Total fructose intake and risk of hypertension: a systematic review and meta-analysis of prospective cohorts

Abstract

Objectives: Although most controlled feeding trials have failed to show an adverse effect of fructose on blood pressure, concerns continue to be raised regarding the role of fructose in hypertension. To quantify the association between fructose-containing sugar (high-fructose corn syrup, sucrose, and fructose) intake and incident hypertension, a systematic review and meta-analysis of prospective cohort studies was undertaken.

Methods: MEDLINE, EMBASE, CINAHL and the Cochrane Library (through February 5, 2014) were searched for relevant studies. Two independent reviewers reviewed and extracted relevant data. Risk estimates were aggregated comparing the lowest (reference) quintile with highest quintile of intake using inverse variance random effect models and expressed as risk ratios (RR) with 95% confidence intervals (CIs). Interstudy heterogeneity was assessed (Cochran Q statistic) and quantified (I(2) statistic). The Newcastle-Ottawa Scale assessed study quality. Clinicaltrials.gov NCT01608620.

Results: Eligibility criteria were met by 3 prospective cohorts (n = 37,375 men and 185,855 women) with 58,162 cases of hypertension observed over 2,502,357 person-years of follow-up. Median fructose intake was 5.7-6.0% total energy in the lowest quintile and 13.9-14.3% total energy in the highest quintile. Fructose intake was not associated with incident hypertension (RR = 1.02, 95% CI, 0.99-1.04), with no evidence of heterogeneity (I(2) = 0%, p = 0.59). Spline curve modeling showed a U-shaped relationship with a negative association at intakes ≤50th percentile (∼10% total energy) and a positive association at higher intakes.

Conclusions: Total fructose intake was not associated with an increased risk of hypertension in 3 large prospective cohorts of U.S. men and women.

Keywords: fructose; hypertension; meta-analysis; prospective cohort; systematic review.

Figures

Fig. 1.. Flow of the literature. Of the 3470 identified articles, 3444 were excluded based on title and abstract. The remaining 26 were reviewed in full. Three prospective cohorts were included in the meta-analysis.

Fig. 2.. Forest plot of prospective cohorts investigating the association between total fructose intake and incident hypertension. The pooled effect estimate is represented as a diamond. p Values are for generic inverse variance random effects models. Interstudy heterogeneity was assessed via Cochrane's Q (χ2) at a significance level of p < 0.10 and quantified by I 2, where I 2 > 50% was considered to be evidence of substantial heterogeneity. All data are presented as relative risks with 95% confidence intervals.(Color figure available online.)

Fig. 3.. Incident hypertension across pooled quintiles of increasing fructose intake. The x-axis represents quintiles; the y-axis represents relative risk. Each point represents a meta-analyzed relative risk of data from all cohorts across each respective quintile, compared to the reference quintile. The vertical line represents each relative risk's respective 95% confidence interval.(Color figure available online.)

Supplementary Fig. 1.. Funnel plot of risk ratios for incident hypertension comparing the lowest with the highest quintiles for fructose intake. The vertical line represents the pooled relative risk. The dashed lines represent the pseudo-95% confidence intervals of the relative risk. The circles represent risk estimates for each cohort, and the horizontal lines represent standard errors of the relative risk.