Consumption of vegetables and fruits and urothelial cancer incidence: a prospective study

Although most epidemiological studies concerning urothelial cancer support a possible protective effect of vegetable and fruit consumption, previous s tudies have been inconsistent with regard to which vegetables and fruits ma y be responsible for an inverse association. The association between the co nsumption of 21 vegetables and nine fruits and urothelial cancer risk was a ssessed in the Netherlands Cohort Study among 120,852 men and women aged 55 -69 years at baseline in 1986. After 6.3 years of follow-up, 538 incident c ases and 2,953 subcohort members with complete vegetable data and 569 cases and 3,123 subcohort members with complete fruit data were available for ca se-cohort analyses. In multivariable case-cohort analyses, the following ag e-, sex-, and smoking-adjusted incidence rate ratios for groups of vegetabl e and fruit consumption were observed (comparing highest versus lowest quin tile of consumption): total vegetables, 0.91 [95% confidence interval (CI): 0.65-1.27]; cooked vegetables, 0.98 (CI: 0.71-1.35); raw vegetables, 1.10 (CI: 0.78-1.53); cooked leafy vegetables, 0.89 (CI: 0.65-1.23); raw leafy v egetables, 0.94 (Cl: 0.73-1.22); pulses, 1.03 (CI: 0.74-1.44); brassicas, 0 .75 (CI: 0.54-1.04); allium vegetables, 0.89 (CI: 0.67-1.19); total fruit, 0.74 (CI: 0.53-1.04); and citrus fruit, 0.85 (CI: 0.62-1.17). For three sep arate items (cauliflower, cooked carrots, and mandarins), a statistically s ignificant inverse association was seen, whereas for other specific vegetab les or fruit, no statistically significant association was observed. The da ta are suggestive of an inverse association between the consumption of bras sicas, total fruit, and urothelial cancer risk, whereas total vegetable con sumption did not appear to be associated with urothelial cancer risk.