systematic review and dose-response meta-analysis.
      ].

      2.1 Search strategy

      We have conducted a systematic literature search in PubMed, Scopus and Web of Science for original studies published up to March 22, 2022 (Supplemental Table 1), that reported the association between the intake of legumes and CVD (comprising any cardiovascular disease), as well as CHD and stroke in particular. The exposure of interest was defined as the intake of beans, peas, lentils or other species belonging to the leguminosae family [
      • Joint FAO/WHO Food Standards Programme C.A.C.
      Codex classification of foods and animal feeds.
      ], in their dried or fresh form, reported individually or in combination. Since the association between the intake of soy and CVD has already been extensively reviewed [
      • Li N.
      • Wu X.
      • Zhuang W.
      • Xia L.
      • Chen Y.
      • Zhao R.
      • et al.
      Soy and isoflavone consumption and multiple health outcomes: umbrella review of systematic reviews and meta-analyses of observational studies and randomized trials in humans.
      ], soy or soy products were not considered under the exposure definition, unless combined with other legume subtypes.

      2.2 Study selection

      A study was considered eligible if 1) it was an intervention or a prospective or retrospective observational study that 2) evaluated the association between the intake of legumes and CVD, CHD or stroke in 3) the general adult population. Additionally, studies were suitable for inclusion if 4) they reported risk estimates as rate ratios (RRs), hazard ratios (HRs), or odds ratios (ORs), with the corresponding 95% confidence intervals (CIs), or data to calculate the respective variance, for all categories of intake. Studies were excluded if 1) not involving human subjects, 2) not presenting original data (e.g., review articles, editorials, reports, comments or guidelines), 3) performed among children (<18 y) and/or pregnant women, 4) having a cross-sectional design, ecological studies, case series, case reports and qualitative studies, 5) addressing exposures other than the intake of legumes as defined above, 6) addressing outcomes other than CVD, CHD or stroke. In addition, studies in which legume intake was expressed as a continuous variable could not be used in the dose–response meta-analysis. Lastly, no restrictions were imposed regarding the study's language, geographical location or publication date.
      The titles/abstracts and full-texts of studies identified in the original search were reviewed independently by two authors (VM and ANi) in accordance with the exclusion criteria. Any disagreements were resolved by consensus or after discussion involving a third author (AN). The literature search was complemented by back and forward citation tracking of eligible papers and other relevant systematic reviews and meta-analyses.

      2.3 Data extraction

      Two authors (VM and ANi) have independently extracted the information using a standardized data collection form. Since no intervention studies were identified at the study selection stage, the following data were extracted: 1) study characteristics (study design, publication year, duration of follow-up when relevant), 2) sample characteristics (country, study cohort, sex, age, sample size), 3) exposure characteristics (definition, data collection methods and categories of intake), 4) outcome characteristics (definition, outcome assessment methods, number of cases) and 5) association measures with CIs for fully-adjusted models and covariates. Estimates were extracted for the whole sample or for different strata, as available. Authors of eligible studies with missing information that was relevant for the calculation of the pooled estimates were contacted and reply was provided by investigators of the Finnish Mobile Clinic Health Examination Survey on legume intake for males and females combined [
      • Mizrahi A.
      • Knekt P.
      • Montonen J.
      • Laasksonen M.A.
      • Heliövaaea M.
      • Järvinen R.
      Plant foods and the risk of cerebrovascular diseases: a potential protection of fruit consumption.
      ].
      For each intake category we extracted the mean or median intake value, depending on the available data. If these were missing, the midpoint of each intake category was computed. When the highest and lowest exposure categories were “open”, a value that was 20% higher or lower than the closest cut-off point was calculated. For studies where the legume intake was only reported as number of servings/portions, intake was converted into grams using a predefined weight of 100 g per portion, in order to be consistent with the serving size used in previous meta-analyses [
      • Bechthold A.
      • Boeing H.
      • Schwedhelm C.
      • Hoffmann G.
      • Knüppel S.
      • Iqbal K.
      • et al.
      Food groups and risk of coronary heart disease, stroke and heart failure: a systematic review and dose-response meta-analysis of prospective studies.
      ,
      • Grosso G.
      • Marventano S.
      • Yang J.
      • Micek A.
      • Pajak A.
      • Scalfi L.
      • et al.
      A comprehensive meta-analysis on evidence of Mediterranean diet and cardiovascular disease: are individual components equal?.
      ,
      • Marventano S.
      • Izquierdo Pulido M.
      • Sánchez-González C.
      • Godos J.
      • Speciani A.
      • Galvano F.
      • et al.
      Legume consumption and CVD risk: a systematic review and meta-analysis.
      ,
      • Becerra-Tomás N.
      • Papandreou C.
      • Salas-Salvadó J.
      Legume consumption and cardiometabolic health.
      ,
      • Viguiliouk E.
      • Glenn A.J.
      • Nishi S.K.
      • Chiavaroli L.
      • Seider M.
      • Khan T.
      • et al.
      Associations between dietary pulses alone or with other legumes and cardiometabolic disease outcomes: an umbrella review and updated systematic review and meta-analysis of prospective cohort studies.
      ,
      U.S. Department of Agriculture, Agricultural Research Service
      USDA food and nutrient database for dietary studies 2017-2018.
      ].

      2.4 Risk of bias assessment

      The internal validity of eligible studies was assessed independently by two authors (VM and ANi) using the ROBINS-I (Risk Of Bias In Non-randomised Studies of Interventions) tool [
      • Sterne J.A.C.
      • Hernán M.A.
      • Reeves B.C.
      • Savović J.
      • Berkman N.D.
      • Viswanathan M.
      • et al.
      ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions.
      ]. In case of disagreement, consensus was achieved involving a third author (AN). The following risk of bias domains were considered: (1) bias due to confounding, (2) bias in selection of participants into the study, (3) bias in classification of interventions, (4) bias due to deviations from intended interventions, (5) bias due to missing data, (6) bias in measurement of outcomes, (7) bias in selection of the reported results. For each domain, studies were judged to be at low, moderate, serious or critical risk of bias. The tool was tailored to address our specific research question and to accommodate the nature and methodology of studies selected for the analysis, in accordance with the ROBINS-I guidance document [
      • Sterne J.A.C.
      • Hernán M.A.
      • Reeves B.C.
      • Savović J.
      • Berkman N.D.
      • Viswanathan M.
      • et al.
      ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions.
      ].

      2.5 Statistical analysis

      We performed traditional meta-analyses to investigate the association between the intake of legumes and CVD, CHD and stroke by comparing the study-specific highest versus the lowest intake category. Summary RRs and respective 95% CIs were calculated by applying random-effects models [
      • DerSimonian R.
      • Laird N.
      Meta-analysis in clinical trials.
      ], as they are generally considered preferable to meta-analyse data presented in peer-reviewed publications [
      • Borenstein M.
      • Hedges L.V.
      • Higgins J.P.
      • Rothstein H.R.
      Introduction to meta-analysis.
      ] and to allow the generalization of results beyond the studies considered [
      • Tufanaru C.
      • Munn Z.
      • Stephenson M.
      • Aromataris E.
      Fixed or random effects meta-analysis? Common methodological issues in systematic reviews of effectiveness.
      ]. Moreover, in the overall analysis ORs were assumed to approximate RRs in accordance with the “rare disease assumption”, a practice commonly followed in this field [
      • Tang G.
      • Wang D.
      • Long J.
      • Yang F.
      • Si L.
      Meta-analysis of the association between whole grain intake and coronary heart disease risk.
      ,
      • Yan Z.
      • Zhang X.
      • Li C.
      • Jiao S.
      • Dong W.
      Association between consumption of soy and risk of cardiovascular disease: a meta-analysis of observational studies.
      ]. Heterogeneity was assessed using the I2 statistic [
      • Higgins J.P.
      • Thompson S.G.
      • Deeks J.J.
      • Altman D.G.
      Measuring inconsistency in meta-analyses.
      ]. In order to investigate possible sources of heterogeneity, we conducted subgroup analyses for each outcome of interest according to the a) study design (case–control vs. prospective cohort studies), b) study location (Asian vs non-Asian studies), c) follow-up time (<10 years vs. ≥10 years) d) validity of the dietary assessment method (validated vs. not validated), e) number of cases (<500 vs. ≥500 cases) and f) sex (females vs. males vs. both). Sensitivity analyses were performed for each outcome by excluding studies that specifically include soy in their exposure definition and studies that were classified as having a serious/critical risk of bias. Additional sensitivity analyses included the application of alternative standard values (i.e., ±15% and ±25% instead of ±20%) to estimate the highest and lowest exposure categories when these were open [
      • Filippini T.
      • Torres D.
      • Lopes C.
      • Carvalho C.
      • Moreira P.
      • Naska A.
      • et al.
      Cadmium exposure and risk of breast cancer: a dose-response meta-analysis of cohort studies.
      ]. Furthermore, all meta-analyses were repeated by removing each single study (leave-one-out method) in order to evaluate its influence on the pooled effect estimates. We assessed the possibility of publication bias through visual inspection of funnel plots and the Egger's regression test.
      Using the methodology established by Greenland and Longnecker (1992) [
      • Greenland S.
      • Longnecker M.P.
      Methods for trend estimation from summarized dose-response data, with applications to meta-analysis.
      ], and advanced by Orsini et al. (2012) [
      • Orsini N.
      • Li R.
      • Wolk A.
      • Khudyaknov P.
      • Spiegelman D.
      Meta-analysis for linear and nonlinear dose-response relations: examples, an evaluation of approximations, and software.
      ] and Crippa et al. (2019) [
      • Crippa A.
      • Discacciati A.
      • Bottai M.
      • Spiegelman D.
      • Orsini N.
      One-stage dose-response meta-analysis for aggregated data.
      ], we have performed a dose–response meta-analysis to identify the association between the intake of legumes and the risk of CVD, CHD and stroke. We applied the ‘one-stage’ approach [
      • Orsini N.
      • Li R.
      • Wolk A.
      • Khudyaknov P.
      • Spiegelman D.
      Meta-analysis for linear and nonlinear dose-response relations: examples, an evaluation of approximations, and software.
      ,
      • Crippa A.
      • Discacciati A.
      • Bottai M.
      • Spiegelman D.
      • Orsini N.
      One-stage dose-response meta-analysis for aggregated data.
      ], that allows to include studies assessing only two levels of exposure, using a restricted cubic spline model with 3 knots at fixed percentiles (10, 50, and 90%) of the intake distribution. The restricted cubic spline model was fitted with a generalized least-squares regression taking into account the correlation within each set of the effect estimates and combining the effect estimates using the restricted maximum likelihood method in a random-effects meta-analysis [
      • Orsini N.
      • Li R.
      • Wolk A.
      • Khudyaknov P.
      • Spiegelman D.
      Meta-analysis for linear and nonlinear dose-response relations: examples, an evaluation of approximations, and software.
      ,
      • Jackson D.
      • White I.R.
      • Thompson S.G.
      Extending DerSimonian and Laird's methodology to perform multivariate random effects meta-analyses.
      ,
      • Orsini N.
      • Bellocco R.
      • Greenland S.
      Generalized least squares for trend estimation of summarized dose–response data.
      ]. All analyses were performed with the Stata statistical software, version 14 (StatCorp, College Station, TX, USA) and RStudio (Version 1.1.456).

      3. Results

      The complete study selection process is described in Fig. 1. A total of 26 eligible articles from 24 unique prospective cohort and case–control studies assessing the association between the intake of legumes and CVD overall, CHD and/or stroke in particular were identified [
      • Mizrahi A.
      • Knekt P.
      • Montonen J.
      • Laasksonen M.A.
      • Heliövaaea M.
      • Järvinen R.
      Plant foods and the risk of cerebrovascular diseases: a potential protection of fruit consumption.
      ,
      • Atkins J.L.
      • Whincup P.H.
      • Morris L.T.
      • Lennon L.T.
      • Papacosta O.
      • Wannamethee S.G.
      High diet quality is associated with a lower risk of cardiovascular disease and all-cause mortality in older men.
      ,
      • Baik I.
      • Cho N.H.
      • Kim S.H.
      • Shin C.
      Dietary information improves cardiovascular disease risk prediction models.
      ,
      • Bazzano L.A.
      • He J.
      • Ogden L.G.
      • Loria C.
      • Vupputuri S.
      • Myers L.
      • et al.
      Legume consumption and risk of coronary heart disease in US men and women: NHANES I Epidemiologic Follow-up Study.
      ,
      • Bernstein A.M.
      • Pan A.
      • Rexrode K.M.
      • Stampfer M.
      • Hu F.B.
      • Mozaffarian D.
      • et al.
      Dietary protein sources and the risk of stroke in men and women.
      ,
      • Bernstein A.M.
      • Sun Q.
      • Hu F.B.
      • Stampfer M.J.
      • Manson J.E.
      • Willett W.C.
      Major dietary protein sources and risk of coronary heart disease in women.
      ,
      • Durga A.V.
      • Manorenj S.
      Dietary pattern in adult patients with acute stroke in South India: a case-control study from a tertiary care center in Hyderabad.
      ,
      • Fraser G.E.
      • Sabaté J.
      • Beeson W.L.
      • Strahan T.M.
      A possible protective effect of nut consumption on risk of coronary heart disease. The Adventist Health Study.
      ,
      • Fung T.T.
      • Isanaka S.
      • Hu F.B.
      • Willett W.C.
      International food group-based diet quality and risk of coronary heart disease in men and women.
      ,
      • Haring B.
      • Gronroos N.
      • Nettleton J.A.
      • von Ballmoos M.C.
      • Selvin E.
      • Alonso A.
      Dietary protein intake and coronary heart disease in a large community based cohort: results from the Atherosclerosis Risk in Communities (ARIC) study [corrected].
      ,
      • Haring B.
      • Misialek J.R.
      • Rebholz C.M.
      • Petruski-Ivleva N.
      • Gottesman R.F.
      • Mosley T.H.
      • et al.
      Association of dietary protein consumption with incident silent cerebral infarcts and stroke: the atherosclerosis risk in Communities (ARIC) study.
      ,
      • Kabagambe E.K.
      • Baylin A.
      • Ruiz-Narvarez E.
      • Siles X.
      • Campos H.
      Decreased consumption of dried mature beans is positively associated with urbanization and nonfatal acute myocardial infarction.
      ,
      • Kokubo Y.
      • Iso H.
      • Ishihara J.
      • Okada K.
      • Inoue M.
      • Tsugane S.
      Association of dietary intake of soy, beans, and isoflavones with risk of cerebral and myocardial infarctions in Japanese populations: the Japan Public Health Center-based (JPHC) study cohort I.
      ,
      • Maher M.A.
      • Gutbi S.S.
      Assessment of dietary pattern among coronary heart disease outpatients attended El-Shaap teaching hospital, Khartoum state.
      ,
      • Martínez-González M.A.
      • Fernández-Jarne E.
      • Martínez-Losa E.
      • Prado-Santamaría M.
      • Brugarolas-Brufau C.
      • Serrano-Martinez M.
      Role of fibre and fruit in the Mediterranean diet to protect against myocardial infarction: a case-control study in Spain.
      ,
      • Martínez-González M.A.
      • García-López M.
      • Bes-Rastollo M.
      • Toledo E.
      • Martínez-Lapiscina E.H.
      • Delgado-Rodriguez M.
      • et al.
      Mediterranean diet and the incidence of cardiovascular disease: a Spanish cohort.
      ,
      • Mata-Fernández A.
      • Hershey M.S.
      • Pastrana-Delgado J.C.
      • Sotos-Prieto M.
      • Ruiz-Canela M.
      • Kales S.N.
      • et al.
      A Mediterranean lifestyle reduces the risk of cardiovascular disease in the “Seguimiento Universidad de Navarra” (SUN) cohort.
      ,
      • Miller V.
      • Mente A.
      • Dehghan M.
      • Rangarajan S.
      • Zhang X.
      • Swaminathan S.
      • et al.
      Fruit, vegetable, and legume intake, and cardiovascular disease and deaths in 18 countries (PURE): a prospective cohort study.
      ,
      • Nouri F.
      • Haghighadoost F.
      • Mohammadifard N.
      • Mansourian M.
      • Sadeghi M.
      • Roohafza H.
      • et al.
      The longitudinal association between soybean and non-soybean legumes intakes and risk of cardiovascular disease: Isfahan cohort study.
      ,
      • Perez-Cornago A.
      • Crowe F.L.
      • Appleby P.N.
      • Bradbury K.E.
      • Wood A.M.
      • Jakobsen M.U.
      • et al.
      Plant foods, dietary fibre and risk of ischaemic heart disease in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort.
      ,
      • Scarmeas N.
      • Luchsinger J.A.
      • Stern Y.
      • Gu Y.
      • He J.
      • DeCarli C.
      • et al.
      Mediterranean diet and magnetic resonance imaging-assessed cerebrovascular disease.
      ,
      • Schröder H.
      • Salas-Salvadó J.
      • Martínez-González M.A.
      • Fíto M.
      • Corella D.
      • Estruch R.
      • et al.
      Baseline adherence to the Mediterranean diet and major cardiovascular events: prevención con Dieta Mediterránea trial.
      ,
      • Tektonidis T.G.
      • Åkesson A.
      • Gigante B.
      • Wolk A.
      • Larsson S.C.
      A Mediterranean diet and risk of myocardial infarction, heart failure and stroke: a population-based cohort study.
      ,
      • Tong T.Y.N.
      • Appleby P.N.
      • Key T.J.
      • Dahm C.C.
      • Overvad K.
      • Olsen A.
      • et al.
      The associations of major foods and fibre with risks of ischaemic and haemorrhagic stroke: a prospective study of 418 329 participants in the EPIC cohort across nine European countries.
      ,
      • Turati F.
      • Pelucchi C.
      • Galeone C.
      • Praud D.
      • Tavani A.
      • La Vecchia C.
      Mediterranean diet and non-fatal acute myocardial infarction: a case-control study from Italy.
      ,
      • Yu D.
      • Zhang X.
      • Gao Y.T.
      • Li H.
      • Yang G.
      • Huang J.
      • et al.
      Fruit and vegetable intake and risk of CHD: results from prospective cohort studies of Chinese adults in Shanghai.
      ]. Study characteristics are summarized in Table 1 and a detailed description is available in Supplemental Tables 2–4.
      Figure 1
      Figure 1Flow-chart describing the systematic literature search and study selection for the association between the intake of legumes and CVD, CHD and Stroke.
      Table 1Summary description of studies included in the systematic review and meta-analysis (N = 26).
      Study referenceCountryCohort (follow-up duration, y)SexSample sizeAge(y) Mean/RangeType of legumes includedOutcomesExposure contrast
      Prospective cohort studies
      Atkins JL et al., 2014 [
      • Atkins J.L.
      • Whincup P.H.
      • Morris L.T.
      • Lennon L.T.
      • Papacosta O.
      • Wannamethee S.G.
      High diet quality is associated with a lower risk of cardiovascular disease and all-cause mortality in older men.
      ]
      UKBRHS (11.3)M332860–79Baked or butter beans, lentils, peas, chickpeas, sweetcornCVD, CHD incidenceHighest vs. lowest
      Baik I et al., 2013 [
      • Baik I.
      • Cho N.H.
      • Kim S.H.
      • Shin C.
      Dietary information improves cardiovascular disease risk prediction models.
      ]
      South KoreaKGES (8)FM902640–69Not specifiedCVD incidenceAbove or below the median
      Bazzano LA et al., 2001 [
      • Bazzano L.A.
      • He J.
      • Ogden L.G.
      • Loria C.
      • Vupputuri S.
      • Myers L.
      • et al.
      Legume consumption and risk of coronary heart disease in US men and women: NHANES I Epidemiologic Follow-up Study.
      ]
      USANHEFS (19)FM963225–74Beans, lentils and chickpeasCVD, CHD incidenceQuintiles
      Bernstein AM et al., 2010 [
      • Bernstein A.M.
      • Sun Q.
      • Hu F.B.
      • Stampfer M.J.
      • Manson J.E.
      • Willett W.C.
      Major dietary protein sources and risk of coronary heart disease in women.
      ]
      USANHS (26)F84 13630–55Beans or lentils, baked or dryCHD incidenceQuintiles
      Bernstein AM et al., 2012 [
      • Bernstein A.M.
      • Pan A.
      • Rexrode K.M.
      • Stampfer M.
      • Hu F.B.
      • Mozaffarian D.
      • et al.
      Dietary protein sources and the risk of stroke in men and women.
      ]
      USANHS (26)F84 01030–55Dry beans, peas, soy, and tofuStroke incidenceQuintiles
      USAHPFS (22)M43 15040–75Dry beans, peas, soy, and tofuStroke incidenceQuintiles
      Fraser GE et al., 1992 [
      • Fraser G.E.
      • Sabaté J.
      • Beeson W.L.
      • Strahan T.M.
      A possible protective effect of nut consumption on risk of coronary heart disease. The Adventist Health Study.
      ]
      USAAHS (6)FM26 47352Beans and peasCHD incidenceTertiles
      Fung TT et al., 2018 [
      • Fung T.T.
      • Isanaka S.
      • Hu F.B.
      • Willett W.C.
      International food group-based diet quality and risk of coronary heart disease in men and women.
      ]
      USANHS II (20)F93 13127–44Dry beans, peas, soy, and tofuCHD incidenceQuintiles
      USAHPFS (24)M43 96640–75Dry beans, peas, soy, and tofuCHD incidenceQuintiles
      Haring B et al., 2014 [
      • Haring B.
      • Gronroos N.
      • Nettleton J.A.
      • von Ballmoos M.C.
      • Selvin E.
      • Alonso A.
      Dietary protein intake and coronary heart disease in a large community based cohort: results from the Atherosclerosis Risk in Communities (ARIC) study [corrected].
      ]
      USAARIC (22)FM12 06645–64Not specifiedCHD incidenceQuintiles
      Haring B et al., 2015 [
      • Haring B.
      • Misialek J.R.
      • Rebholz C.M.
      • Petruski-Ivleva N.
      • Gottesman R.F.
      • Mosley T.H.
      • et al.
      Association of dietary protein consumption with incident silent cerebral infarcts and stroke: the atherosclerosis risk in Communities (ARIC) study.
      ]
      USAARIC (22.7)FM11 60145–64Not specifiedStroke incidenceQuintiles
      Kokubo Y et al., 2007 [
      • Kokubo Y.
      • Iso H.
      • Ishihara J.
      • Okada K.
      • Inoue M.
      • Tsugane S.
      Association of dietary intake of soy, beans, and isoflavones with risk of cerebral and myocardial infarctions in Japanese populations: the Japan Public Health Center-based (JPHC) study cohort I.
      ]
      JapanJPHC (12.5)FM40 45040–59BeansCVD, CHD, Stroke incidenceTertiles
      Mata-Fernández A et al., 2021 [
      • Mata-Fernández A.
      • Hershey M.S.
      • Pastrana-Delgado J.C.
      • Sotos-Prieto M.
      • Ruiz-Canela M.
      • Kales S.N.
      • et al.
      A Mediterranean lifestyle reduces the risk of cardiovascular disease in the “Seguimiento Universidad de Navarra” (SUN) cohort.
      ]
      SpainSUN (11.5)FM18 63138.1Lentils, peas, chickpeas and beansCVD incidenceAbove or below the median
      Miller V et al., 2017 [
      • Miller V.
      • Mente A.
      • Dehghan M.
      • Rangarajan S.
      • Zhang X.
      • Swaminathan S.
      • et al.
      Fruit, vegetable, and legume intake, and cardiovascular disease and deaths in 18 countries (PURE): a prospective cohort study.
      ]
      MultinationalPURE (7.4)FM135 33535–70Beans, black beans, lentils, peas, chickpeas and black-eyed peasCVD, CHD, Stroke incidenceQuintiles
      Martínez-González MA et al., 2011 [
      • Martínez-González M.A.
      • García-López M.
      • Bes-Rastollo M.
      • Toledo E.
      • Martínez-Lapiscina E.H.
      • Delgado-Rodriguez M.
      • et al.
      Mediterranean diet and the incidence of cardiovascular disease: a Spanish cohort.
      ]
      SpainSUN (4.9)FM13 60938Lentils, peas, chickpeas and beansCHD incidenceMedian
      Mizrahi A et al., 2009 [
      • Mizrahi A.
      • Knekt P.
      • Montonen J.
      • Laasksonen M.A.
      • Heliövaaea M.
      • Järvinen R.
      Plant foods and the risk of cerebrovascular diseases: a potential protection of fruit consumption.
      ]
      FinlandFMCHES (24)FM393240–74Not specifiedStroke incidenceQuartiles
      Nouri F et al., 2021 [
      • Nouri F.
      • Haghighadoost F.
      • Mohammadifard N.
      • Mansourian M.
      • Sadeghi M.
      • Roohafza H.
      • et al.
      The longitudinal association between soybean and non-soybean legumes intakes and risk of cardiovascular disease: Isfahan cohort study.
      ]
      IranICS (13)FM5432≥35Beans, chickpeas, lentilsCVD incidenceTertiles
      Perez-Cornago A et al., 2021 [
      • Perez-Cornago A.
      • Crowe F.L.
      • Appleby P.N.
      • Bradbury K.E.
      • Wood A.M.
      • Jakobsen M.U.
      • et al.
      Plant foods, dietary fibre and risk of ischaemic heart disease in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort.
      ]
      Multinational/EuropeEPIC (12.6)FM490 31135–70Not specifiedCHD incidenceQuintiles
      Scarmeas N et al., 2011 [
      • Scarmeas N.
      • Luchsinger J.A.
      • Stern Y.
      • Gu Y.
      • He J.
      • DeCarli C.
      • et al.
      Mediterranean diet and magnetic resonance imaging-assessed cerebrovascular disease.
      ]
      USAWHICAP (5.8)FM707≥65Not specifiedStroke incidenceAbove or below the median
      Schröder H et al., 2014 [
      • Schröder H.
      • Salas-Salvadó J.
      • Martínez-González M.A.
      • Fíto M.
      • Corella D.
      • Estruch R.
      • et al.
      Baseline adherence to the Mediterranean diet and major cardiovascular events: prevención con Dieta Mediterránea trial.
      ]
      SpainPREDIMED (4.8)FM744767Not specifiedCVD incidenceHighest vs. lowest
      Tektonidis TG et al., 2015 [
      • Tektonidis T.G.
      • Åkesson A.
      • Gigante B.
      • Wolk A.
      • Larsson S.C.
      A Mediterranean diet and risk of myocardial infarction, heart failure and stroke: a population-based cohort study.
      ]
      SwedenSMC (10.4)F32 92148–83Not specifiedCHD incidence, Ischaemic and haemorrhagic stroke incidenceAbove or below the median
      Tong TYN et al., 2020 [
      • Tong T.Y.N.
      • Appleby P.N.
      • Key T.J.
      • Dahm C.C.
      • Overvad K.
      • Olsen A.
      • et al.
      The associations of major foods and fibre with risks of ischaemic and haemorrhagic stroke: a prospective study of 418 329 participants in the EPIC cohort across nine European countries.
      ]
      Multinational/EuropeEPIC (12.7)FM418 32951.2Beans, chickpeas, split peas, lentilsStroke incidenceQuintiles
      Yu D et al., 2014 [
      • Yu D.
      • Zhang X.
      • Gao Y.T.
      • Li H.
      • Yang G.
      • Huang J.
      • et al.
      Fruit and vegetable intake and risk of CHD: results from prospective cohort studies of Chinese adults in Shanghai.
      ]
      ChinaSWHS (9.8)F67 21140–70Fresh & snow peas, soya beans, broad, long and green beansCHD incidenceQuartiles
      ChinaSMHS (5.4)M55 47440–74Fresh & snow peas, soya beans, broad, long and green beansCHD incidenceQuartiles
      Case-control Studies
      Durga AV & Manorenj S 2019 [
      • Durga A.V.
      • Manorenj S.
      Dietary pattern in adult patients with acute stroke in South India: a case-control study from a tertiary care center in Hyderabad.
      ]
      IndiaFM30019–80Pulses (Red, green, black and bengal gram dhal & whole pulses)StrokeHighest vs. lowest
      Kabagambe, EK et al., 2005 [
      • Kabagambe E.K.
      • Baylin A.
      • Ruiz-Narvarez E.
      • Siles X.
      • Campos H.
      Decreased consumption of dried mature beans is positively associated with urbanization and nonfatal acute myocardial infarction.
      ]
      Costa RicaFM4238<75Dried mature beansCHDQuartiles
      Maher, MA & Gutbi SS 2017 [
      • Maher M.A.
      • Gutbi S.S.
      Assessment of dietary pattern among coronary heart disease outpatients attended El-Shaap teaching hospital, Khartoum state.
      ]
      SudanFM100∼53Not specifiedCHDHighest vs. lowest
      Martínez-González MA et al., 2002 [
      • Martínez-González M.A.
      • Fernández-Jarne E.
      • Martínez-Losa E.
      • Prado-Santamaría M.
      • Brugarolas-Brufau C.
      • Serrano-Martinez M.
      Role of fibre and fruit in the Mediterranean diet to protect against myocardial infarction: a case-control study in Spain.
      ],
      SpainFM342<80Not specifiedCHDQuintiles
      Turati F et al., 2015 [
      • Turati F.
      • Pelucchi C.
      • Galeone C.
      • Praud D.
      • Tavani A.
      • La Vecchia C.
      Mediterranean diet and non-fatal acute myocardial infarction: a case-control study from Italy.
      ]
      ItalyFM144219–79Not specifiedCHDAbove or below the median
      AHS: Adventist Health Study; ARIC: Atherosclerosis Risk in Communities; BRHS: British Regional Heart Study; CHD: Coronary Heart Disease; CI: Confidence Interval; CVD: Cardiovascular Disease; EPIC: European Prospective Investigation into Cancer and Nutrition; FMCHES: Finnish Mobile Clinic Health Examination Survey; HPFS: Health Professionals' Follow-up Study; ICS: Isfahan Cohort Study; JPHC: Japan Public Health Center–Based Study; KGES: Korean Genome Epidemiology Study; NHEFS: National Health and Nutrition Examination Survey Epidemiologic Follow-up Study; NHS: Nurses' Health Study; PREDIMED: Prevention with Mediterranean Diet; PURE: Prospective Urban Rural Epidemiology; RoB: Risk of bias; SMC: Swedish Mammography Cohort; SMHS: Shangai Men's Health Study; SUN: Seguimiento University of Navarra; SWHS: Shangai Women's Health Study; WHICAP: Washington Heights/Hamilton Heights Columbia Aging Project.

      3.1 Legumes and cardiovascular disease

      Out of the total of 26 studies identified in the systematic review, Martínez-González MA et al., 2011 [
      • Martínez-González M.A.
      • García-López M.
      • Bes-Rastollo M.
      • Toledo E.
      • Martínez-Lapiscina E.H.
      • Delgado-Rodriguez M.
      • et al.
      Mediterranean diet and the incidence of cardiovascular disease: a Spanish cohort.
      ] was excluded from the analyses addressing the association between the intake of legumes and CVD overall for being performed using the same sample as Mata-Fernández A et al., 2021 [
      • Mata-Fernández A.
      • Hershey M.S.
      • Pastrana-Delgado J.C.
      • Sotos-Prieto M.
      • Ruiz-Canela M.
      • Kales S.N.
      • et al.
      A Mediterranean lifestyle reduces the risk of cardiovascular disease in the “Seguimiento Universidad de Navarra” (SUN) cohort.
      ]. Twenty-five studies were thus included, comprising as outcomes CVD as a composite measure, CHD or stroke (Supplemental Table 2). Of these, twenty were prospective cohorts [
      • Mizrahi A.
      • Knekt P.
      • Montonen J.
      • Laasksonen M.A.
      • Heliövaaea M.
      • Järvinen R.
      Plant foods and the risk of cerebrovascular diseases: a potential protection of fruit consumption.
      ,
      • Atkins J.L.
      • Whincup P.H.
      • Morris L.T.
      • Lennon L.T.
      • Papacosta O.
      • Wannamethee S.G.
      High diet quality is associated with a lower risk of cardiovascular disease and all-cause mortality in older men.
      ,
      • Baik I.
      • Cho N.H.
      • Kim S.H.
      • Shin C.
      Dietary information improves cardiovascular disease risk prediction models.
      ,
      • Bazzano L.A.
      • He J.
      • Ogden L.G.
      • Loria C.
      • Vupputuri S.
      • Myers L.
      • et al.
      Legume consumption and risk of coronary heart disease in US men and women: NHANES I Epidemiologic Follow-up Study.
      ,
      • Bernstein A.M.
      • Pan A.
      • Rexrode K.M.
      • Stampfer M.
      • Hu F.B.
      • Mozaffarian D.
      • et al.
      Dietary protein sources and the risk of stroke in men and women.
      ,
      • Bernstein A.M.
      • Sun Q.
      • Hu F.B.
      • Stampfer M.J.
      • Manson J.E.
      • Willett W.C.
      Major dietary protein sources and risk of coronary heart disease in women.
      ,
      • Fraser G.E.
      • Sabaté J.
      • Beeson W.L.
      • Strahan T.M.
      A possible protective effect of nut consumption on risk of coronary heart disease. The Adventist Health Study.
      ,
      • Fung T.T.
      • Isanaka S.
      • Hu F.B.
      • Willett W.C.
      International food group-based diet quality and risk of coronary heart disease in men and women.
      ,
      • Haring B.
      • Gronroos N.
      • Nettleton J.A.
      • von Ballmoos M.C.
      • Selvin E.
      • Alonso A.
      Dietary protein intake and coronary heart disease in a large community based cohort: results from the Atherosclerosis Risk in Communities (ARIC) study [corrected].
      ,
      • Haring B.
      • Misialek J.R.
      • Rebholz C.M.
      • Petruski-Ivleva N.
      • Gottesman R.F.
      • Mosley T.H.
      • et al.
      Association of dietary protein consumption with incident silent cerebral infarcts and stroke: the atherosclerosis risk in Communities (ARIC) study.
      ,
      • Kokubo Y.
      • Iso H.
      • Ishihara J.
      • Okada K.
      • Inoue M.
      • Tsugane S.
      Association of dietary intake of soy, beans, and isoflavones with risk of cerebral and myocardial infarctions in Japanese populations: the Japan Public Health Center-based (JPHC) study cohort I.
      ,
      • Mata-Fernández A.
      • Hershey M.S.
      • Pastrana-Delgado J.C.
      • Sotos-Prieto M.
      • Ruiz-Canela M.
      • Kales S.N.
      • et al.
      A Mediterranean lifestyle reduces the risk of cardiovascular disease in the “Seguimiento Universidad de Navarra” (SUN) cohort.
      ,
      • Miller V.
      • Mente A.
      • Dehghan M.
      • Rangarajan S.
      • Zhang X.
      • Swaminathan S.
      • et al.
      Fruit, vegetable, and legume intake, and cardiovascular disease and deaths in 18 countries (PURE): a prospective cohort study.
      ,
      • Nouri F.
      • Haghighadoost F.
      • Mohammadifard N.
      • Mansourian M.
      • Sadeghi M.
      • Roohafza H.
      • et al.
      The longitudinal association between soybean and non-soybean legumes intakes and risk of cardiovascular disease: Isfahan cohort study.
      ,
      • Perez-Cornago A.
      • Crowe F.L.
      • Appleby P.N.
      • Bradbury K.E.
      • Wood A.M.
      • Jakobsen M.U.
      • et al.
      Plant foods, dietary fibre and risk of ischaemic heart disease in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort.
      ,
      • Scarmeas N.
      • Luchsinger J.A.
      • Stern Y.
      • Gu Y.
      • He J.
      • DeCarli C.
      • et al.
      Mediterranean diet and magnetic resonance imaging-assessed cerebrovascular disease.
      ,
      • Schröder H.
      • Salas-Salvadó J.
      • Martínez-González M.A.
      • Fíto M.
      • Corella D.
      • Estruch R.
      • et al.
      Baseline adherence to the Mediterranean diet and major cardiovascular events: prevención con Dieta Mediterránea trial.
      ,
      • Tektonidis T.G.
      • Åkesson A.
      • Gigante B.
      • Wolk A.
      • Larsson S.C.
      A Mediterranean diet and risk of myocardial infarction, heart failure and stroke: a population-based cohort study.
      ,
      • Tong T.Y.N.
      • Appleby P.N.
      • Key T.J.
      • Dahm C.C.
      • Overvad K.
      • Olsen A.
      • et al.
      The associations of major foods and fibre with risks of ischaemic and haemorrhagic stroke: a prospective study of 418 329 participants in the EPIC cohort across nine European countries.
      ,
      • Yu D.
      • Zhang X.
      • Gao Y.T.
      • Li H.
      • Yang G.
      • Huang J.
      • et al.
      Fruit and vegetable intake and risk of CHD: results from prospective cohort studies of Chinese adults in Shanghai.
      ] (mean follow-up: 14.6 y) and five were case–control studies [
      • Durga A.V.
      • Manorenj S.
      Dietary pattern in adult patients with acute stroke in South India: a case-control study from a tertiary care center in Hyderabad.
      ,
      • Kabagambe E.K.
      • Baylin A.
      • Ruiz-Narvarez E.
      • Siles X.
      • Campos H.
      Decreased consumption of dried mature beans is positively associated with urbanization and nonfatal acute myocardial infarction.
      ,
      • Maher M.A.
      • Gutbi S.S.
      Assessment of dietary pattern among coronary heart disease outpatients attended El-Shaap teaching hospital, Khartoum state.
      ,
      • Martínez-González M.A.
      • Fernández-Jarne E.
      • Martínez-Losa E.
      • Prado-Santamaría M.
      • Brugarolas-Brufau C.
      • Serrano-Martinez M.
      Role of fibre and fruit in the Mediterranean diet to protect against myocardial infarction: a case-control study in Spain.
      ,
      • Turati F.
      • Pelucchi C.
      • Galeone C.
      • Praud D.
      • Tavani A.
      • La Vecchia C.
      Mediterranean diet and non-fatal acute myocardial infarction: a case-control study from Italy.
      ]. One study was performed in Africa [
      • Maher M.A.
      • Gutbi S.S.
      Assessment of dietary pattern among coronary heart disease outpatients attended El-Shaap teaching hospital, Khartoum state.
      ], six in Asia [
      • Baik I.
      • Cho N.H.
      • Kim S.H.
      • Shin C.
      Dietary information improves cardiovascular disease risk prediction models.
      ,
      • Durga A.V.
      • Manorenj S.
      Dietary pattern in adult patients with acute stroke in South India: a case-control study from a tertiary care center in Hyderabad.
      ,
      • Kokubo Y.
      • Iso H.
      • Ishihara J.
      • Okada K.
      • Inoue M.
      • Tsugane S.
      Association of dietary intake of soy, beans, and isoflavones with risk of cerebral and myocardial infarctions in Japanese populations: the Japan Public Health Center-based (JPHC) study cohort I.
      ,
      • Miller V.
      • Mente A.
      • Dehghan M.
      • Rangarajan S.
      • Zhang X.
      • Swaminathan S.
      • et al.
      Fruit, vegetable, and legume intake, and cardiovascular disease and deaths in 18 countries (PURE): a prospective cohort study.
      ,
      • Nouri F.
      • Haghighadoost F.
      • Mohammadifard N.
      • Mansourian M.
      • Sadeghi M.
      • Roohafza H.
      • et al.
      The longitudinal association between soybean and non-soybean legumes intakes and risk of cardiovascular disease: Isfahan cohort study.
      ,
      • Yu D.
      • Zhang X.
      • Gao Y.T.
      • Li H.
      • Yang G.
      • Huang J.
      • et al.
      Fruit and vegetable intake and risk of CHD: results from prospective cohort studies of Chinese adults in Shanghai.
      ], nine in Europe [
      • Mizrahi A.
      • Knekt P.
      • Montonen J.
      • Laasksonen M.A.
      • Heliövaaea M.
      • Järvinen R.
      Plant foods and the risk of cerebrovascular diseases: a potential protection of fruit consumption.
      ,
      • Atkins J.L.
      • Whincup P.H.
      • Morris L.T.
      • Lennon L.T.
      • Papacosta O.
      • Wannamethee S.G.
      High diet quality is associated with a lower risk of cardiovascular disease and all-cause mortality in older men.
      ,
      • Martínez-González M.A.
      • Fernández-Jarne E.
      • Martínez-Losa E.
      • Prado-Santamaría M.
      • Brugarolas-Brufau C.
      • Serrano-Martinez M.
      Role of fibre and fruit in the Mediterranean diet to protect against myocardial infarction: a case-control study in Spain.
      ,
      • Mata-Fernández A.
      • Hershey M.S.
      • Pastrana-Delgado J.C.
      • Sotos-Prieto M.
      • Ruiz-Canela M.
      • Kales S.N.
      • et al.
      A Mediterranean lifestyle reduces the risk of cardiovascular disease in the “Seguimiento Universidad de Navarra” (SUN) cohort.
      ,
      • Perez-Cornago A.
      • Crowe F.L.
      • Appleby P.N.
      • Bradbury K.E.
      • Wood A.M.
      • Jakobsen M.U.
      • et al.
      Plant foods, dietary fibre and risk of ischaemic heart disease in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort.
      ,
      • Schröder H.
      • Salas-Salvadó J.
      • Martínez-González M.A.
      • Fíto M.
      • Corella D.
      • Estruch R.
      • et al.
      Baseline adherence to the Mediterranean diet and major cardiovascular events: prevención con Dieta Mediterránea trial.
      ,
      • Tektonidis T.G.
      • Åkesson A.
      • Gigante B.
      • Wolk A.
      • Larsson S.C.
      A Mediterranean diet and risk of myocardial infarction, heart failure and stroke: a population-based cohort study.
      ,
      • Tong T.Y.N.
      • Appleby P.N.
      • Key T.J.
      • Dahm C.C.
      • Overvad K.
      • Olsen A.
      • et al.
      The associations of major foods and fibre with risks of ischaemic and haemorrhagic stroke: a prospective study of 418 329 participants in the EPIC cohort across nine European countries.
      ,
      • Turati F.
      • Pelucchi C.
      • Galeone C.
      • Praud D.
      • Tavani A.
      • La Vecchia C.
      Mediterranean diet and non-fatal acute myocardial infarction: a case-control study from Italy.
      ] and nine in North America [
      • Bazzano L.A.
      • He J.
      • Ogden L.G.
      • Loria C.
      • Vupputuri S.
      • Myers L.
      • et al.
      Legume consumption and risk of coronary heart disease in US men and women: NHANES I Epidemiologic Follow-up Study.
      ,
      • Bernstein A.M.
      • Pan A.
      • Rexrode K.M.
      • Stampfer M.
      • Hu F.B.
      • Mozaffarian D.
      • et al.
      Dietary protein sources and the risk of stroke in men and women.
      ,
      • Bernstein A.M.
      • Sun Q.
      • Hu F.B.
      • Stampfer M.J.
      • Manson J.E.
      • Willett W.C.
      Major dietary protein sources and risk of coronary heart disease in women.
      ,
      • Fraser G.E.
      • Sabaté J.
      • Beeson W.L.
      • Strahan T.M.
      A possible protective effect of nut consumption on risk of coronary heart disease. The Adventist Health Study.
      ,
      • Fung T.T.
      • Isanaka S.
      • Hu F.B.
      • Willett W.C.
      International food group-based diet quality and risk of coronary heart disease in men and women.
      ,
      • Haring B.
      • Gronroos N.
      • Nettleton J.A.
      • von Ballmoos M.C.
      • Selvin E.
      • Alonso A.
      Dietary protein intake and coronary heart disease in a large community based cohort: results from the Atherosclerosis Risk in Communities (ARIC) study [corrected].
      ,
      • Haring B.
      • Misialek J.R.
      • Rebholz C.M.
      • Petruski-Ivleva N.
      • Gottesman R.F.
      • Mosley T.H.
      • et al.
      Association of dietary protein consumption with incident silent cerebral infarcts and stroke: the atherosclerosis risk in Communities (ARIC) study.
      ,
      • Kabagambe E.K.
      • Baylin A.
      • Ruiz-Narvarez E.
      • Siles X.
      • Campos H.
      Decreased consumption of dried mature beans is positively associated with urbanization and nonfatal acute myocardial infarction.
      ,
      • Scarmeas N.
      • Luchsinger J.A.
      • Stern Y.
      • Gu Y.
      • He J.
      • DeCarli C.
      • et al.
      Mediterranean diet and magnetic resonance imaging-assessed cerebrovascular disease.
      ]. The pooled sample included a total of 1 703 121 participants and 44 181 cases and age ranged between 19 and 83 years. All studies used Food Frequency Questionnaires (FFQs) for the dietary assessment, with the exception of Mizrahi, A et al. (2009) [
      • Mizrahi A.
      • Knekt P.
      • Montonen J.
      • Laasksonen M.A.
      • Heliövaaea M.
      • Järvinen R.
      Plant foods and the risk of cerebrovascular diseases: a potential protection of fruit consumption.
      ] who relied on interviewer-administered diet history. Three studies have explicitly included soy-legumes in their definition along with non-soy ones [
      • Bernstein A.M.
      • Pan A.
      • Rexrode K.M.
      • Stampfer M.
      • Hu F.B.
      • Mozaffarian D.
      • et al.
      Dietary protein sources and the risk of stroke in men and women.
      ,
      • Fung T.T.
      • Isanaka S.
      • Hu F.B.
      • Willett W.C.
      International food group-based diet quality and risk of coronary heart disease in men and women.
      ,
      • Yu D.
      • Zhang X.
      • Gao Y.T.
      • Li H.
      • Yang G.
      • Huang J.
      • et al.
      Fruit and vegetable intake and risk of CHD: results from prospective cohort studies of Chinese adults in Shanghai.
      ], while eleven have explicitly excluded them [
      • Atkins J.L.
      • Whincup P.H.
      • Morris L.T.
      • Lennon L.T.
      • Papacosta O.
      • Wannamethee S.G.
      High diet quality is associated with a lower risk of cardiovascular disease and all-cause mortality in older men.
      ,
      • Bazzano L.A.
      • He J.
      • Ogden L.G.
      • Loria C.
      • Vupputuri S.
      • Myers L.
      • et al.
      Legume consumption and risk of coronary heart disease in US men and women: NHANES I Epidemiologic Follow-up Study.
      ,
      • Bernstein A.M.
      • Sun Q.
      • Hu F.B.
      • Stampfer M.J.
      • Manson J.E.
      • Willett W.C.
      Major dietary protein sources and risk of coronary heart disease in women.
      ,
      • Durga A.V.
      • Manorenj S.
      Dietary pattern in adult patients with acute stroke in South India: a case-control study from a tertiary care center in Hyderabad.
      ,
      • Fraser G.E.
      • Sabaté J.
      • Beeson W.L.
      • Strahan T.M.
      A possible protective effect of nut consumption on risk of coronary heart disease. The Adventist Health Study.
      ,
      • Kabagambe E.K.
      • Baylin A.
      • Ruiz-Narvarez E.
      • Siles X.
      • Campos H.
      Decreased consumption of dried mature beans is positively associated with urbanization and nonfatal acute myocardial infarction.
      ,
      • Kokubo Y.
      • Iso H.
      • Ishihara J.
      • Okada K.
      • Inoue M.
      • Tsugane S.
      Association of dietary intake of soy, beans, and isoflavones with risk of cerebral and myocardial infarctions in Japanese populations: the Japan Public Health Center-based (JPHC) study cohort I.
      ,
      • Mata-Fernández A.
      • Hershey M.S.
      • Pastrana-Delgado J.C.
      • Sotos-Prieto M.
      • Ruiz-Canela M.
      • Kales S.N.
      • et al.
      A Mediterranean lifestyle reduces the risk of cardiovascular disease in the “Seguimiento Universidad de Navarra” (SUN) cohort.
      ,
      • Miller V.
      • Mente A.
      • Dehghan M.
      • Rangarajan S.
      • Zhang X.
      • Swaminathan S.
      • et al.
      Fruit, vegetable, and legume intake, and cardiovascular disease and deaths in 18 countries (PURE): a prospective cohort study.
      ,
      • Nouri F.
      • Haghighadoost F.
      • Mohammadifard N.
      • Mansourian M.
      • Sadeghi M.
      • Roohafza H.
      • et al.
      The longitudinal association between soybean and non-soybean legumes intakes and risk of cardiovascular disease: Isfahan cohort study.
      ,
      • Tong T.Y.N.
      • Appleby P.N.
      • Key T.J.
      • Dahm C.C.
      • Overvad K.
      • Olsen A.
      • et al.
      The associations of major foods and fibre with risks of ischaemic and haemorrhagic stroke: a prospective study of 418 329 participants in the EPIC cohort across nine European countries.
      ]. For the remaining eleven studies, no details on the exposure definition were provided. CVD cases were ascertained using medical records and vital statistics, with the exception of: Baik, I et al. (2013) [
      • Baik I.
      • Cho N.H.
      • Kim S.H.
      • Shin C.
      Dietary information improves cardiovascular disease risk prediction models.
      ] (based on self-reporting and vital statistics); Scarmeas N et al. (2011) [
      • Scarmeas N.
      • Luchsinger J.A.
      • Stern Y.
      • Gu Y.
      • He J.
      • DeCarli C.
      • et al.
      Mediterranean diet and magnetic resonance imaging-assessed cerebrovascular disease.
      ] (using magnetic resonance imaging (MRI)); and Durga AV & Manorenj S (2019) [
      • Durga A.V.
      • Manorenj S.
      Dietary pattern in adult patients with acute stroke in South India: a case-control study from a tertiary care center in Hyderabad.
      ] (using MRI or computed tomography scans). Maher MA & Gutbi SS (2017) [
      • Maher M.A.
      • Gutbi S.S.
      Assessment of dietary pattern among coronary heart disease outpatients attended El-Shaap teaching hospital, Khartoum state.
      ] did not report on the method used to identify new cases.
      A summary of the risk of bias assessment is presented in Supplemental Table 5. Four studies were judged to be at serious risk of bias due to either limited control of confounding (i.e. failure to adjust for important confounders, including age, sex, body mass index (BMI) and energy intake; in the absence of BMI, controlling for energy intake and physical activity levels was sufficient; in the absence of energy intake, adjusting for overall food intake was acceptable) [
      • Durga A.V.
      • Manorenj S.
      Dietary pattern in adult patients with acute stroke in South India: a case-control study from a tertiary care center in Hyderabad.
      ,
      • Schröder H.
      • Salas-Salvadó J.
      • Martínez-González M.A.
      • Fíto M.
      • Corella D.
      • Estruch R.
      • et al.
      Baseline adherence to the Mediterranean diet and major cardiovascular events: prevención con Dieta Mediterránea trial.
      ] or concerns regarding exposure classification [
      • Durga A.V.
      • Manorenj S.
      Dietary pattern in adult patients with acute stroke in South India: a case-control study from a tertiary care center in Hyderabad.
      ,
      • Fraser G.E.
      • Sabaté J.
      • Beeson W.L.
      • Strahan T.M.
      A possible protective effect of nut consumption on risk of coronary heart disease. The Adventist Health Study.
      ,
      • Turati F.
      • Pelucchi C.
      • Galeone C.
      • Praud D.
      • Tavani A.
      • La Vecchia C.
      Mediterranean diet and non-fatal acute myocardial infarction: a case-control study from Italy.
      ], while one study was judged to be at critical risk of bias [
      • Maher M.A.
      • Gutbi S.S.
      Assessment of dietary pattern among coronary heart disease outpatients attended El-Shaap teaching hospital, Khartoum state.
      ] due to possible bias in exposure classification (not using a validated dietary assessment method and not quantifying intake levels).
      The meta-analysis comparing the highest versus the lowest category of intake (31 comparisons) showed a protective association between the consumption of legumes and CVD risk (RR = 0.94; 95%CI:0.89, 0.99), with a moderate level of heterogeneity across included estimates (I2 = 43.1%) (Fig. 2). The Egger's regression asymmetry test showed no publication bias (Supplemental Figs. 1–B), supported by the substantially symmetric distribution of the corresponding funnel plot (Supplemental Figs. 1–A). The pooled estimate was largely determined by prospective cohort studies (20 out of the 25 studies considered) and as expected, the RR remained materially unchanged when the analysis relied only on this study design (n = 20, RR = 0.96, 95%CI: 0.91; 1.00; I2 = 39.5) (Fig. 2 and Supplemental Table 8). Conversely, the association was stronger when only case–control studies were considered (n = 5, RR = 0.72, 95% CI: 0.60, 0.85; I2 = 0.0%). An inverse association was also observed among studies conducted in Asia (n = 6, RR = 0.82, 95%CI: 0.74, 0.90; I2 = 9.9%) and the pooled estimate for non-Asian studies was similar to that observed for the main analysis (n = 19, RR = 0.97, 95%CI: 0.92, 1.02; I2 = 34.0%) (Supplemental Table 8). The association was stronger considering studies that did not rely on a validated dietary assessment method [
      • World Cancer Research Fund/American Institute for Cancer Research
      Continuous update Project expert report 2018.
      ,
      • Mizrahi A.
      • Knekt P.
      • Montonen J.
      • Laasksonen M.A.
      • Heliövaaea M.
      • Järvinen R.
      Plant foods and the risk of cerebrovascular diseases: a potential protection of fruit consumption.
      ,
      • Bazzano L.A.
      • He J.
      • Ogden L.G.
      • Loria C.
      • Vupputuri S.
      • Myers L.
      • et al.
      Legume consumption and risk of coronary heart disease in US men and women: NHANES I Epidemiologic Follow-up Study.
      ,
      • Durga A.V.
      • Manorenj S.
      Dietary pattern in adult patients with acute stroke in South India: a case-control study from a tertiary care center in Hyderabad.
      ,
      • Fraser G.E.
      • Sabaté J.
      • Beeson W.L.
      • Strahan T.M.
      A possible protective effect of nut consumption on risk of coronary heart disease. The Adventist Health Study.
      ,
      • Maher M.A.
      • Gutbi S.S.
      Assessment of dietary pattern among coronary heart disease outpatients attended El-Shaap teaching hospital, Khartoum state.
      ,
      • Turati F.
      • Pelucchi C.
      • Galeone C.
      • Praud D.
      • Tavani A.
      • La Vecchia C.
      Mediterranean diet and non-fatal acute myocardial infarction: a case-control study from Italy.
      ] (n = 6, RR = 0.85, 95%CI: 0.74; 0.96; I2 = 30.0%), when compared with those using validated methods [
      • Atkins J.L.
      • Whincup P.H.
      • Morris L.T.
      • Lennon L.T.
      • Papacosta O.
      • Wannamethee S.G.
      High diet quality is associated with a lower risk of cardiovascular disease and all-cause mortality in older men.
      ,
      • Baik I.
      • Cho N.H.
      • Kim S.H.
      • Shin C.
      Dietary information improves cardiovascular disease risk prediction models.
      ,
      • Bazzano L.A.
      • He J.
      • Ogden L.G.
      • Loria C.
      • Vupputuri S.
      • Myers L.
      • et al.
      Legume consumption and risk of coronary heart disease in US men and women: NHANES I Epidemiologic Follow-up Study.
      ,
      • Bernstein A.M.
      • Pan A.
      • Rexrode K.M.
      • Stampfer M.
      • Hu F.B.
      • Mozaffarian D.
      • et al.
      Dietary protein sources and the risk of stroke in men and women.
      ,
      • Bernstein A.M.
      • Sun Q.
      • Hu F.B.
      • Stampfer M.J.
      • Manson J.E.
      • Willett W.C.
      Major dietary protein sources and risk of coronary heart disease in women.
      ,
      • Fung T.T.
      • Isanaka S.
      • Hu F.B.
      • Willett W.C.
      International food group-based diet quality and risk of coronary heart disease in men and women.
      ,
      • Haring B.
      • Gronroos N.
      • Nettleton J.A.
      • von Ballmoos M.C.
      • Selvin E.
      • Alonso A.
      Dietary protein intake and coronary heart disease in a large community based cohort: results from the Atherosclerosis Risk in Communities (ARIC) study [corrected].
      ,
      • Haring B.
      • Misialek J.R.
      • Rebholz C.M.
      • Petruski-Ivleva N.
      • Gottesman R.F.
      • Mosley T.H.
      • et al.
      Association of dietary protein consumption with incident silent cerebral infarcts and stroke: the atherosclerosis risk in Communities (ARIC) study.
      ,
      • Kabagambe E.K.
      • Baylin A.
      • Ruiz-Narvarez E.
      • Siles X.
      • Campos H.
      Decreased consumption of dried mature beans is positively associated with urbanization and nonfatal acute myocardial infarction.
      ,
      • Kokubo Y.
      • Iso H.
      • Ishihara J.
      • Okada K.
      • Inoue M.
      • Tsugane S.
      Association of dietary intake of soy, beans, and isoflavones with risk of cerebral and myocardial infarctions in Japanese populations: the Japan Public Health Center-based (JPHC) study cohort I.
      ,
      • Martínez-González M.A.
      • Fernández-Jarne E.
      • Martínez-Losa E.
      • Prado-Santamaría M.
      • Brugarolas-Brufau C.
      • Serrano-Martinez M.
      Role of fibre and fruit in the Mediterranean diet to protect against myocardial infarction: a case-control study in Spain.
      ,
      • Mata-Fernández A.
      • Hershey M.S.
      • Pastrana-Delgado J.C.
      • Sotos-Prieto M.
      • Ruiz-Canela M.
      • Kales S.N.
      • et al.
      A Mediterranean lifestyle reduces the risk of cardiovascular disease in the “Seguimiento Universidad de Navarra” (SUN) cohort.
      ,
      • Miller V.
      • Mente A.
      • Dehghan M.
      • Rangarajan S.
      • Zhang X.
      • Swaminathan S.
      • et al.
      Fruit, vegetable, and legume intake, and cardiovascular disease and deaths in 18 countries (PURE): a prospective cohort study.
      ,
      • Nouri F.
      • Haghighadoost F.
      • Mohammadifard N.
      • Mansourian M.
      • Sadeghi M.
      • Roohafza H.
      • et al.
      The longitudinal association between soybean and non-soybean legumes intakes and risk of cardiovascular disease: Isfahan cohort study.
      ,
      • Perez-Cornago A.
      • Crowe F.L.
      • Appleby P.N.
      • Bradbury K.E.
      • Wood A.M.
      • Jakobsen M.U.
      • et al.
      Plant foods, dietary fibre and risk of ischaemic heart disease in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort.
      ,
      • Scarmeas N.
      • Luchsinger J.A.
      • Stern Y.
      • Gu Y.
      • He J.
      • DeCarli C.
      • et al.
      Mediterranean diet and magnetic resonance imaging-assessed cerebrovascular disease.
      ,
      • Schröder H.
      • Salas-Salvadó J.
      • Martínez-González M.A.
      • Fíto M.
      • Corella D.
      • Estruch R.
      • et al.
      Baseline adherence to the Mediterranean diet and major cardiovascular events: prevención con Dieta Mediterránea trial.
      ,
      • Tektonidis T.G.
      • Åkesson A.
      • Gigante B.
      • Wolk A.
      • Larsson S.C.
      A Mediterranean diet and risk of myocardial infarction, heart failure and stroke: a population-based cohort study.
      ,
      • Tong T.Y.N.
      • Appleby P.N.
      • Key T.J.
      • Dahm C.C.
      • Overvad K.
      • Olsen A.
      • et al.
      The associations of major foods and fibre with risks of ischaemic and haemorrhagic stroke: a prospective study of 418 329 participants in the EPIC cohort across nine European countries.
      ,
      • Yu D.
      • Zhang X.
      • Gao Y.T.
      • Li H.
      • Yang G.
      • Huang J.
      • et al.
      Fruit and vegetable intake and risk of CHD: results from prospective cohort studies of Chinese adults in Shanghai.
      ] (n = 19, RR = 0.96, 95%CI: 0.91, 1.01; I2 = 42.2%) (Supplemental Table 8). Subgroup analyses based on follow-up duration, number of included cases and participants' sex did not reveal significant changes (Supplemental Table 8). The inverse association between the intake of legumes and CVD risk also remained after the exclusion of studies specifically assessing soy intake [
      • Bernstein A.M.
      • Pan A.
      • Rexrode K.M.
      • Stampfer M.
      • Hu F.B.
      • Mozaffarian D.
      • et al.
      Dietary protein sources and the risk of stroke in men and women.
      ,
      • Fung T.T.
      • Isanaka S.
      • Hu F.B.
      • Willett W.C.
      International food group-based diet quality and risk of coronary heart disease in men and women.
      ,
      • Yu D.
      • Zhang X.
      • Gao Y.T.
      • Li H.
      • Yang G.
      • Huang J.
      • et al.
      Fruit and vegetable intake and risk of CHD: results from prospective cohort studies of Chinese adults in Shanghai.
      ] (RR = 0.93; 95%CI: 0.88; 0.98; I2 = 47.7%) and judged to be at serious/critical risk of bias [
      • Durga A.V.
      • Manorenj S.
      Dietary pattern in adult patients with acute stroke in South India: a case-control study from a tertiary care center in Hyderabad.
      ,
      • Fraser G.E.
      • Sabaté J.
      • Beeson W.L.
      • Strahan T.M.
      A possible protective effect of nut consumption on risk of coronary heart disease. The Adventist Health Study.
      ,
      • Maher M.A.
      • Gutbi S.S.
      Assessment of dietary pattern among coronary heart disease outpatients attended El-Shaap teaching hospital, Khartoum state.
      ,
      • Schröder H.
      • Salas-Salvadó J.
      • Martínez-González M.A.
      • Fíto M.
      • Corella D.
      • Estruch R.
      • et al.
      Baseline adherence to the Mediterranean diet and major cardiovascular events: prevención con Dieta Mediterránea trial.
      ,
      • Turati F.
      • Pelucchi C.
      • Galeone C.
      • Praud D.
      • Tavani A.
      • La Vecchia C.
      Mediterranean diet and non-fatal acute myocardial infarction: a case-control study from Italy.
      ] (RR = 0.94; 95%CI: 0.90; 0.99; I2 = 37.5%). Furthermore, after selectively removing each individual study (Supplemental Fig. 2), the pooled estimates have consistently remained within the 95% CI of the main analysis. In addition, pooled estimates remained the same after the application of ±15% or ±25% (instead of ±20%) to the closest available boundary (lower or upper) in open categories of intake, indicating that results are robust to the choice of a standard value for these estimations.
      Figure 2
      Figure 2Meta-analysis for the association between the intake of legumes (highest vs. lowest, as specified in the individual studies) and CVD overall in prospective cohort (PC) and case–control studies (CC). Weights and between-subgroup heterogeneity are from random-effects model.
      Of the twenty-five studies included in the analysis comparing extreme categories of intake, seventeen provided data that were considered in the dose–response analysis [
      • Mizrahi A.
      • Knekt P.
      • Montonen J.
      • Laasksonen M.A.
      • Heliövaaea M.
      • Järvinen R.
      Plant foods and the risk of cerebrovascular diseases: a potential protection of fruit consumption.
      ,
      • Baik I.
      • Cho N.H.
      • Kim S.H.
      • Shin C.
      Dietary information improves cardiovascular disease risk prediction models.
      ,
      • Bazzano L.A.
      • He J.
      • Ogden L.G.
      • Loria C.
      • Vupputuri S.
      • Myers L.
      • et al.
      Legume consumption and risk of coronary heart disease in US men and women: NHANES I Epidemiologic Follow-up Study.
      ,
      • Bernstein A.M.
      • Pan A.
      • Rexrode K.M.
      • Stampfer M.
      • Hu F.B.
      • Mozaffarian D.
      • et al.
      Dietary protein sources and the risk of stroke in men and women.
      ,
      • Bernstein A.M.
      • Sun Q.
      • Hu F.B.
      • Stampfer M.J.
      • Manson J.E.
      • Willett W.C.
      Major dietary protein sources and risk of coronary heart disease in women.
      ,
      • Haring B.
      • Gronroos N.
      • Nettleton J.A.
      • von Ballmoos M.C.
      • Selvin E.
      • Alonso A.
      Dietary protein intake and coronary heart disease in a large community based cohort: results from the Atherosclerosis Risk in Communities (ARIC) study [corrected].
      ,
      • Haring B.
      • Misialek J.R.
      • Rebholz C.M.
      • Petruski-Ivleva N.
      • Gottesman R.F.
      • Mosley T.H.
      • et al.
      Association of dietary protein consumption with incident silent cerebral infarcts and stroke: the atherosclerosis risk in Communities (ARIC) study.
      ,
      • Kabagambe E.K.
      • Baylin A.
      • Ruiz-Narvarez E.
      • Siles X.
      • Campos H.
      Decreased consumption of dried mature beans is positively associated with urbanization and nonfatal acute myocardial infarction.
      ,
      • Martínez-González M.A.
      • Fernández-Jarne E.
      • Martínez-Losa E.
      • Prado-Santamaría M.
      • Brugarolas-Brufau C.
      • Serrano-Martinez M.
      Role of fibre and fruit in the Mediterranean diet to protect against myocardial infarction: a case-control study in Spain.
      ,
      • Mata-Fernández A.
      • Hershey M.S.
      • Pastrana-Delgado J.C.
      • Sotos-Prieto M.
      • Ruiz-Canela M.
      • Kales S.N.
      • et al.
      A Mediterranean lifestyle reduces the risk of cardiovascular disease in the “Seguimiento Universidad de Navarra” (SUN) cohort.
      ,
      • Miller V.
      • Mente A.
      • Dehghan M.
      • Rangarajan S.
      • Zhang X.
      • Swaminathan S.
      • et al.
      Fruit, vegetable, and legume intake, and cardiovascular disease and deaths in 18 countries (PURE): a prospective cohort study.
      ,
      • Perez-Cornago A.
      • Crowe F.L.
      • Appleby P.N.
      • Bradbury K.E.
      • Wood A.M.
      • Jakobsen M.U.
      • et al.
      Plant foods, dietary fibre and risk of ischaemic heart disease in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort.
      ,
      • Schröder H.
      • Salas-Salvadó J.
      • Martínez-González M.A.
      • Fíto M.
      • Corella D.
      • Estruch R.
      • et al.
      Baseline adherence to the Mediterranean diet and major cardiovascular events: prevención con Dieta Mediterránea trial.
      ,
      • Tektonidis T.G.
      • Åkesson A.
      • Gigante B.
      • Wolk A.
      • Larsson S.C.
      A Mediterranean diet and risk of myocardial infarction, heart failure and stroke: a population-based cohort study.
      ,
      • Tong T.Y.N.
      • Appleby P.N.
      • Key T.J.
      • Dahm C.C.
      • Overvad K.
      • Olsen A.
      • et al.
      The associations of major foods and fibre with risks of ischaemic and haemorrhagic stroke: a prospective study of 418 329 participants in the EPIC cohort across nine European countries.
      ,
      • Turati F.
      • Pelucchi C.
      • Galeone C.
      • Praud D.
      • Tavani A.
      • La Vecchia C.
      Mediterranean diet and non-fatal acute myocardial infarction: a case-control study from Italy.
      ,
      • Yu D.
      • Zhang X.
      • Gao Y.T.
      • Li H.
      • Yang G.
      • Huang J.
      • et al.
      Fruit and vegetable intake and risk of CHD: results from prospective cohort studies of Chinese adults in Shanghai.
      ], comprising a total of 1 489 243 participants and 37 166 CVD cases. Considering an intake level of 0 g/week as reference, one portion of cooked legumes (100 g) per week was associated with a reduced CVD risk (RR = 0.97; 95%CI: 0.90, 1.04), which was further attenuated per each increasing portion (Fig. 3). The subgroup analysis based on study design is presented in supplemental figures 3-A and 3-B. Based on evidence from 14 prospective cohort studies, the inverse association between legume intake and CVD risk remains but is attenuated (Supplemental Figs. 3–A). Conversely, the dose–response analysis performed using data from case–control studies only has indicated a U-shaped association between the intake of legumes and CVD (Supplemental Figs. 3–B). However, the number of included studies was small (n = 3) and the interpretation of this result should be made with caution. Dose–response associations between legume intake and CVD risk by study location are presented in Supplemental Fig. 4. The gradually increasing benefit for every additional weekly portion was observed in non-Asian populations (14 studies, Supplemental Figs. 4–B) and it was more pronounced in Asians (3 studies, Supplemental Figs. 4–A), who have also reported higher intake levels.
      Figure 3
      Figure 3Summary dose–response association (solid line) and respective confidence interval (dashed lines) between the intake of legumes (g/week) and the risk of CVD (17 studies). Estimates were obtained through weighted mixed-effects models with restricted cubic splines with 3 knots at fixed percentiles of legume intake. The value of 0 g/week served as reference.

      3.2 Legumes and coronary heart disease

      Sixteen eligible studies assessed the association between the intake of legumes and CHD [
      • Atkins J.L.
      • Whincup P.H.
      • Morris L.T.
      • Lennon L.T.
      • Papacosta O.
      • Wannamethee S.G.
      High diet quality is associated with a lower risk of cardiovascular disease and all-cause mortality in older men.
      ,
      • Bazzano L.A.
      • He J.
      • Ogden L.G.
      • Loria C.
      • Vupputuri S.
      • Myers L.
      • et al.
      Legume consumption and risk of coronary heart disease in US men and women: NHANES I Epidemiologic Follow-up Study.
      ,
      • Bernstein A.M.
      • Sun Q.
      • Hu F.B.
      • Stampfer M.J.
      • Manson J.E.
      • Willett W.C.
      Major dietary protein sources and risk of coronary heart disease in women.
      ,
      • Fraser G.E.
      • Sabaté J.
      • Beeson W.L.
      • Strahan T.M.
      A possible protective effect of nut consumption on risk of coronary heart disease. The Adventist Health Study.
      ,
      • Fung T.T.
      • Isanaka S.
      • Hu F.B.
      • Willett W.C.
      International food group-based diet quality and risk of coronary heart disease in men and women.
      ,
      • Haring B.
      • Gronroos N.
      • Nettleton J.A.
      • von Ballmoos M.C.
      • Selvin E.
      • Alonso A.
      Dietary protein intake and coronary heart disease in a large community based cohort: results from the Atherosclerosis Risk in Communities (ARIC) study [corrected].
      ,
      • Kabagambe E.K.
      • Baylin A.
      • Ruiz-Narvarez E.
      • Siles X.
      • Campos H.
      Decreased consumption of dried mature beans is positively associated with urbanization and nonfatal acute myocardial infarction.
      ,
      • Kokubo Y.
      • Iso H.
      • Ishihara J.
      • Okada K.
      • Inoue M.
      • Tsugane S.
      Association of dietary intake of soy, beans, and isoflavones with risk of cerebral and myocardial infarctions in Japanese populations: the Japan Public Health Center-based (JPHC) study cohort I.
      ,
      • Maher M.A.
      • Gutbi S.S.
      Assessment of dietary pattern among coronary heart disease outpatients attended El-Shaap teaching hospital, Khartoum state.
      ,
      • Martínez-González M.A.
      • Fernández-Jarne E.
      • Martínez-Losa E.
      • Prado-Santamaría M.
      • Brugarolas-Brufau C.
      • Serrano-Martinez M.
      Role of fibre and fruit in the Mediterranean diet to protect against myocardial infarction: a case-control study in Spain.
      ,
      • Martínez-González M.A.
      • García-López M.
      • Bes-Rastollo M.
      • Toledo E.
      • Martínez-Lapiscina E.H.
      • Delgado-Rodriguez M.
      • et al.
      Mediterranean diet and the incidence of cardiovascular disease: a Spanish cohort.
      ,
      • Miller V.
      • Mente A.
      • Dehghan M.
      • Rangarajan S.
      • Zhang X.
      • Swaminathan S.
      • et al.
      Fruit, vegetable, and legume intake, and cardiovascular disease and deaths in 18 countries (PURE): a prospective cohort study.
      ,
      • Perez-Cornago A.
      • Crowe F.L.
      • Appleby P.N.
      • Bradbury K.E.
      • Wood A.M.
      • Jakobsen M.U.
      • et al.
      Plant foods, dietary fibre and risk of ischaemic heart disease in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort.
      ,
      • Tektonidis T.G.
      • Åkesson A.
      • Gigante B.
      • Wolk A.
      • Larsson S.C.
      A Mediterranean diet and risk of myocardial infarction, heart failure and stroke: a population-based cohort study.
      ,
      • Turati F.
      • Pelucchi C.
      • Galeone C.
      • Praud D.
      • Tavani A.
      • La Vecchia C.
      Mediterranean diet and non-fatal acute myocardial infarction: a case-control study from Italy.
      ,
      • Yu D.
      • Zhang X.
      • Gao Y.T.
      • Li H.
      • Yang G.
      • Huang J.
      • et al.
      Fruit and vegetable intake and risk of CHD: results from prospective cohort studies of Chinese adults in Shanghai.
      ], including a total of 1 114 165 participants and 28 258 cases (Supplemental Table 3). Twelve studies were prospective cohorts [
      • Atkins J.L.
      • Whincup P.H.
      • Morris L.T.
      • Lennon L.T.
      • Papacosta O.
      • Wannamethee S.G.
      High diet quality is associated with a lower risk of cardiovascular disease and all-cause mortality in older men.
      ,
      • Bazzano L.A.
      • He J.
      • Ogden L.G.
      • Loria C.
      • Vupputuri S.
      • Myers L.
      • et al.
      Legume consumption and risk of coronary heart disease in US men and women: NHANES I Epidemiologic Follow-up Study.
      ,
      • Bernstein A.M.
      • Sun Q.
      • Hu F.B.
      • Stampfer M.J.
      • Manson J.E.
      • Willett W.C.
      Major dietary protein sources and risk of coronary heart disease in women.
      ,
      • Fraser G.E.
      • Sabaté J.
      • Beeson W.L.
      • Strahan T.M.
      A possible protective effect of nut consumption on risk of coronary heart disease. The Adventist Health Study.
      ,
      • Fung T.T.
      • Isanaka S.
      • Hu F.B.
      • Willett W.C.
      International food group-based diet quality and risk of coronary heart disease in men and women.
      ,
      • Haring B.
      • Gronroos N.
      • Nettleton J.A.
      • von Ballmoos M.C.
      • Selvin E.
      • Alonso A.
      Dietary protein intake and coronary heart disease in a large community based cohort: results from the Atherosclerosis Risk in Communities (ARIC) study [corrected].
      ,
      • Kokubo Y.
      • Iso H.
      • Ishihara J.
      • Okada K.
      • Inoue M.
      • Tsugane S.
      Association of dietary intake of soy, beans, and isoflavones with risk of cerebral and myocardial infarctions in Japanese populations: the Japan Public Health Center-based (JPHC) study cohort I.
      ,
      • Martínez-González M.A.
      • García-López M.
      • Bes-Rastollo M.
      • Toledo E.
      • Martínez-Lapiscina E.H.
      • Delgado-Rodriguez M.
      • et al.
      Mediterranean diet and the incidence of cardiovascular disease: a Spanish cohort.
      ,
      • Miller V.
      • Mente A.
      • Dehghan M.
      • Rangarajan S.
      • Zhang X.
      • Swaminathan S.
      • et al.
      Fruit, vegetable, and legume intake, and cardiovascular disease and deaths in 18 countries (PURE): a prospective cohort study.
      ,
      • Perez-Cornago A.
      • Crowe F.L.
      • Appleby P.N.
      • Bradbury K.E.
      • Wood A.M.
      • Jakobsen M.U.
      • et al.
      Plant foods, dietary fibre and risk of ischaemic heart disease in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort.
      ,
      • Tektonidis T.G.
      • Åkesson A.
      • Gigante B.
      • Wolk A.
      • Larsson S.C.
      A Mediterranean diet and risk of myocardial infarction, heart failure and stroke: a population-based cohort study.
      ,
      • Turati F.
      • Pelucchi C.
      • Galeone C.
      • Praud D.
      • Tavani A.
      • La Vecchia C.
      Mediterranean diet and non-fatal acute myocardial infarction: a case-control study from Italy.
      ] (mean follow-up: 13.6 y) and four studies followed a case–control design [
      • Kabagambe E.K.
      • Baylin A.
      • Ruiz-Narvarez E.
      • Siles X.
      • Campos H.
      Decreased consumption of dried mature beans is positively associated with urbanization and nonfatal acute myocardial infarction.
      ,
      • Maher M.A.
      • Gutbi S.S.
      Assessment of dietary pattern among coronary heart disease outpatients attended El-Shaap teaching hospital, Khartoum state.
      ,
      • Martínez-González M.A.
      • Fernández-Jarne E.
      • Martínez-Losa E.
      • Prado-Santamaría M.
      • Brugarolas-Brufau C.
      • Serrano-Martinez M.
      Role of fibre and fruit in the Mediterranean diet to protect against myocardial infarction: a case-control study in Spain.
      ,
      • Turati F.
      • Pelucchi C.
      • Galeone C.
      • Praud D.
      • Tavani A.
      • La Vecchia C.
      Mediterranean diet and non-fatal acute myocardial infarction: a case-control study from Italy.
      ]. One study was performed in Africa [
      • Maher M.A.
      • Gutbi S.S.
      Assessment of dietary pattern among coronary heart disease outpatients attended El-Shaap teaching hospital, Khartoum state.
      ], three were performed in Asia [
      • Kokubo Y.
      • Iso H.
      • Ishihara J.
      • Okada K.
      • Inoue M.
      • Tsugane S.
      Association of dietary intake of soy, beans, and isoflavones with risk of cerebral and myocardial infarctions in Japanese populations: the Japan Public Health Center-based (JPHC) study cohort I.
      ,
      • Miller V.
      • Mente A.
      • Dehghan M.
      • Rangarajan S.
      • Zhang X.
      • Swaminathan S.
      • et al.
      Fruit, vegetable, and legume intake, and cardiovascular disease and deaths in 18 countries (PURE): a prospective cohort study.
      ,
      • Yu D.
      • Zhang X.
      • Gao Y.T.
      • Li H.
      • Yang G.
      • Huang J.
      • et al.
      Fruit and vegetable intake and risk of CHD: results from prospective cohort studies of Chinese adults in Shanghai.
      ], six in Europe [
      • Atkins J.L.
      • Whincup P.H.
      • Morris L.T.
      • Lennon L.T.
      • Papacosta O.
      • Wannamethee S.G.
      High diet quality is associated with a lower risk of cardiovascular disease and all-cause mortality in older men.
      ,
      • Martínez-González M.A.
      • Fernández-Jarne E.
      • Martínez-Losa E.
      • Prado-Santamaría M.
      • Brugarolas-Brufau C.
      • Serrano-Martinez M.
      Role of fibre and fruit in the Mediterranean diet to protect against myocardial infarction: a case-control study in Spain.
      ,
      • Martínez-González M.A.
      • García-López M.
      • Bes-Rastollo M.
      • Toledo E.
      • Martínez-Lapiscina E.H.
      • Delgado-Rodriguez M.
      • et al.
      Mediterranean diet and the incidence of cardiovascular disease: a Spanish cohort.
      ,
      • Perez-Cornago A.
      • Crowe F.L.
      • Appleby P.N.
      • Bradbury K.E.
      • Wood A.M.
      • Jakobsen M.U.
      • et al.
      Plant foods, dietary fibre and risk of ischaemic heart disease in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort.
      ,
      • Tektonidis T.G.
      • Åkesson A.
      • Gigante B.
      • Wolk A.
      • Larsson S.C.
      A Mediterranean diet and risk of myocardial infarction, heart failure and stroke: a population-based cohort study.
      ,
      • Turati F.
      • Pelucchi C.
      • Galeone C.
      • Praud D.
      • Tavani A.
      • La Vecchia C.
      Mediterranean diet and non-fatal acute myocardial infarction: a case-control study from Italy.
      ] and six in North America [
      • Bazzano L.A.
      • He J.
      • Ogden L.G.
      • Loria C.
      • Vupputuri S.
      • Myers L.
      • et al.
      Legume consumption and risk of coronary heart disease in US men and women: NHANES I Epidemiologic Follow-up Study.
      ,
      • Bernstein A.M.
      • Sun Q.
      • Hu F.B.
      • Stampfer M.J.
      • Manson J.E.
      • Willett W.C.
      Major dietary protein sources and risk of coronary heart disease in women.
      ,
      • Fraser G.E.
      • Sabaté J.
      • Beeson W.L.
      • Strahan T.M.
      A possible protective effect of nut consumption on risk of coronary heart disease. The Adventist Health Study.
      ,
      • Fung T.T.
      • Isanaka S.
      • Hu F.B.
      • Willett W.C.
      International food group-based diet quality and risk of coronary heart disease in men and women.
      ,
      • Haring B.
      • Gronroos N.
      • Nettleton J.A.
      • von Ballmoos M.C.
      • Selvin E.
      • Alonso A.
      Dietary protein intake and coronary heart disease in a large community based cohort: results from the Atherosclerosis Risk in Communities (ARIC) study [corrected].
      ,
      • Kabagambe E.K.
      • Baylin A.
      • Ruiz-Narvarez E.
      • Siles X.
      • Campos H.
      Decreased consumption of dried mature beans is positively associated with urbanization and nonfatal acute myocardial infarction.
      ]. Participants' age ranged from 19 to 83 years old. All studies used FFQs for the dietary assessment. Two studies have included soy and non-soy legumes in their exposure definition [
      • Fung T.T.
      • Isanaka S.
      • Hu F.B.
      • Willett W.C.
      International food group-based diet quality and risk of coronary heart disease in men and women.
      ,
      • Yu D.
      • Zhang X.
      • Gao Y.T.
      • Li H.
      • Yang G.
      • Huang J.
      • et al.
      Fruit and vegetable intake and risk of CHD: results from prospective cohort studies of Chinese adults in Shanghai.
      ], eight included only non-soy legumes [
      • Atkins J.L.
      • Whincup P.H.
      • Morris L.T.
      • Lennon L.T.
      • Papacosta O.
      • Wannamethee S.G.
      High diet quality is associated with a lower risk of cardiovascular disease and all-cause mortality in older men.
      ,
      • Bazzano L.A.
      • He J.
      • Ogden L.G.
      • Loria C.
      • Vupputuri S.
      • Myers L.
      • et al.
      Legume consumption and risk of coronary heart disease in US men and women: NHANES I Epidemiologic Follow-up Study.
      ,
      • Bernstein A.M.
      • Sun Q.
      • Hu F.B.
      • Stampfer M.J.
      • Manson J.E.
      • Willett W.C.
      Major dietary protein sources and risk of coronary heart disease in women.
      ,
      • Fraser G.E.
      • Sabaté J.
      • Beeson W.L.
      • Strahan T.M.
      A possible protective effect of nut consumption on risk of coronary heart disease. The Adventist Health Study.
      ,
      • Kabagambe E.K.
      • Baylin A.
      • Ruiz-Narvarez E.
      • Siles X.
      • Campos H.
      Decreased consumption of dried mature beans is positively associated with urbanization and nonfatal acute myocardial infarction.
      ,
      • Kokubo Y.
      • Iso H.
      • Ishihara J.
      • Okada K.
      • Inoue M.
      • Tsugane S.
      Association of dietary intake of soy, beans, and isoflavones with risk of cerebral and myocardial infarctions in Japanese populations: the Japan Public Health Center-based (JPHC) study cohort I.
      ,
      • Martínez-González M.A.
      • García-López M.
      • Bes-Rastollo M.
      • Toledo E.
      • Martínez-Lapiscina E.H.
      • Delgado-Rodriguez M.
      • et al.
      Mediterranean diet and the incidence of cardiovascular disease: a Spanish cohort.
      ,
      • Miller V.
      • Mente A.
      • Dehghan M.
      • Rangarajan S.
      • Zhang X.
      • Swaminathan S.
      • et al.
      Fruit, vegetable, and legume intake, and cardiovascular disease and deaths in 18 countries (PURE): a prospective cohort study.
      ] and six did not specify the type of legumes included [
      • Haring B.
      • Gronroos N.
      • Nettleton J.A.
      • von Ballmoos M.C.
      • Selvin E.
      • Alonso A.
      Dietary protein intake and coronary heart disease in a large community based cohort: results from the Atherosclerosis Risk in Communities (ARIC) study [corrected].
      ,
      • Maher M.A.
      • Gutbi S.S.
      Assessment of dietary pattern among coronary heart disease outpatients attended El-Shaap teaching hospital, Khartoum state.
      ,
      • Martínez-González M.A.
      • Fernández-Jarne E.
      • Martínez-Losa E.
      • Prado-Santamaría M.
      • Brugarolas-Brufau C.
      • Serrano-Martinez M.
      Role of fibre and fruit in the Mediterranean diet to protect against myocardial infarction: a case-control study in Spain.
      ,
      • Perez-Cornago A.
      • Crowe F.L.
      • Appleby P.N.
      • Bradbury K.E.
      • Wood A.M.
      • Jakobsen M.U.
      • et al.
      Plant foods, dietary fibre and risk of ischaemic heart disease in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort.
      ,
      • Tektonidis T.G.
      • Åkesson A.
      • Gigante B.
      • Wolk A.
      • Larsson S.C.
      A Mediterranean diet and risk of myocardial infarction, heart failure and stroke: a population-based cohort study.
      ,
      • Turati F.
      • Pelucchi C.
      • Galeone C.
      • Praud D.
      • Tavani A.
      • La Vecchia C.
      Mediterranean diet and non-fatal acute myocardial infarction: a case-control study from Italy.
      ]. CHD cases were assessed using medical records and vital statistics; Maher MA & Gutbi SS (2017) [
      • Maher M.A.
      • Gutbi S.S.
      Assessment of dietary pattern among coronary heart disease outpatients attended El-Shaap teaching hospital, Khartoum state.
      ] did not report the method used.
      The risk of bias assessment is presented in Supplemental Table 6. Two studies were judged to be at serious risk of bias due to concerns in exposure classification [
      • Fraser G.E.
      • Sabaté J.
      • Beeson W.L.
      • Strahan T.M.
      A possible protective effect of nut consumption on risk of coronary heart disease. The Adventist Health Study.
      ,
      • Turati F.
      • Pelucchi C.
      • Galeone C.
      • Praud D.
      • Tavani A.
      • La Vecchia C.
      Mediterranean diet and non-fatal acute myocardial infarction: a case-control study from Italy.
      ], while one [
      • Maher M.A.
      • Gutbi S.S.
      Assessment of dietary pattern among coronary heart disease outpatients attended El-Shaap teaching hospital, Khartoum state.
      ] was judged to be at critical risk of bias (not using a validated dietary assessment method and not quantifying intake levels).
      The meta-analysis comparing extreme categories of intake (19 comparisons) showed an inverse association between the intake of legumes and CHD, with a 10% decrease in disease risk (RR = 0.90; 95%CI: 0.85, 0.96) and a low level of heterogeneity (I2 = 27.9%) (Fig. 4). Visual inspection of the funnel plot (Supplemental Figs. 5–A) suggested an underrepresentation of small studies with positive associations, whereas the Egger's regression asymmetry test showed no publication bias (Supplemental Figs. 5–B). The analysis considering only prospective cohort studies (n = 12) resulted in a pooled RR similar to that observed in the main analysis (Fig. 4 and Supplemental Table 8), whereas a stronger association was observed among case–control studies (n = 4, RR = 0.72, 95%CI: 0.60, 0.87; I2 = 0.0%). A stronger association was also observed in studies conducted in Asia (n = 3, RR = 0.84, 95%CI: 0.71, 1.00; I2 = 0.0%) comparing with non-Asians (n = 13, RR = 0.91, 95%CI: 0.84, 0.97; I2 = 42.2%), and among studies that did not use validated dietary assessment methods [
      • Bazzano L.A.
      • He J.
      • Ogden L.G.
      • Loria C.
      • Vupputuri S.
      • Myers L.
      • et al.
      Legume consumption and risk of coronary heart disease in US men and women: NHANES I Epidemiologic Follow-up Study.
      ,
      • Fraser G.E.
      • Sabaté J.
      • Beeson W.L.
      • Strahan T.M.
      A possible protective effect of nut consumption on risk of coronary heart disease. The Adventist Health Study.
      ,
      • Maher M.A.
      • Gutbi S.S.
      Assessment of dietary pattern among coronary heart disease outpatients attended El-Shaap teaching hospital, Khartoum state.
      ,
      • Turati F.
      • Pelucchi C.
      • Galeone C.
      • Praud D.
      • Tavani A.
      • La Vecchia C.
      Mediterranean diet and non-fatal acute myocardial infarction: a case-control study from Italy.
      ] (n = 4, RR = 0.79, 95%CI: 0.67; 0.94; I2 = 31.3%), in comparison with those using validated methods [
      • Atkins J.L.
      • Whincup P.H.
      • Morris L.T.
      • Lennon L.T.
      • Papacosta O.
      • Wannamethee S.G.
      High diet quality is associated with a lower risk of cardiovascular disease and all-cause mortality in older men.
      ,
      • Bernstein A.M.
      • Sun Q.
      • Hu F.B.
      • Stampfer M.J.
      • Manson J.E.
      • Willett W.C.
      Major dietary protein sources and risk of coronary heart disease in women.
      ,
      • Fung T.T.
      • Isanaka S.
      • Hu F.B.
      • Willett W.C.
      International food group-based diet quality and risk of coronary heart disease in men and women.
      ,
      • Haring B.
      • Gronroos N.
      • Nettleton J.A.
      • von Ballmoos M.C.
      • Selvin E.
      • Alonso A.
      Dietary protein intake and coronary heart disease in a large community based cohort: results from the Atherosclerosis Risk in Communities (ARIC) study [corrected].
      ,
      • Kabagambe E.K.
      • Baylin A.
      • Ruiz-Narvarez E.
      • Siles X.
      • Campos H.
      Decreased consumption of dried mature beans is positively associated with urbanization and nonfatal acute myocardial infarction.
      ,
      • Kokubo Y.
      • Iso H.
      • Ishihara J.
      • Okada K.
      • Inoue M.
      • Tsugane S.
      Association of dietary intake of soy, beans, and isoflavones with risk of cerebral and myocardial infarctions in Japanese populations: the Japan Public Health Center-based (JPHC) study cohort I.
      ,
      • Martínez-González M.A.
      • Fernández-Jarne E.
      • Martínez-Losa E.
      • Prado-Santamaría M.
      • Brugarolas-Brufau C.
      • Serrano-Martinez M.
      Role of fibre and fruit in the Mediterranean diet to protect against myocardial infarction: a case-control study in Spain.
      ,
      • Martínez-González M.A.
      • García-López M.
      • Bes-Rastollo M.
      • Toledo E.
      • Martínez-Lapiscina E.H.
      • Delgado-Rodriguez M.
      • et al.
      Mediterranean diet and the incidence of cardiovascular disease: a Spanish cohort.
      ,
      • Miller V.
      • Mente A.
      • Dehghan M.
      • Rangarajan S.
      • Zhang X.
      • Swaminathan S.
      • et al.
      Fruit, vegetable, and legume intake, and cardiovascular disease and deaths in 18 countries (PURE): a prospective cohort study.
      ,
      • Perez-Cornago A.
      • Crowe F.L.
      • Appleby P.N.
      • Bradbury K.E.
      • Wood A.M.
      • Jakobsen M.U.
      • et al.
      Plant foods, dietary fibre and risk of ischaemic heart disease in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort.
      ,
      • Tektonidis T.G.
      • Åkesson A.
      • Gigante B.
      • Wolk A.
      • Larsson S.C.
      A Mediterranean diet and risk of myocardial infarction, heart failure and stroke: a population-based cohort study.
      ,
      • Yu D.
      • Zhang X.
      • Gao Y.T.
      • Li H.
      • Yang G.
      • Huang J.
      • et al.
      Fruit and vegetable intake and risk of CHD: results from prospective cohort studies of Chinese adults in Shanghai.
      ] (n = 19, RR = 0.94, 95%CI: 0.89; 0.99; I2 = 0.0%) (Supplemental Table 8). Subgroup analyses based on follow-up duration, number of included cases and sex of the participants did not reveal substantial changes (Supplemental Table 8). The sensitivity analyses performed excluding studies comprising soy legumes (RR = 0.89; 95%CI: 0.89; 0.96; I2 = 39.3%) and studies judged to be at a serious/critical risk of bias (RR = 0.91; 95%CI: 0.86; 0.97; I2 = 18.5%) have not altered substantially the magnitude and the precision of the association. The selective exclusion of each individual study did not influence the pooled estimate observed in the main analysis (Supplemental Fig. 6). The pooled RRs also remained the same after the application of ±15% or ±25% (instead of ±20%) to the closest available boundary (lower or upper) in open categories of legume intake, indicating that results are robust to the choice of a standard value for these estimations.
      Figure 4
      Figure 4Meta-analysis for the association between the intake of legumes (highest vs. lowest, as specified in the individual studies) and CHD in prospective cohort (PC) and case–control studies (CC). Weights and between-subgroup heterogeneity are from random-effects model.
      Of the total of eligible studies, eleven were included in the dose–response analysis [
      • Bazzano L.A.
      • He J.
      • Ogden L.G.
      • Loria C.
      • Vupputuri S.
      • Myers L.
      • et al.
      Legume consumption and risk of coronary heart disease in US men and women: NHANES I Epidemiologic Follow-up Study.
      ,
      • Bernstein A.M.
      • Sun Q.
      • Hu F.B.
      • Stampfer M.J.
      • Manson J.E.
      • Willett W.C.
      Major dietary protein sources and risk of coronary heart disease in women.
      ,
      • Haring B.
      • Gronroos N.
      • Nettleton J.A.
      • von Ballmoos M.C.
      • Selvin E.
      • Alonso A.
      Dietary protein intake and coronary heart disease in a large community based cohort: results from the Atherosclerosis Risk in Communities (ARIC) study [corrected].
      ,
      • Kabagambe E.K.
      • Baylin A.
      • Ruiz-Narvarez E.
      • Siles X.
      • Campos H.
      Decreased consumption of dried mature beans is positively associated with urbanization and nonfatal acute myocardial infarction.
      ,
      • Kokubo Y.
      • Iso H.
      • Ishihara J.
      • Okada K.
      • Inoue M.
      • Tsugane S.
      Association of dietary intake of soy, beans, and isoflavones with risk of cerebral and myocardial infarctions in Japanese populations: the Japan Public Health Center-based (JPHC) study cohort I.
      ,
      • Martínez-González M.A.
      • Fernández-Jarne E.
      • Martínez-Losa E.
      • Prado-Santamaría M.
      • Brugarolas-Brufau C.
      • Serrano-Martinez M.
      Role of fibre and fruit in the Mediterranean diet to protect against myocardial infarction: a case-control study in Spain.
      ,
      • Martínez-González M.A.
      • García-López M.
      • Bes-Rastollo M.
      • Toledo E.
      • Martínez-Lapiscina E.H.
      • Delgado-Rodriguez M.
      • et al.
      Mediterranean diet and the incidence of cardiovascular disease: a Spanish cohort.
      ,
      • Miller V.
      • Mente A.
      • Dehghan M.
      • Rangarajan S.
      • Zhang X.
      • Swaminathan S.
      • et al.
      Fruit, vegetable, and legume intake, and cardiovascular disease and deaths in 18 countries (PURE): a prospective cohort study.
      ,
      • Perez-Cornago A.
      • Crowe F.L.
      • Appleby P.N.
      • Bradbury K.E.
      • Wood A.M.
      • Jakobsen M.U.
      • et al.
      Plant foods, dietary fibre and risk of ischaemic heart disease in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort.
      ,
      • Tektonidis T.G.
      • Åkesson A.
      • Gigante B.
      • Wolk A.
      • Larsson S.C.
      A Mediterranean diet and risk of myocardial infarction, heart failure and stroke: a population-based cohort study.
      ,
      • Turati F.
      • Pelucchi C.
      • Galeone C.
      • Praud D.
      • Tavani A.
      • La Vecchia C.
      Mediterranean diet and non-fatal acute myocardial infarction: a case-control study from Italy.
      ,
      • Yu D.
      • Zhang X.
      • Gao Y.T.
      • Li H.
      • Yang G.
      • Huang J.
      • et al.
      Fruit and vegetable intake and risk of CHD: results from prospective cohort studies of Chinese adults in Shanghai.
      ] (927 701 participants and 23 294 cases). Considering no consumption (0 g/week) as the reference category, an intake of 100 g (about one portion) of legumes per week was associated with a reduction of 9% in CHD risk (RR = 0.91; 95%CI: 0.85, 0.98), with an increased benefit for each additional 100 g/week increment up to an intake of 400 g/week, upon which the association seems to reach a plateau (Fig. 5). Again, as expected, the pattern of association based on prospective cohort studies only was similar to that observed for the main dose–response analysis (Supplemental Figs. 7–B). When considering case–control studies only, we observed a U-shaped association between the intake of legumes and CHD, with the highest benefit level being noted for an intake of 200 g/week, when compared with a consumption of 0 g/week (RR = 0.53, 95%CI: 0.30, 0.95) (Supplemental Figs. 7–A). Considering two Asian studies and no consumption as the reference category, an inverse association was observed for an intake of 100 g/week (RR = 0.93, 95%CI: 0.86, 1.01), which increased further with an additional 100 g (200 g/week: RR = 0.88, 95%CI: 0.76, 1.02) and subsequently reached a plateau as an additional increase in intake did not relate to a further reduction in disease risk (300 g/week: RR = 0.87, 95%CI: 0.75,1.01) (Supplemental Figs. 8–A). For non-Asian studies (n = 9), the relationship pattern observed was, as expected, similar to that of the main dose–response analysis (Supplemental Figs. 8–B).
      Figure 5
      Figure 5Summary dose–response association (solid line) and respective confidence interval (dashed lines) between the intake of legumes (g/week) and the risk of CHD (11 studies). Estimates were obtained through weighted mixed-effects models with restricted cubic splines with 3 knots at fixed percentiles of legume intake. The value of 0 g/week served as reference.

      3.3 Legumes and stroke

      Nine studies on the association between the intake of legumes and stroke were identified [
      • Mizrahi A.
      • Knekt P.
      • Montonen J.
      • Laasksonen M.A.
      • Heliövaaea M.
      • Järvinen R.
      Plant foods and the risk of cerebrovascular diseases: a potential protection of fruit consumption.
      ,
      • Bernstein A.M.
      • Pan A.
      • Rexrode K.M.
      • Stampfer M.
      • Hu F.B.
      • Mozaffarian D.
      • et al.
      Dietary protein sources and the risk of stroke in men and women.
      ,
      • Durga A.V.
      • Manorenj S.
      Dietary pattern in adult patients with acute stroke in South India: a case-control study from a tertiary care center in Hyderabad.
      ,
      • Haring B.
      • Misialek J.R.
      • Rebholz C.M.
      • Petruski-Ivleva N.
      • Gottesman R.F.
      • Mosley T.H.
      • et al.
      Association of dietary protein consumption with incident silent cerebral infarcts and stroke: the atherosclerosis risk in Communities (ARIC) study.
      ,
      • Kokubo Y.
      • Iso H.
      • Ishihara J.
      • Okada K.
      • Inoue M.
      • Tsugane S.
      Association of dietary intake of soy, beans, and isoflavones with risk of cerebral and myocardial infarctions in Japanese populations: the Japan Public Health Center-based (JPHC) study cohort I.
      ,
      • Miller V.
      • Mente A.
      • Dehghan M.
      • Rangarajan S.
      • Zhang X.
      • Swaminathan S.
      • et al.
      Fruit, vegetable, and legume intake, and cardiovascular disease and deaths in 18 countries (PURE): a prospective cohort study.
      ,
      • Scarmeas N.
      • Luchsinger J.A.
      • Stern Y.
      • Gu Y.
      • He J.
      • DeCarli C.
      • et al.
      Mediterranean diet and magnetic resonance imaging-assessed cerebrovascular disease.
      ,
      • Tektonidis T.G.
      • Åkesson A.
      • Gigante B.
      • Wolk A.
      • Larsson S.C.
      A Mediterranean diet and risk of myocardial infarction, heart failure and stroke: a population-based cohort study.