CRD summary

The authors concluded that intake of chocolate, soy protein isolate, green tea and black tea were associated with changes in some cardiovascular risk factors. In light of methodological weaknesses in the review process and individual studies, the unclear suitability of the analyses and a lack of information on individual studies, the authors' conclusions should be treated with caution.

Authors' objectives

To evaluate the effects of flavonoids and flavonoid-rich foods on cardiovascular disease and associated risk factors.

Searching

The Cochrane Library, MEDLINE and EMBASE were searched to June 2007. Search terms were reported. Bibliographies of relevant reviews were checked and experts were contacted. The search was restricted to published articles. Articles not in English were translated where possible.

Study selection

Randomised controlled trials (RCT) on cardiovascular disease or cardiovascular disease risk factors that evaluated the effect of advising participants to take or eat more flavonoids or flavonoid-rich foods were eligible for inclusion. Studies of children, pregnant women or patients critically ill with illnesses other than cardiovascular disease were excluded. Studies of multiple interventions where the effect of flavonoids could not be isolated or studies in which the control condition did not allow the effects to be reasonably ascribed to flavonoids were excluded.

Included studies were of a wide range of flavonoids, the most common of which were soy or soy isoflavone containing products. Study duration ranged from a few hours to 52 weeks. Outcomes reported in the review were flow-mediated dilation, systolic and diastolic blood pressure and levels of high density lipoproteins (HDL) and low density lipoproteins (LDL). None of the included studies reported cardiovascular events.

Study selection was performed independently by two reviewers. Disagreements were resolved through discussion with a third reviewer.

Assessment of study quality

Methodological quality of included studies was assessed according to allocation concealment, blinding of participants, researchers and outcome assessors, reporting of dropouts, funding from industry and whether the saturated fat intake in the intervention and control arms were within 2% of one another. Twenty per cent of the validity assessment was performed independently by two reviewers and 80% per cent was performed by a single reviewer.

Data extraction

The mean and standard deviation of the changes in each outcome from baseline to the end of intervention or, where not available, for the outcome measurement, were extracted for intervention and control arms. Where necessary, standard deviations were calculated from standard errors or confidence intervals (CI). Where numerical values were not provided, these were estimated from figures. If the intervention and control arms differed at baseline by more than the pre and post measurement for either of the groups, those data were not used. In crossover studies, the mean and standard deviation were used separately for intervention and control groups. Twenty per cent of the data extraction was performed independently by two reviewers and 80% was performed by a single reviewer.

Methods of synthesis

Interventions were grouped according to food sources or major flavonoid groups and pooled weighted mean differences (WMD) with 95% CI were calculated using the DerSimonian and Laird random-effects model. Subgroup analyses were carried out where data from at least five studies were available to investigate type of control or placebo, type of intervention, dosage, duration, participants' sex, menopausal status and baseline risk of cardiovascular disease. Sensitivity analyses were conducted for each of the quality criteria by excluding trials that did not meet the criteria. Cochran's test and the I² statistic were used to assess heterogeneity. Funnel plots were used to assess for publication bias.

Results of the review

The meta analysis included 133 RCTs (n=6,980), 78 parallel trials (n unclear) and 92 crossover trials (n unclear). Methodological weaknesses evident in most studies were: unclear or no allocation concealment; lack of blinding of participants, researchers or outcome assessors; a difference of more than 2% in the saturated fat intake for intervention and control arms; and receipt of industry funding. Dropouts were clearly reported in most studies.

Chronic intake of chocolate or cocoa increased flow-mediated dilation (WMD 1.45%, 95% CI 0.62% to 2.28%; two studies, n= 78), reduced systolic blood pressure (WMD -5.88, 95% CI -9.55 to -2.21; five studies, n=194) and reduced diastolic blood pressure (WMD -3.30, 95% CI -5.77 to -0.83; four studies n=140).

Acute chocolate or cocoa intake increased flow-mediated dilation (WMD 3.99%, 95% CI 2.86% to 5.12%, I²=46%; six studies, n=140). Funnel plots of chocolate and cocoa studies suggested publication bias, the findings for flow-mediated dilation were not robust to sensitivity analyses and there was evidence of significant heterogeneity for the blood pressure analyses (I²=81% and I²=70%). P values were not reported for the meta-analyses.

Chronic black tea intake was associated with a significant increase in flow-mediated dilation (WMD 3.40%, 95% CI 1.85% to 4.95%; one study, n= 100). However, black tea was associated with an acute increase in systolic (WMD 5.69, 95% CI 1.52 to 9.86; four studies, n unclear) and diastolic blood pressure (WMD 2.56, 95% CI 1.03 to 4.10; four studies, n unclear). The effect on blood pressure remained significant when caffeine intake was controlled for.

Soy protein isolate intake was associated with a significant reduction in diastolic blood pressure (WMD -1.99, 95% CI -2.86 to -1.12; nine studies, n=562) and LDL (WMD -0.19, 95% CI -0.24 to -0.14; 39 studies, n= 2,747). There was no evidence of significant heterogeneity for these outcomes.

Green tea intake was associated with a significant reduction in LDL (WMD -0.23, 95% CI -0.34 to -0.12; four studies, n=378). P values were not reported for the meta-analyses.

There was no significant association between other flavonoid groups and flow-mediated dilation, blood pressure, HDL or LDL.

Authors' conclusions

Chronic and acute intake of chocolate increased flow-mediated dilation and reduced systolic and diastolic blood pressure. Soy protein isolate reduced diastolic blood pressure and both soy protein isolate and green tea reduced LDL. Acute intake of black tea increased both systolic and diastolic blood pressure. For most other flavonoid subclasses, there was insufficient evidence to determine any effects on cardiovascular risk factors.

CRD commentary

The review addressed a clear question, although the inclusion criteria were generally stated in terms of exclusion criteria. Three relevant databases were searched. The search was restricted to published articles and evidence of reporting bias was found for some results. Although appropriate methods were used in the study selection process to minimise reviewer error and bias, these were not applied to all studies for data extraction and validity assessment, therefore, the possibility of reviewer error and bias could not be ruled out. Insufficient information was provided on the individual studies, particularly the control conditions. Methodological weaknesses of the included studies may have affected the reliability of the data. Given the presence of statistical heterogeneity for some outcomes, clinical heterogeneity between studies and inclusion of both parallel and crossover trials in the analysis, the results may have been better treated in a narrative synthesis. The absence of p values for the meta-analyses made it difficult to assess the statistical significance of the findings.

In light of methodological weaknesses in the review process and included studies, the scarcity of information on individual studies and weaknesses in the analysis, the authors' conclusions should be treated with caution.

Implications of the review for practice and research

Practice: The authors stated that the possible risks of flavonoid consumption should be considered in addition to the clinical benefits.

Research: The authors stated that further research was required on commonly consumed subclasses other than soy or cocoa, such as anthocyanins and flavanones. Future research should investigate dose-response effects and cardiovascular disease risk factors, should be of sufficient duration and should report all outcomes.

Funding

Not stated.

Bibliographic details

Hooper L, Kroon PA, Rimm EB, Cohn JS, Harvey I, Le Cornu KA, Ryder JJ, Hall WL, Cassidy A. Flavonoids, flavonoid-rich foods, and cardiovascular risk: a meta-analysis of randomized controlled trials. American Journal of Clinical Nutrition 2008; 88(1): 38-50. [PubMed: 18614722]

Original Paper URL

http://www.ajcn.org/cgi/content/full/88/1/38

Indexing Status

Subject indexing assigned by NLM

MeSH

Blood Pressure /drug effects; Cacao /chemistry; Cardiovascular Diseases /blood /epidemiology; Diet; Endothelium, Vascular /drug effects; Flavonoids /pharmacology; Humans; Lipid Metabolism /drug effects; Randomized Controlled Trials as Topic; Risk Factors; Soybean Proteins /pharmacology; Tea /chemistry; Vitis /chemistry

AccessionNumber

12008105709

Database entry date

16/12/2009

Record Status

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.