Lavender and tea tree oils may cause inappropriate breast growth in boys

Between the news offices for New England Journal of Medicine and NIH’s National Institute for Environmental Health Sciences (NIEHS), you have no doubt already heard the report that topical application of lavender and tea tree oil-containing products has been linked with gynecomastia in three boys.
Yes, imagine being a private practice pediatrician whose 10-year-old male patient presents with “firm, tender breast buds measuring 3.5 cm by 4.0 cm in length and width and 3.5 cm in depth, with stretching of the areolae.” In fact, imagine being the parent…or the boy himself.

In all three cases, the boys were of average weight (obesity can sometimes be linked to gynecomastia) and had normal blood levels of sex hormones and pituitary sex hormone-releasing hormones. Other confounding endocrine disruptors associated with prepubertal gynecomastia were also lacking: oral contraceptives, marijuana, or soy products. Stopping use of the lavender and/or tea tree-containing products led to resolution of the symptoms in all three cases.
What is so beautiful to me about this study is that the pedatric group collaborated with molecular endocrinology researchers at NIEHS to use relatively simple, cell culture-based assays to demonstrate that very low concentrations of both lavender and tea tree oils are weak stimulators of estrogen receptor-alpha transcriptional activation resulting in increased production of three well-known estrogen responsive genes. More dramatically, the oils had anti-androgenic activity, substantially repressing the activity of four genes normally stimulated by the primary circulating androgen, dihydrotestosterone (DHT).
When I say “potent” or “low concentrations,” I mean that these effects are observed with the oils dissolved in cell culture at concentrations of 0.005% to 0.025% (v/v). What this means is that topical application of the oils seems to provides enough systemic absorption of the offending components to produce a combination of pro-estrogenic and anti-androgenic effects to cause development of breast tissue in these boys.
However, not so satisfying in this paper is that the offending components of each oil are not investigated or mentioned by name. Lavender (Lavandula angustifolia or L. officinalis) and tea tree (Melaleuca alternafolia) are among the most highly investigated medicinal and fragrance plants because their monoterpene compounds largely smell good and have been linked with anti-anxiety (lavender) and anti-bacterial (tea tree) effects. Howes et al. surveyed the estrogenic activity of many essential oil constituents and found that some compounds were weak estrogen agonists while others antagonized estradiol. Similarly, an editorial ten years ago in the Journal of the National Cancer Institute noted that a lavender component, perillyl alcohol, was under active investigation as an anti-estrogen to prevent or treat breast cancer.
I suspect that the investigators don’t have the exact answer at this point and are probably fractionating each oil, then mixing and matching individual pure components to find out if there are additive or synergistic among the components that resulted in these clinical effects. Lavender and tea tree contain compounds called monoterpenes, 10-carbon molecules that are made from 5-carbon branched-chain isoprene units that probably had you convulsing or glazed-over in biochemistry class. Incidentally, these compounds also form the basis of the 20-carbon skeleton that give rise to the sex hormones, so it’s really no surprise that monoterpenes would have positive or negative effects on male and female sex hormone pathways.
What remains surprising is that while some of these effects were observed with essential oil products that are highly-concentrated extracts of the plants, at least one case reported the use of only a lavender-containing shampoo and hair gel. So, the systemic bioavailability of some of these compounds must be quite good, even when applied to the scalp or other areas of skin. It also goes without saying how the expansion of herbal scents in cosmetic products might be having untoward effects on sex hormones across the population and may have also played a role in the sometimes confusing results of the Women’s Health Initiative project..
Specifically, I worry about these results being manipulated by purveyors of breast-enhancing cosmetics for young women. I believe we (as a scientific community) visited this issue a couple of years ago with concerns about the long-term breast cancer-promoting effects of such products in young women. Only time will tell how soon the hucksters will jump on this bandwagon again.
But one final note: the results of this study emphasize the need for communication between physicians and basic scientists. The pediatricians involved clearly did superb differential diagnosis of the cause of these cases, eliminating the most obvious causes and focusing on lavender and tea tree products.
But then, the reverse, bedside-to-bench relationship with NIEHS researchers led to a convincing series of cellular experiments that have provided the molecular basis of causality in these seemingly idiopathic cases of gynecomastia. Yes, I am a bit let down that the precise chemical constituents were not identified in this NEJM report and am actually surprised that the paper was accepted with studies of only the essential oil mixtures.
We’ll continue to keep an eye on the followup to this study and hope that our colleague, Molecule of the Day, has more to say about bioactive monoterpenes.
Primary source: Henley DV, Lipson N, Korach KS, and Bloch CA (2007) Prepubertal gynecomastia linked to lavender and tea tree oils. New Engl J Med 356:479-485

8 thoughts on “Lavender and tea tree oils may cause inappropriate breast growth in boys

  1. Wow. This is really fascinating and yet kind of scary at the same time. So many of our toiletries are scented in some way. Any thoughts on what other nice-smelling plants have similar monoterpenes?
    I can’t imagine the sleuthing involved on the part of the doctors to figure out this connection.

  2. Lavender and tea tree oils may cause inappropriate breast growth in boys

    *Facetious alert*
    You mean there is such a thing as appropriate breast growth in boys??

  3. knobody: I resisted the urge to link to Orac’s post because of the seriousness of gynecomastia in young boys but, of course, I leave it to my readers to provide this bit of levity. I’ve read Orac’s post and I still don’t understand why anyone would have the desire to do such a thing.
    Melissa: you raise some great points. I’m not so surprised that the effects were seen with highly-concentrated essential oil preps, but it is quite shocking that lotions, hair gels, and shampoos had enough of the offending substance(s) to cause such physiological effects. To answer your question, monoterpenes and monoterpenoids are literally everywhere in the plant kingdom, both in long-chain and cyclized form. Examples are citronella, camphor, eucalyptol, etc., but the question is which ones are estrogenic/anti-androgenic and in what combinations?

  4. Interesting, but considering the pervasiveness of lavender-scented body-care products for all ages, including infants, you would expect to see much more of this. Three cases do not equal a conclusive result. The real question, to me, is why did these three boys respond this way? It may be that only SOME people, with a particular body chemistry, will have this response.

  5. Agreed, brownrice. Case reports, just like any anecdote, should guide hypothesis-generation and do not necessarily demonstrate causality. There were trends in some of the cases that pointed to causality (such as symptoms were worse around the times of day when the lotions or “healing balms” were applied.).
    It may also be due to individual responses, as you note. My money, however, is on the pharmacokinetics of the products used, hence, I was surprised that the paper did not list the abundance of key components in the products used.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s