| Description |
Caffeinated and fructose-rich beverages are widely consumed among women of reproductive age but their association with reproductive hormones is not well understood, due in part to inadequate exposure assessment. Our objectives were to 1) assess the relationship between caffeine and fructose intake and reproductive hormones in healthy premenopausal women, evaluating potential effect modification by race; and 2) determine the validity of the Food Frequency Questionnaire (FFQ) for measuring monthly caffeinated beverage intake compared to multiple 24-hour dietary recalls (24HDR). The BioCycle Study (2005-2007) prospective cohort (n=259) included women, ages 18-44, who were followed for 2 menstrual cycles, providing fasting blood specimens at up to 8 visits per cycle, 4 24HDRs per cycle, and an FFQ at the end of each cycle. Caffeine intake ≥200 mg/day was inversely associated with free estradiol (E2) concentrations among white women (β=-0.15 [95% confidence interval (CI): -0.26, - 0.05] and positively associated among Asian women (β=0.61 [95% CI: 0.31, 0.92] after taking into account potential confounders. Women who consumed more added sugar than an average American woman (≥ 73.2 grams/day) or above the 66th percentile in fructose intake (≥ 41.5 grams/day) had elevated free E2 concentrations compared to women who consumed less. Women who consumed ≥1 cup/day of sweetened soda had elevated free E2 (β=0.15 [95% CI: 0.06, 0.24]. Neither artificially sweetened soda intake nor fruit juice iv intake ≥1 cup/day was significantly associated with reproductive hormones. Caffeine intake reported in the FFQ was greater than that reported in the 24HDRs (mean=114.1 versus 92.6 mg/day; P=0.006) despite high correlation (r=0.80, P<0.001) and moderate agreement (kappa=0.56, 95% CI: 0.42-0.70). In summary, moderate caffeine consumption was associated with reduced E2 among white women and elevated E2 among Asian women. Added sugars, total fructose, and sweetened soda were associated with elevated E2 among all races. Further research on the association between caffeine, caffeinated beverage components and reproductive hormones, and whether these relationships differ by race, is warranted. Additionally, although caffeine exposures are highly correlated, absolute intakes differ significantly between measurement tools, highlighting the importance of considering potential misclassification of caffeine exposure when conducting women's health epidemiologic studies. |
