Risk assessment of isoflavones in food supplements: a graphical approach to qualitative synthesis




Poster session 3


Tuesday 25 October 2016 - 10:30 to 11:00


All authors in correct order:

Smeraldi C1, Gundert-Remy U2, Pirow R3, Aiassa E4, Barizzone F5, Roncancio Pena C1
1 Food Ingredients and Packaging Unit, European Food Safety Authority (EFSA), Parma, Italy
2 Panel on Food Additives and Nutrient Sources added to Food, European Food Safety Authority (EFSA), Parma, Italy
3 Department Chemicals and Product Safety, Federal Institute for Risk Assessment, Berlin, Germany
4 Assessment and Methodological Support Unit, European Food Safety Authority (EFSA), Parma, Italy
5 Assessment and Methodological Support Unit, European Food Safety Authority (EFSA), Parma, Italy
Presenting author and contact person

Presenting author:

Camilla Smeraldi

Contact person:

Abstract text
Background and objectives: In 2015 the Panel on Food Additives and Nutrient Sources added to Food (ANS Panel) of the European Safety Authority (EFSA) completed an assessment on the potential harmful effect of isoflavones from soy, red clover and kudzu root contained in food supplements targeted at peri- and post-menopausal women on the three target organs: mammary gland, uterus and thyroid.

Methods: A systematic-review of the evidence, including both human and animal studies, was performed. Owing to the large heterogeneity of the data a formal meta-analysis was considered not to be feasible. In many of the animal studies included in our review, other endpoints were measured for their primary objectives (e.g. bone mineral density, cognitive function, behavioural patterns, etc) and effects on uterine weight were only recorded as an ancillary measure, in many cases simply to verify the oestrogenic property of the preparations tested. The systematic collection of data on this specific endpoint within our review allowed us to generate a dataset that could be transferred into a graphical representation of the results from 42 different studies. The studies were grouped according to the type of isoflavones tested and then sorted for their duration.

Results: The graphical representation of this evidence synthesis is presented in Figure 1. A statistically significant increase in uterine weight versus ovariectomised (OVX) control group of animals was represented as a full orange dot, the size of the dot being proportionate to the relative effect on the uterine weight compared with the control group within the same study. The bars on the left side of the figure represent the duration of each study (in days).

Conclusions: This graphical representation can be an effective way for synthesing evidence from a large number of animal studies reporting on the same endpoints at comparable doses. In interpreting dose-response relationships however, caution should be exercised, since in some cases default values had to be applied to convert all the doses as originally reported in the publications to mg/kg bodyweight/day of isoflavones.