Objectives: to predict flavonols content of the habitual diets of free-living subjects from urine and plasma concentrations of flavonols.
Design: Ten type 2 diabetic patients (five male, five female), mean age 60 (s.e.m. 7) y and BMI 30.2 (s.e.m. 3.5) kg/m2 were treated in a random crossover design for a 2 week period on either a low flavonoid diet or on the same diet supplemented at one of two high flavonols levels (total 77.3 or 110.4 mg/day) provided by supplements of 1500 ml tea daily and 400 g fried white onion in olive oil with and without tomato ketchup and herbs.
Setting: Glasgow Royal Infirmary, University of Glasgow, Scotland.
Main outcome measures: Fasting plasma concentration, urine concentration and 24 h excretion of quercetin, isorhamnetin, kaempferol and myricetin.
Results: Plasma flavonol concentration (r=0.750, P=0.001), 24 h urine concentration (r=0.847, P=0.001) and 24 h urine excretion (r=0.728, P=<0.001) were all highly significantly related to dietary intake and gave similar estimates of intakes. Fasting plasma flavonols concentrations on habitual diets ranged from 0 to 43.7 ng/ml mean. Regression equations were constricted: total flavonols intake r=0.74, P<0.001 and quercetin intake r=0.744, P<0. 001. From these equations, flavonol intakes from habitual diets were estimated at 17-50, mean 35 mg/day. Of this, 91% was from quercetin.
Conclusions: Dietary flavonols are absorbed and appear in plasma and urine as potential biomarkers in concentrations related quantitatively to intake. Estimation of dietary intake from plasma or urine concentrations appears possible.
Sponsorship: Rank Prize Funds and Rank Foundation of the Department of Human Nutrition; Ministry of Health and Medical Education, IR Iran. European Journal of Clinical Nutrition (2000) 54, 143-149