Abstract

OBJECTIVES:

• To review all prospective studies which provided data about total fruits or vegetables, their subgroups, and specific types of vegetables and fruits, in relationship with risk of type 2 diabetes mellitus.
• To define the amount of consumption found to be related with the described effects on type 2 diabetes mellitus.
• To define possible effect modification by confounders.

DATA SOURCE: The Pubmed database was searched (No start date - July 6, 2009) for relevant articles using the keywords "prospective, cohort, follow-up, longitudinal" combined with "vegetables, fruit", and a fair amount of other keywords. The exact search term is described in the methods.
Prospective studies published in the English language were included. Reference lists were searched for additional articles.
RESULTS: 25 articles were found which provided information about 15 different cohorts.
Of these, 0 articles were excluded.
Results are described when any evidence for an association - as defined in the Methods - was found. In addition, data about total vegetables or fruits is described.

• Data about total vegetables was provided by 9 cohorts, including 12,206 cases. No evidence was found for an association.
• Data about total fruits was provided by 10 cohorts, including 12,731 cases. Suggestive evidence was found for a protective effect of high vs low consumption. Inclusion of effects from intermediate levels of consumption, strengthened the evidence: Total fruit possibly protects against type 2 diabetes risk at the range of consumption of ≥ 156-170 g/day, but ≤ 239 g/day.
• Both green-yellow vegetables, and yellow vegetables possibly protect against risk of type 2 diabetes. The association with green-yellow vegetables was only examined among Asian populations, and the association with yellow vegetables was only examined among women.
• Suggestive evidence was found for a protective effect of green (leafy) vegetables and apples. No level of consumption could be defined for these effects.
• Potatoes and fruit punch possibly increase risk of type 2 diabetes. No level of consumption could be defined for these effects. The association with fruit punch was only examined among US women.
• Suggestive evidence was found for an increased risk of type 2 diabetes with consumption of ≥ 1 drink of sweetened fruit juices/day.

CONCLUSION: Total fruit possibly protects against risk of type 2 diabetes at an intermediate level of consumption (≥ 156-170 g/day, but ≤ 239 g/day), but no evidence was found for an association with total vegetables. Both green-yellow vegetables, and yellow vegetables are possibly protective, and suggestive evidence was found for a protective effect of green (leafy) vegetables and apples.
Potatoes and fruit punch possibly increase risk of type 2 diabetes, and suggestive evidence was found for an increased risk of sweetened fruit juices (at consumption of ≥ 1 drink/day).

Total vegetables and type 2 diabetes mellitus.

Data was provided by 9 cohorts, including a total of 12,106 cases.
Protective effects were found in 2 cohorts (11, 14), including 2,340 cases (19% of all cases). No other (non)significant associations were found in any cohort.

Inclusion of intermediate levels of consumption:
Significant protective effects at any level of consumption were confined to results from 2 cohorts, including 1,988 cases (Montonen J, at 42-78 g/day; and Villegas R, at ≥ 182 g/day). No other (non)significant effects were found at any level of consumption.

Conclusion: Protective effects were found in 2 out of 9 cohorts, including 19% of all cases. No evidence was found for an association between total vegetables and risk of diabetes type 2.

Total fruit and type 2 diabetes mellitus.

Data was provided by 10 cohorts, including a total of 12,731 cases.
Significant protective associations were found in 3 cohorts, incuding 5,647 cases (44% of all cases). No other (non)significant associations were found in any cohort.

Inclusion of intermediate levels of consumption:
Significant protective associations at any level of consumption were found in 3 cohorts (Bazzano LA, at the highest quintile of consumption [median 200 g/d]; Montonen J, at ≥ 156 g/d; and Villegas R, at the 2nd-3rd quintile of consumption [median 170 and 239 g/d]), these cohorts included 6,517 cases (51% of all cases).
Significant protective effects with intermediate levels of consumption were found in 2 other cohorts: among overweight women at the 3rd quintile of consumption [median 153 g/d] in one cohort (Liu S), and with fruit consumption 2-5.9 times/day in another cohort (Hodge AM). Fruit consumption was defined in frequency of consumption, rather than g/servings/cups over a given time unit in the latter cohort. But if frequency equals servings, the range of consumption for a protective effect would overlap with the range from the first 3 cohorts (≥ 156-170 g/day, but ≤ 239 g/day), and thereby stengthen the possibility of an effect.

Conclusion: Significant protective associations of high vs low consumption were found in 3 out of 10 cohorts, including 44% of all cases. Suggestive evidence was found for a protective effect of high fruit consumption against type 2 diabetes risk.
Significant protective effects at an intermediate range of consumption were found in 3 cohorts - two of which were of moderate-large size -, including 51% of all cases. Findings from 2 other cohorts seem to strenghten the possibility of an effect at this range of consumption. Fruit possibly protects against type 2 diabetes risk at the range of consumption of ≥ 156-170 g/day, but ≤ 239 g/day.