The increasing use of botanical food supplements has raised concerns among scientific and regulatory communities. Occasional cases of intoxication have occurred from misuse, misidentification of the botanical species or contamination with extraneous plants. Consequently, risk assessment of botanical products requires adequate specification of identity and composition. Sources vary from staple food plants to herbals used in folk medicine; the supplement may comprise the whole plant, extracts thereof or purified components. This variability poses problems in adopting a generic approach to their risk assessment. The nature and extent of toxicological testing required will depend on: nature of the supplement, prior knowledge of human consumption, likely exposure and nutritional impact, and intended beneficial effects. Generally, for herbs or complex extracts, it is not possible to make a risk assessment on the basis of a single active component as more than one may be of toxicological significance and matrix effects may affect bioavailability. Nevertheless, studies on single components may be useful in elucidating potential interactions. Botanical supplements are intended to produce physiological effects, so there is a need to distinguish a No Observed Effect Level from a No Observed 'Adverse' Effect Level and the margin of exposure between that producing the desired effect and the upper safe level may be smaller than that adopted for food additives and contaminants. Human studies of efficacy and possible side effects may help in determining the acceptable margin of exposure. A decision tree will be presented to assist in determining the extent of data requirements based on the nature of the product.
[1]
J. Betz,et al.
Determination of pyrrolizidine alkaloids in commercial comfrey products (Symphytum sp.).
,
1994,
Journal of pharmaceutical sciences.
[2]
E. Ernst.
The RiskBenefit Profile of Commonly Used Herbal Therapies: Ginkgo, St. John's Wort, Ginseng, Echinacea, Saw Palmetto, and Kava
,
2002,
Annals of Internal Medicine.
[3]
A G Renwick,et al.
Human variability and noncancer risk assessment--an analysis of the default uncertainty factor.
,
1998,
Regulatory toxicology and pharmacology : RTP.
[4]
M. Chambolle.
Assessment of extreme levels of chronic food intakes.
,
1999,
Regulatory toxicology and pharmacology : RTP.
[5]
M. Gershwin,et al.
Variability in commercial ginseng products: an analysis of 25 preparations.
,
2001,
The American journal of clinical nutrition.
[6]
P. Vereerstraeten,et al.
Urothelial carcinoma associated with the use of a Chinese herb (Aristolochia fangchi)
,
2000,
The New England journal of medicine.
[7]
Elizabeth Matisoo-Smith,et al.
Taste sensitivity to 6-n-propylthiouracil predicts acceptance of bitter-tasting spinach in 3-6-y-old children.
,
2002,
The American journal of clinical nutrition.