Abstract P389: Flavonoid intake and incident hypertension in French women

Abstract

Background: Flavonoids have been shown to lower blood pressure, however, there is limited information on the relation of long-term intake of flavonoids and the incidence of hypertension. Methods and Results: We evaluated the relations of total flavonoid and flavonoid subclasses (flavonols, flavones, flavanones, anthocyanins, flavan-3-ols and polymers) intake to self-reported hypertension in a prospective cohort study among 40,574 disease-free French women in the E3N cohort, a component of the EPIC cohort. Diet was assessed in 1993 using a validated diet history questionnaire. We used the Phenol-Explorer database to estimate total flavonoid and flavonoid subclass intake and categorized them in quintiles. Between 1993 and 2007, 9,348 cases of incident self-reported hypertension were identified using follow-up questionnaires. The validity of self-reported hypertension was evaluated using a health claims database and the positive predictive value was 82%. In multivariable analyses we adjusted for age, education, smoking, BMI, physical activity, menopausal status, hormone use, diabetes, hypercholesterolemia and intake of calories, alcohol, processed meat, caffeine, magnesium, potassium and omega-3. In multivariable Cox regression models, comparing the highest quintile of total flavonoid intake to the lowest, high flavonoid intake was inversely associated with the rate of self-reported hypertension (HR=0.91; 95%CI 0.85, 0.98; p-trend=0.01). When comparing extreme quintiles of intake, we observed inverse associations with flavonols (HR=0.90; 95%CI 0.84, 0.97; p-trend=0.03), anthocyanins (HR=0.91; 95%CI 0.84, 0.97; p-trend=0.008) and polymers (HR=0.91; 95%CI 0.85, 0.97; p-trend=0.005). We evaluated these relations restricting analyses to individuals ≤ 60 years and results did not change. The main foods contributing to total flavonoid intake were tea, chocolate, plums and wine. Comparing the highest to the lowest quintile of plum intake we observed a 14% lower rate of self-reported hypertension (HR=0.86; 95%CI 0.81, 0.93; p-trend=<0.0001). No associations were observed for the other main contributors. Conclusion: In this large prospective cohort of French women, we observed a weak association between intake of total flavonoid, flavonols, anthocyanins and polymers intake and self-reported hypertension. Plum consumption appears to be associated with a lower rate of hypertension.