Association of Total Nut, Tree Nut, Peanut, and Peanut Butter Consumption with Cancer Incidence and Mortality: A Comprehensive Systematic Review and Dose-Response Meta-Analysis of Observational Studies

Abstract

Data on the association of nut intake with risk of cancer and its mortality are conflicting. Although previous meta-analyses summarized available findings in this regard, some limitations may distort their findings. Moreover, none of these meta-analyses examined the dose-response associations of total nut intake with the risk of specific cancers as well as associations between specific types of nuts and cancer mortality. Therefore, this study aimed to summarize available findings on the associations of total nut (tree nuts and peanuts), tree nut (walnuts, pistachios, macadamia nuts, pecans, cashews, almonds, hazelnuts, and Brazil nuts), peanut (whole peanuts without considering peanut butter), and peanut butter consumption with risk of cancer and its mortality by considering the above-mentioned points. We searched the online databases until March 2020 to identify eligible articles. In total, 43 articles on cancer risk and 9 articles on cancer mortality were included in the current systematic review and meta-analysis. The summary effect size (ES) for risk of cancer, comparing the highest with lowest intakes of total nuts, was 0.86 (95% CI: 0.81, 0.92, P < 0.001, I2 = 58.1%; P < 0.01), indicating a significant inverse association. Such a significant inverse association was also seen for tree nut intake (pooled ES: 0.87, 95% CI: 0.78-0.96, P < 0.01, I2 = 15.8%; P = 0.28). Based on the dose-response analysis, a 5-g/d increase in total nut intake was associated with 3%, 6%, and 25% lower risks of overall, pancreatic, and colon cancers, respectively. In terms of cancer mortality, we found 13%, 18%, and 8% risk reductions with higher intakes of total nuts, tree nuts, and peanuts, respectively. In addition, a 5-g/d increase in total nut intake was associated with a 4% lower risk of cancer mortality. In conclusion, our findings support the protective association between total nut and tree nut intake and the risk of cancer and its mortality.

Keywords: cancer; meta-analysis; mortality; nuts; peanuts.

FIGURE 1

Flow diagram of the study selection.

FIGURE 2

Nonlinear dose-response associations between total nut (A, B, C), tree nut (D), peanut butter (E), and peanut (F) intake and risk of overall cancer in adults aged ≥18 y among all (A), cohort (B, D, E, F), and case-control studies (C). Dietary intake of nuts was modeled in a random-effects model using restricted cubic splines with knots fixed at the 5th, 35th, 65th, and 95th percentiles of the distribution. The dotted line indicates the linear model. The solid line represents point estimates of the association between nut intake and ESs. The dashed line presents the 95% CI. ES, effect size.

FIGURE 3

Nonlinear dose-response associations between total nut intake and risk of pancreatic (A), lung (B), colon (C), and colorectal (D) cancers in adults aged ≥18 y. Dietary intake of total nuts was modeled in a random-effects model using restricted cubic splines with knots fixed at the 5th, 35th, 65th, and 95th percentiles of the distribution. The dotted line indicates the linear model. The solid line represents point estimates of the association between total nut intake and ESs. The dashed line presents the 95% CI. ES, effect size.

FIGURE 4

Nonlinear dose-response associations between total nut intake and risk of rectal (A), esophageal (B), gastric (C), and breast (D) cancers in adults aged ≥18 y. Dietary intake of total nuts was modeled in a random-effects model using restricted cubic splines with knots fixed at the 5th, 35th, 65th, and 95th percentiles of the distribution. The dotted line indicates the linear model. The solid line represents point estimates of the association between total nut intake and ESs. The dashed line presents the 95% CI. ES, effect size.

FIGURE 5

Nonlinear dose-response association between total nut (A), peanut (B), and peanut butter (C) intake and risk of cancer mortality in adults aged ≥18 y. All included studies were prospective cohorts. Dietary intake of nuts was modeled in a random-effects model using restricted cubic splines with knots fixed at the 5th, 35th, 65th, and 95th percentiles of the distribution. The dotted line indicates the linear model. The solid line represents point estimates of the association between nut intake and ESs. The dashed line presents the 95% CI. ES, effect size.