I get often get asked why I do not endorse the consumption of stevia (see my post Is Sugar Paleo? for more information on what sugars/sweeteners I do endorse). So, as I found myself including a section on the trouble with stevia for The Paleo Approach, I felt like this was a good topic to include as a book teaser on the blog. I have a section of Chapter 3 that describes the role that sugars, blood sugar regulation, insulin sensitivity, fructose, sugar alcohols and nonnutritive sweeteners play in propagating inflammation in autoimmune disease. This excerpt is included as a standalone text box following the subsection on nonnutritive sweeteners.
This excerpt is from Chapter 3 (The Diet Link to Autoimmune Disease chapter).
Stevia is often recommended as a natural sugar substitute because it comes from the leaf of a plant (Stevia rebaudiana Bertoni). It tastes sweet on the tongue, requires very small quantities to sweeten baking, and contains no sugar. While some experts advise caution against purified and manufactured forms of stevia, green leaf stevia is typically endorsed. On the surface, it sounds like a perfect solution. However, I do not recommend the consumption of stevia, even in its most natural form. The chemicals responsible for the sweet taste of stevia are called steviol glycosides (there are at least ten different steviol glycosides present in the stevia plant). Purified/manufactured forms of stevia often isolate one or two of these steviol glycosides whereas green leaf stevia (which is simply the dried and powdered leaves of the stevia plant) contain all ten.
Steviol glycosides are synthesized in the same pathway and end up being structurally very similar to the plant hormones gibberellin and kaurene. This means that steviol glycosides have a hormone structure. The majority of toxicological studies establish that stevia is safe, however there are some studies showing that it can act as a mutagen and may increase the risk of cancer (these studies are in the minority and tend to use quite high concentrations, so they are readily discarded in discussions of the overall safety of consuming stevia). Whether or not stevia causes genetic mutations is not the only cause for concern, however (even if safety studies focus on this particular property). For those with autoimmune disease, in which hormones have such a dramatic impact on disease development and progression, the impact of consuming stevia on hormone regulation is relevant.
There is evidence that steviol glycosides have contraceptive effects in both males and females. In particular, one specific steviol glycoside, called stevioside, has been shown to have potent contraceptive properties in female rats, implying that stevia may have an impact on estrogen, progesterone or both. In another study, male rats fed stevia extracts showed a decrease in fertility, reduced testosterone levels and testicular atrophy, potentially attributable binding of steviol glycosides with an androgen receptor. Although no studies have been conducted evaluating the impact of stevia on fertility in humans, the stevia plant was traditionally used to control the fertility of women by the Guarani Indians in southern Brazil. While small and occasional consumption of stevia likely has little to no impact on general health, it should not be consumed on a regular basis especially by those with altered hormone balance and dysfunctional immune systems.
Brusick DJ. A critical review of the genetic toxicity of steviol and steviol glycosides. Food Chem Toxicol. 2008 Jul;46 Suppl 7:S83-91.
Mazzei Planas G and Kuć J. Contraceptive properties of Stevia rebaudiana. Science. 1968 Nov 29;162(3857):1007.
Melis MS Effects of chronic administration of Stevia rebaudiana on fertility in rats Journal of Ethnopharmacology 1999 Nov 67(2):157–161
Melis MS. Chronic administration of aqueous extract of Stevia rebaudiana in rats: renal effects. Journal of Ethnopharmacology 1995. July 47(3):129–134
Oliveira-Filho RM et al. Chronic administration of aqueous extract of Stevia rebaudiana (Bert.) Bertoni in rats: Endocrine effects. General Pharmacology: The Vascular System. 1989. 20(2):187–191