Association between Plant-Based Dietary Patterns and Risk of Cardiovascular Disease: A Systematic Review and Meta-Analysis of Prospective Cohort Studies

Abstract

Plant-based diets, characterized by a higher consumption of plant foods and a lower consumption of animal foods, are associated with a favorable cardiovascular disease (CVD) risk, but evidence regarding the association between plant-based diets and CVD (including coronary heart disease (CHD) and stroke) incidence remain inconclusive. A literature search was conducted using the PubMed, EMBASE and Web of Science databases through December 2020 to identify prospective observational studies that examined the associations between plant-based diets and CVD incidence among adults. A systematic review and a meta-analysis using random effects models and dose-response analyses were performed. Ten studies describing nine unique cohorts were identified with a total of 698,707 participants (including 137,968 CVD, 41,162 CHD and 13,370 stroke events). Compared with the lowest adherence, the highest adherence to plant-based diets was associated with a lower risk of CVD (RR 0.84; 95% CI 0.79-0.89) and CHD (RR 0.88; 95% CI 0.81-0.94), but not of stroke (RR 0.87; 95% CI 0.73-1.03). Higher overall plant-based diet index (PDI) and healthful PDI scores were associated with a reduced CVD risk. These results support the claim that diets lower in animal foods and unhealthy plant foods, and higher in healthy plant foods are beneficial for CVD prevention. Protocol was published in PROSPERO (No. CRD42021223188).

Keywords: cardiovascular disease; coronary heart disease; meta-analysis; plant-based diet.

Figure 1

Flow diagram of the literature search.

Figure 2

Forest plot of the adjusted relative risk (RR) of cardiovascular disease for the highest versus the lowest adherence to plant-based dietary patterns. Pooled risk estimates and 95% confidence intervals (CI) using a random effects model for meta-analysis are in bold. Abbreviations: ARIC, Atherosclerosis Risk in Communities Study; CI, confidence interval; EPIC, European Prospective Investigation into Cancer and Nutrition; NHS, Nurses’ Health Study, NHS2, Nurses’ Health Study II; REGARDS, Reasons for Geographic and Racial Differences in Stroke; RR, relative risk; seTE, standard error of treatment effect; TCHS, Tzuchi Health Study; TCVS, Tzuchi Vegetarian Study; TE, treatment effect; UK, United Kingdom.

Figure 3

Forest plot of the adjusted relative risk (RR) of coronary heart disease for the highest versus the lowest adherence to plant-based dietary patterns. Pooled risk estimates and 95% confidence intervals (CI) using a random effects model for meta-analysis are in bold. Abbreviations: CI, confidence interval; EPIC, European Prospective Investigation into Cancer and Nutrition; NHS, Nurses’ Health Study, NHS2, Nurses’ Health Study II; REGARDS, Reasons for Geographic and Racial Differences in Stroke; RR, relative risk; seTE, standard error of treatment effect; TE, treatment effect; UK, United Kingdom.

Figure 4

Forest plot of the adjusted relative risk (RR) of stroke for the highest versus the lowest adherence to plant-based dietary patterns. Pooled risk estimates and 95% confidence intervals (CI) using a random effects model for meta-analysis are in bold. Abbreviations: CI, confidence interval; EPIC, European Prospective Investigation into Cancer and Nutrition; NHS, Nurses’ Health Study, NHS2, Nurses’ Health Study II; REGARDS, Reasons for Geographic and Racial Differences in Stroke; RR, relative risk; seTE, standard error of treatment effect; TCHS, Tzuchi Health Study; TCVS, Tzuchi Vegetarian Study; TE, treatment effect; UK, United Kingdom.

Figure 5

Dose–response analyses for the potential linear and nonlinear associations between plant-based diet indices (PDI) and incident cardiovascular disease (CVD). The shaded areas represent the 95% confidence intervals (CI) for the fitted linear trend (gray solid lines). The dashed line areas represent the 95% confidence intervals (CI) for the fitted nonlinear trend (black solid lines). (a). Overall PDI was associated with a lower risk of CVD in a linear fashion (RR: 0.85 (95% CI 0.80 to 0.90) per 25% increase, p for nonlinearity <0.01; p for significance of the curve = 0.53; p for linear association < 0.01). (b). Healthful PDI was associated with a lower risk of CVD in a linear fashion (RR: 0.84 (95% CI 0.75 to 0.94) per 25% increase, p for nonlinearity = 0.01; p for significance of the curve = 0.97; p for linear association < 0.01). (c). Unhealthful PDI was associated with a higher risk of CVD in a linear fashion (RR: 1.13 (95% CI 1.02 to 1.26) per 25% increase, p for nonlinearity < 0.01; p for significance of the curve = 0.13; p for linear association < 0.01).