Abstract MP21: Consumption of Nuts and Beans and Risk of Incident Coronary Heart Disease, Stroke, and Diabetes Mellitus: A Systematic Review and Meta-analysis
BACKGROUND Higher consumption of nuts is linked to lower risk of CHD, but relationships with stroke or diabetes are less established. Further, whether consuming beans (legumes) is associated with CHD, stroke, or diabetes has not been systematically reviewed.
OBJECTIVE To systematically investigate and quantify the relationships of nut and bean intake with incident CHD, stroke, and diabetes.
METHODS Using MOOSE and PRISMA guidelines, we systematically searched multiple databases to identify randomized control trials or observational studies examining associations of nut or bean intake with incident CHD, stroke, or diabetes. Studies were excluded if reporting only intermediate physiologic measures, soft CVD outcomes (e.g., angina), or crude risk estimates. Data were extracted independently and in duplicate. Generalized least-squares trend estimation was used to assess dose-response relationships. Pooled summary estimates were calculated by fixed (I2<30%) or random (I2≥30%) effects meta-analysis. Prespecified sources of heterogeneity were analyzed using meta-regression. Potential for publication bias was explored using funnel plots and Begg’s test.
RESULTS From 2,737 identified abstracts, 24 observational studies and 0 trials met inclusion criteria, comprising 813,590 individuals and 21,149 CHD, 8,531 stroke, and 24,044 diabetes cases. Nut intake was inversely associated with fatal CHD (n=4 studies, 2,794 cases; RR per 1oz/wk: 0.88; 95% CI 0.83, 0.93; I2 0%), non-fatal CHD (n=3, 4,280 cases; RR per 1oz/wk: 0.94; 95% CI 0.89, 0.98; I2 14%), and diabetes (n=4, 16,371 cases; RR per 1oz/wk: 0.94; 95% CI 0.89, 0.99; I2 61%); but not ischemic or hemorrhagic stroke (n=2, 4,553 cases). Bean intake was also inversely associated with total CHD (n= 4, 10,164 cases; RR 1oz/day: 0.94; 95% CI 0.89, 0.98; I2 6%) but not diabetes (n=2; 5561 cases) or stroke (n=2; 2356 cases). Meta-regression identified age (p 0.04) and study duration (p 0.01) as sources of heterogeneity, with less benefits in older individuals or with greater length of follow-up. Evidence for publication bias was not seen.
CONCLUSIONS Our findings support benefits of eating nuts and beans to reduce CHD, and nuts to reduce diabetes. More evidence from long-term prospective studies and trials is needed to assess whether eating nuts or beans influences stroke and whether eating beans influences diabetes.
- © 2013 by American Heart Association, Inc.