Exploring the impact of 3,3'-diindolylmethane on the urinary estrogen profile of premenopausal women

Background 3,3'-diindolylmethane (DIM) is a phytonutrient derived from cruciferous vegetables that is an often-used supplement in the complementary and alternative medicine space. The most common goal for providers when recommending DIM to their patients is to alter estrogen metabolism, yet research into DIM's effect on the estrogen profile is lacking in the published literature. The objective of this study was to comprehensively evaluate DIM's effect on the urinary estrogen profile. Methods In this retrospective cohort study, we analyzed data from a clinical laboratory, including urinary estrogen and estrogen metabolite concentrations. Analyte concentrations were determined from dried urine samples using a gas chromatography-tandem mass spectrometry assay. Individuals were separated into two groups, either reporting taking DIM (N = 909) or reporting not taking DIM (N = 18,385). Comparisons between individuals in these two groups were made using the Wilcoxon rank sum test. Additionally, we were also able to explore a subset of women who had laboratory results in the database before and after initiating DIM treatment (N = 53). In this subset, differences were assessed with Wilcoxon signed rank tests. Results In the larger group that was separated into women reporting either DIM use or no use, significant differences were observed in the concentrations of almost every urinary estrogen and estrogen metabolite (with the only exception being 2-methoxyestrone) in the urinary estrogen profiles of those taking DIM compared to those not taking DIM (all P values < 0.001). In the smaller subset of individuals with results before and after initiating DIM use, differences were only seen in 4 of the urinary estrogens and estrogen metabolites (P < 0.001 for estrone, estradiol, estriol, and 16-hydroxyestrone). Differences in total estrogens were significant in both the larger group and the smaller subset (both with P < 0.001). Additionally, observed differences in the ratios of metabolites followed a similar trend with more significant differences observed in the larger group. Notably, the 2-hydroxyestrone:16-hydroxyestrone ratio increased significantly in both the larger group and the smaller subset with results before and after DIM use. Conclusions The results of this study provide the most comprehensive evaluation to date of DIM's effect on the urinary estrogen profile. Additionally, the results demonstrate that the dried urine collection and accompanying assay used capture changes that are similar in direction, but not necessarily magnitude, to previous reports in the literature. Considered together, these two things highlight the clinical validity and utility of this approach to the evaluation of DIM supplementation and suggest the need for additional studies using this approach to fully understand the potential clinical utility of DIM.