1. Objectives

To assess the effectiveness of both advice to increase fruit and vegetable intake and the provision of fruit and vegetables to increase their consumption for the primary prevention of cardiovascular disease (CVD)

2. How studies were identified

The following databases were searched up to September 2012:

• CENTRAL (The Cochrane Library 2012, Issue 9)
• MEDLINE
• EMBASE
• Conference Proceedings Citation Index – Science (Web of Science)
• metaRegister of controlled trials (mRCT)
• WHO International Clinical Trials Registry Platform (ICTRP)

Reference lists were also screened and citation searches were performed

3. Criteria for including studies in the review

3.1 Study type

Randomized controlled trials, including cluster-randomised trials and cross-over trials

3.2 Study participants

Adults 18 years of age or over from the general population or who at increased risk of CVD due to factors such as smoking, dyslipidaemia, or hypertension

(Studies where over 25% of the participants had CVD or type II diabetes at baseline were excluded as the focus of the review is on the primary prevention of CVD and interventions for diabetes are covered by a separate review group)

3.3 Interventions

In comparison to no intervention or minimal intervention (e.g., leaflets), i) dietary advice to increase the intake of whole fruit and vegetables, and ii) the provision of whole fruit and vegetables as a means to increase their consumption, were evaluated for the primary prevention of CVD

(Multi-factorial lifestyle and dietary interventions and interventions aimed specifically at weight loss were excluded)

(Minimum duration of follow-up from intervention initiation was three months)

3.4 Primary outcomes

Clinical Events

• Cardiovascular mortality
• All-cause mortality
• Non-fatal endpoints: myocardial infarction, coronary artery bypass graft, percutaneous transluminal coronary angioplasty, angina, angiographically-defined coronary heart disease, stroke, carotid endarterectomy, peripheral artery disease

Secondary outcomes included change in cardiovascular risk factors (blood pressure, blood lipids, incidence of type II diabetes), quality of life, adverse effects and costs

4. Main results

4.1 Included studies

Ten trials with 1730 participants were included in this review

• Four of the ten trials enrolled only women, and three were of healthy adults
• Seven studies enrolled people with various conditions/risk factors: metabolic syndrome (two studies), dyslipidaemia, obesity, colorectal adenomatous polyps, pre-metabolic syndrome, family history of breast cancer
• Four trials provided dietary advice only to increase fruit and vegetable intake
• Dietary advice varied between studies, with recommendations ranging from five to nine servings of fruit and vegetables per day, and dietary counselling also varied in terms of delivery and frequency
• Six trials provided fruit and vegetables, of which five studies provided a single fruit or vegetable only
• Intervention fruit and vegetables included roasted soybean flour (kinako), cooked pinto beans, grapefruit, raw garlic, tomato, and legumes
• Duration of follow-up ranged from three months to one year

4.2 Study settings

• Trials were conducted in Brazil (1 trial), Canada (1 trial), the United Kingdom of Great Britain and Northern Ireland (2 trials) and in the United States of America (6 trials)

4.3 Study settings

How the data were analysed
Two comparisons were made: i) advice to increase fruit and vegetable intake versus no advice or minimal advice, and ii) the provision of fruit and vegetables versus no provision of fruit and vegetables, for the primary prevention of CVD. For continuous outcomes, mean differences (MD) and 95% confidence intervals (CI) were calculated. Sensitivity analyses were performed excluding studies at high risk of bias. Subgroup analyses based on study duration and baseline CVD risk were planned.

Results
Clinical Events
None of the included trials of either dietary advice or fruit and vegetable provision contributed data on clinical events.

Cardiovascular Risk Factors
Blood pressure
Advice to increase the intake of fruit and vegetables statistically significantly reduced systolic blood pressure, with a MD of -3.0 mmHg when compared with no or minimal advice (95% CI [-4.92 to -1.09], p=0.002; 2 studies/891 participants). Diastolic blood pressure, however, was not significantly improved (MD -0.9 mmHg, 95% CI [-2.03 to 0.24], p=0.21; 2 studies/891 participants).

The provision of roasted soy flour alone resulted in a statistically significant reduction in diastolic blood pressure in women with metabolic syndrome (p<0.05); however, this difference was in comparison to the roasted soy flour and fish oil group. Provision of tomatoes was found to statistically significantly increase both systolic (MD 1.0 mmHg, 95% CI [0.45 to 1.55], p<0.0001; 1 study/157 participants) and diastolic (MD 1.5 mmHg, 95% CI [1.18 to 1.82], p<0.0001; 1 study/157 participants) blood pressure. No other studies reported a significant effect of fruit and vegetables on either systolic or diastolic blood pressure.

Lipid levels
Overall, advice to increase the intake of fruit and vegetables had no effect on total cholesterol (MD -0.01, 95%CI [-0.11 to 0.09], p=0.81; 4 studies/970 participants), LDL-cholesterol (MD -0.17, 95% CI [-0.38 to 0.03], p=0.10; 2 studies/251 participants), HDL-cholesterol (MD -0.01, 95%CI [-0.10 to 0.08], p=0.79; 2 trials/251 participants) or triglyceride levels (MD 0.10, 95% CI [-0.06 to 0.27], p=0.20; 3 trials/280 participants) compared with no or minimal dietary advice.

Pinto beans, when compared with chicken soup, reduced total cholesterol, LDL-cholesterol and HDL-cholesterol in one trial; however, the data were presented graphically (all p≤0.05; 80 participants). Total cholesterol was not statistically significantly reduced with the provision of fruit and vegetables (MD -0.10 mmol/L, 95% CI [-0.24 to 0.04], p=0.17; 2 studies/187 participants). Heterogeneity prevented pooling of data for LDL-cholesterol analyses; one study found tomatoes statistically significantly reduced LDL-cholesterol (MD -0.09 mmol/L, 95% CI [-0.12 to -0.06], 157 participants), while the other two studies showed no effect of fruit and vegetable provision. Heterogeneity also prevented the pooled analysis of data on HDL-cholesterol. Three studies reported no effect of fruit and vegetable provision on HDL-cholesterol, while two studies reported a significant increase, one with roasted soy flour (MD 0.17 mmol/L, 95% CI [0.02 to 0.32]), and the second with raw garlic (MD 0.08 mmol/L, 95% CI [0.00 to 0.16]). A further study showed a significant decrease in HDL-cholesterol with the provision of tomatoes (MD -0.06 mmol/L, 95% CI [-0.08 to -0.04]). Three studies provided useable data for the pooled analysis of the effect of fruit and vegetable provision on triglyceride levels. Overall, there was no effect of the treatment on triglyceride levels (MD -0.01 mmol/L, 95% CI [-0.03 to 0.01], p=0.32; 3 trials/284 participants). Three further studies reported no effect of the intervention on triglyceride levels.

Adverse effects
One study of dietary advice to increase fruit and vegetable consumption reported on adverse effects and noted that bowel movements and flatulence were significantly increased with the intervention, but this did not persist after three months. Two studies providing fruit and vegetables recorded adverse effects; while one trial reported no serious effects, in the other trial 57% of those in the raw garlic treatment group reported bad breath and body odour.

Additional outcomes
None of the included studies provided data on costs.

5. Additional author observations*

Few trials meeting the specified inclusion criteria were identified, and none of the identified studies reported on the primary clinical outcomes of interest. All studies were of relatively short duration, and thus the effectiveness of interventions to increase fruit and vegetable consumption on cardiovascular risk factors also remains uncertain. In addition, few trials contributed data to each analysis, and several planned pooled analyses were not performed due to heterogeneity. Across the trials of both dietary advice to increase the intake of fruit and vegetables and trials in which fruit and vegetables were provided, there was considerable variability in the interventions and in the cardiovascular risk profile of the participants.

Although a beneficial effect was observed for the cardiovascular risk factor outcome systolic blood pressure in the trials of dietary advice to increase the intake of fruit and vegetables, no effects were found for diastolic blood pressure or blood lipids. In the pooled analyses of fruit and vegetable provision, no effects on blood lipids were observed. Further trials of longer duration are needed to confirm these findings. Given the paucity of randomized controlled trials examining solely the effects of advice to consume more fruit and vegetables or the provision of fruit and vegetables on cardiovascular risk factors, and none examining these effects on CVD clinical endpoints, this review does not make any recommendations about changing current recommendations or practice.