Fluid intake and the risk of urothelial cell carcinomas in the European Prospective Investigation into Cancer and Nutrition (EPIC).
International journal of cancer 2010 ; 128: 2695-708.
Ros MM, Bas Bueno-de-Mesquita HB, Büchner FL, Aben KK, Kampman E, Egevad L, Overvad K, Tjønneland A, Roswall N, Clavel-Chapelon F, Kaaks R, Chang-Claude J, Boeing H, Weikert S, Trichopoulou A, Orfanos P, Stasinopulou G, Saieva C, Krogh V, Vineis P, Tumino R, Mattiello A, Peeters PH, van Duijnhoven FJ, Lund E, Gram IT, Chirlaque MD, Barricarte A, Rodriguez L, Molina E, Gonzalez C, Dorronsoro M, Manjer J, Ehrnström R, Ljungberg B, Allen NE, Roddam AW, Khaw KT, Wareham NJ, Boffetta P, Slimani N, Michaud DS, Kiemeney LA, Riboli E
DOI : 10.1002/ijc.25592
PubMed ID : 20715171
PMCID : 0
Abstract
Results from previous studies investigating the association between fluid intake and urothelial cell carcinomas (UCC) are inconsistent. We evaluated this association among 233,236 subjects in the European Prospective Investigation into Cancer and Nutrition (EPIC), who had adequate baseline information on water and total fluid intake. During a mean follow-up of 9.3 years, 513 first primary UCC occurred. At recruitment, habitual fluid intake was assessed by a food frequency questionnaire. Multivariable hazard ratios were estimated using Cox regression stratified by age, sex and center and adjusted for energy intake, smoking status, duration of smoking and lifetime intensity of smoking. When using the lowest tertile of intake as reference, total fluid intake was not associated with risk of all UCC (HR 1.12; 95%CI 0.86-1.45, p-trend = 0.42) or with risk of prognostically high-risk UCC (HR 1.28; 95%CI 0.85-1.93, p-trend = 0.27) or prognostically low-risk UCC (HR 0.93; 95%CI 0.65-1.33, p-trend = 0.74). No associations were observed between risk of UCC and intake of water, coffee, tea and herbal tea and milk and other dairy beverages. For prognostically low-risk UCC suggestions of an inverse association with alcoholic beverages and of a positive association with soft drinks were seen. Increased risks were found for all UCC and prognostically low-risk UCC with higher intake of fruit and vegetable juices. In conclusion, total usual fluid intake is not associated with UCC risk in EPIC. The relationships observed for some fluids may be due to chance, but further investigation of the role of all types of fluid is warranted.