Background Fruit intake has been associated with a reduced risk of cardiovascular disease (CVD). However, the effects of individual sub-groups of fruit on CVD have rarely been investigated. The aim of the current study is to explore the association between total fruit intake, sub-groups of fruit intake according to polyphenol content and CVD mortality using data from the UK Women’s Cohort Study (UKWCS).

Method Total fruit intake (g/day) was obtained from 30,458 participants (mean age 52 years) who completed a 217 item food frequency questionnaire between 1995–1998 at baseline. Fruit intakes were further sub-categorised into groups according to similarities in polyphenol content listed in Phenol Explorer (http://www.phenol-explorer.eu/), including berries, citrus, drupes, pomes and tropical fruits. For inclusion in the analysis, participants were free from history of CVD. Mortality records for participants were reported via the NHS Central Register, and during the follow-up period of 18.9 years, 286 fatal CVD cases [138 coronary heart disease (CHD), 148 stroke] were observed. Survival analysis was undertaken using Cox proportional hazard ratios. Adjustments included age, body mass index, alcohol, physical activity, smoking and socio-economic status.

Results Total fruit intake was significantly, negatively associated with risk of fatal CVD, with a 6% decrease in risk (CI 0.90–0.99, p-value = 0.013) for every additional 80g portion of fruit consumed. Total fruit intake was also significantly associated with fatal CHD when investigated separately, with a 7% decrease in risk (CI 0.87–0.99, p-value = 0.031) for every additional portion of fruit consumed. Total fruit intake was not associated with risk of fatal stroke, however citrus fruit consumers had a significantly lower risk of fatal stroke, [HR 0.44 (CI 0.26–0.76)] than non-consumers. A significant dose response was not observed for this association. Other sub-groups of fruit analysis as mentioned above did not show any association with fatal CVD.

Conclusion These analyses suggest that consumption of total fruit intake may be protective against fatal CVD. However, whilst these data do not suggest a differential association with certain sub-groups of fruit, they do lend support to widely promoted guidelines promoting fruit consumption for health.