Is hand soap dangerous to you?
A news alert is making the rounds through the popular press these days; I noticed it on the tech blog io9.com in a summary with the provocative title “Hand soap is killing you.” The article references a paper published in the highly regarded medical journal Proceedings of the National Academy of Sciences. I haven’t read the article beyond the abstract (it is behind a paywall at pnas.org), but the abstract does summarize the methods and results sufficiently to be able to make some educated conclusions about the risks of antibacterial soaps.
Researchers at the University of Colorado, and at the University of California-Davis examined the effects of the common antibacterial compound triclosan on muscle function in mice and fish. Triclosan is a chlorinated derivative of phenol that is widely used for its antibacterial and antifungal abilities. It affects microorganisms by inhibiting the biosynthesis of fatty acids, and in low concentration exhibits significant bacteriostatic properties. It itself it is not significantly toxic to mammals, except when it is present in surface waters in the environment it can degrade to form dioxins, which are carcinogenic to a wide range of species. According to my bottle of Dawn brand antibacterial dish detergent, triclosan is the active ingredient present at a concentration of 0.1% weight/volume, or 0.7 grams of triclosan per 24 ounce bottle.
The PNAS paper uses two animal model systems and one in vitro tissue culture system to examine the physiologic effects of triclosan on muscle cells. In the mouse model, triclosan was administered intraperitoneally via injection at a concentration of at least 12.5 mg per kg of body weight, and was assessed by measuring grip strength following exposure. In the fish model, triclosan was present in the water at a concentration greater than 0.52 µM, and was assessed by measuring swimming performance. Finally, in an isolated myotube tissue culture model, triclosan was present at micromolar concentrations, and the effects were assessed by measuring the ability of the cells to respond to stimuli.
Let’s look at the mouse model first; the authors only found effects on mice when triclosan was administered via i.p. injection at greater than 12.5 mg per kg of body weight. First, that is not the standard mechanism that a human might expect to be exposed to triclosan. Second, by working out the math to extrapolate to a human (12.5 mg times 65 kg mass) this would require roughly 800 mg triclosan for an average human being to reproduce these experimental conditions. Recall that my 1.5 pint bottle of dish detergent only has 700 mg of triclosan in the entire container. This result suggests that the triclosan risk is pretty unlikely for a human.
The fish and tissue culture models are a bit different. Triclosan is present and exposed via aqueous solution at a concentration of around 1 micromolar. Again by working out the math (290 g/mol mass times 10-6 mol/L) works out to a concentration about 200 fold more dilute than that found in my dish detergent. The fish model is difficult to put into perspective, as the abstract does not clearly indicate the exposure time necessary to produce the effects, but presumably it is a continuous exposure which is not the case with me and my dish soap. The tissue model is perhaps the cleanest; experiments like these use short time exposures to the compound of interest, and directly measure the effects on the cells in question. However, we have to consider delivery of the compound in the real world. Triclosan in my dish soap is present on my skin at a certain level, and by the time the compound gets into my body it will be at a far lower level.
So what is the take home message from this? It appears that triclosan, a very common antimicrobial compound, has bad effects on the human body. This is not news; it has been known that triclosan in the environment is a contributor to the presence of dioxins, which in turn are potent carcinogens. This present study outlines a very clear effect that triclosan is having on a specific set of tissues (muscles) in the animal body, but the effects of triclosan in the laboratory are at levels that are significantly higher than those encountered in the home. Consequently, I don’t feel that the individual health risk of triclosan at nominal exposure levels is particularly worrisome. It might be more useful though to reexamine the benefits and overall costs of using antibacterial compounds in the home from the point of view that they are contributing to the continued emergence of antimicrobial resistance in common microorganisms.