You may have heard the rumors: a clinical trial proves that lavender oil and tea tree oil have estrogenic-like effects in young boys, causing prepubertal gynecomastia!
In the early 2000s, a pediatrician noticed a trend within his practice. Three boys, of different ages and from different families, all presented with the same extremely rare disorder: prepubertal gynecomastia. Noticing this cluster of cases, he began looking for similarities among these boys. Were they exposed to any of the myriad of hormone disruptors in the environment? No. Upon investigating further, he found that the only discernible trend among the patients was exposure to lavender oil containing products in the form of a healing balm, haircare system (shampoo and leave-in gel containing both lavender oil and tea tree oil), and body soap.
As the cases became curiouser and curiouser, discontinuation of these products reversed the condition and all three boys returned to normal shortly thereafter. The pediatrician notified a team of researchers about the experience and they examined lavender essential oil and tea tree essential oil for potential estrogen-like and anti-androgenic effects (in other words, hormone disruption similar to what is seen in many environmental toxins) that could possibly explain this bizarre coincidence. Such actions were found, though in a dose-dependent manner requiring extremely large doses, and the researchers published the findings in a case study followed by a press release from the National Institute of Environmental Health Sciences.
When this case report was published in 2007, it drew both excitement and ire from a wide range of individuals, including those who leapt at the chance to both villainize and defend aromatherapy. Those eager to discredit the relatively large range of safety with the use of lavender essential oil and tea tree essential oil insisted that they should be avoided by children, pregnant women, women prone to hormonally related cancers–just about anyone due to these documented risks. After all, they claimed, the New England Journal of Medicine published proof.
Others, however, were eager to defend lavender and tea tree oils, attempting to discredit the validity of the study or pick apart perceived flaws of the study. The crux of this argument centered on the lack of depth and detail regarding the lavender oil-containing products which comprised the exposures in the case report. One key point being that the specific products were not identified, thus perhaps the researchers failed to identify the actual exposure. (Though it should be noted that in a summer 2007 response to such claims, the authors clarify that the ingredients were identified directly from the ingredient label on the products being used in the practice.) Another key point of the argument is that other ingredients in the products must be responsible for the outcome.
So who is right? And more specifically, are lavender oil and tea tree oil safe? Or are they potentially dangerous?
The Missing Piece of the Puzzle
This is where a basic understanding of research methods saves the day. While we could easily create a lengthy list of reasons that the case reports don’t establish causation and point out every instance the study lacks clarity, a more accurate approach would be to utilize an elementary level of health research interpretation and let the paper speak for itself. With this second approach, we utilize the scientific evidence within the context it was intended to be interpreted, increasing accuracy by employing an evidence based approach instead of succumbing to cognitive bias.
The first step to understanding the relevance or significance of any research study is to identify the type of study and how it is designed to be interpreted. That critical step provides the context required to properly understand the study’s findings and helps us to avoid incorrectly identifying inherent qualities of a study as errors or poor methodology.
This article is a case report. Case reports are studies which analyze individual outcomes which are novel or unique so that researchers can evaluate them further. In other words, case studies are merely analyzed and polished anecdotes which have been evaluated by professionals. They typically cover a single case but they can include as many as 5-10. They are not, however, epidemiological or experimental studies. They are never intended to be a substitute for the type of research methods which establish causation, and they do not even employ the type of methods used in descriptive research to establish firm correlations.
Case reports share unique and novel cases so that both researchers and providers can be aware of the unusual outcomes that take place, building upon those events scientifically until clear evidence about an issue is obtained. By definition, they cover unusual outcomes which are not typically seen among the general population. It is unusual to see lavender oil and tea tree oil linked to breast growth in young boys, hence the value of this report documenting a potential subject for further research. This point is emphasized by the researchers:
…our findings should be interpreted carefully. . . .We would remind readers that we observed an unusual clinical phenomenon in prepubertal boys that resolved on discontinuation of the topically applied products. (Author response, June 2007)
A good case study then includes information about potential mechanisms of action which may explain the novel or unusual outcome. In this example, the researchers took it a step further and looked to see whether or not it was even possible for lavender essential oil and tea tree essential oil to exhibit estrogen-like actions on the body. And, as it turns out, in a dose-dependent relationship, such actions are possible. Though possible and probable are far from being synonyms. In fact, the dose-dependent relationship demonstrates that the estrogen-like actions are extremely weak and that achieving such actions in the body would be extremely difficult. That is why the researchers proposed in the article that perhaps an abnormally low threshold exists among certain individuals, causing them to experience adverse outcomes at levels which would not typically impact the general population.
This is a crucial and valuable type of research when interpreted in the proper context and in light of what is already know within the literature as a whole. Identifying rare cases where a small portion of the population exhibits reactions or outcomes from specific exposures that are not experienced by the population as a whole tells us how to establish contraindications for products, whether those products are environmental exposures, drugs, or even natural botanicals.
Poor Study Quality? Or Poor Interpretation?
Researchers know that protocols, safety recommendations, and procedures are not established through the use of case reports, rather through the use of a wide array of research studies. Case reports are to be interpreted within the limitations and abilities of this particular type of research, which does not include any notion that they provide evidence of causation in any way.
Which means that both the use of this study to villainize the uses of these essential oils and the loud criticisms of the way in which this study was conducted both reflect basic misunderstandings of the way research works and how it should be interpreted.
Case reports are uncontrolled environments reflecting unique outcomes without the consistency we would use in a clinical trial or the analysis of all possible variables we would use in an epidemiological study. They are not designed to establish any form of causation, rather to report on unusual cases and provide some insight in the way of hypotheses regarding how such outcomes could be possible. It is a common mistake to assume that all studies are to be evaluated under the same lens, which is why research methods as a whole comprise such a critical role in aromatherapy education.
So what would be the appropriate interpretation for this study when it comes to the use of lavender oil and tea tree oil? In this case, we don’t have to spend a great deal of time or utilize a significant amount of expertise in research methods to interpret this one. The authors conclude the article with a fairly specific call to action:
Until epidemiologic studies are performed to determine the prevalence of gynecomastia associated with exposure to lavender oil and tea tree oil, we suggest that the medical community should be aware of the possibility of endocrine disruption and should caution patients about repeated exposure to any products containing these oils. – Henley, et.al., 2007 (emphasis ours)
In other words, in the absence of the type of research studies which are used to establish causation, if you are a medical professional, you may want to know that this occurred if you work with patients who may find this relevant. This is a common conclusion of case studies as it is well understood among the consumers of scientific research that causation has been neither established nor implied.
Why publish it at all?
Imagine if a pharmaceutical appeared to be linked to a rare disorder in children. A pediatrician noted this apparent trend and a team of researchers investigated further, finding that there might be a dose that could cause such results. And then they hid their findings, never allowing them to see the light of day. We would be furious. Failure to share these findings, whether they involve lavender oil and tea tree oil or some new pharmaceutical, means that a potential threat remains hidden. The findings are not studied further and we never obtain clear answers about this potential risk while children are subjected to the exposure on a regular basis. Case studies are published so that these potential threats do not remain hidden below the surface. Attention is brought to the findings so that researchers can dig deeper to enhance safety guidelines, protecting those who may be at risk.
After publication and the following press release, aromatherapists and essential oil manufacturers have called the report junk science and insisted that it should never have been published, especially not in such a prestigious journal. After all, they argue, causation is not established in this study so it should be verified first. Yet, the novel findings of this study and the fact that it has been discussed and cited numerous times explain exactly why such a study is published and why it contributes valuable information to the field.
Most noteworthy, the study did exactly what case reports are designed to do–it promoted further research in the potential actions of lavender oil and tea tree oil. Since its publication, other researchers have investigated this claim and expanded the body of evidence that exist surrounding the potential estrogen-like actions of lavender essential oil providing large bodies of evidence that such actions do not pose a threat (Politano, et al, 2013).
And equally as important, doctors have responded to the findings as well, looking for these potential exposures in additional cases of idiopathic (meaning the cause is unknown) prepubertal gynecomastia and published findings of cases which existed without a history of such exposures (Demirbilek et al, 2014). This is how scientific discovery is designed to work. Low level observations are shared so that in-depth research can take place to expand our knowledge about how these products work and don’t work in the body. And where there is potential for risk, medical providers consider such caveats and look for additional cases that researchers may find interesting.
The Story Doesn’t End There!
The additional research prompted by this study found that lavender essential oil and tea tree essential oil do not typically present actions which are strong enough to have a measurable effect on the body, even at large doses, based upon rat models (Politano, et al, 2013). Additionally, one study found that products containing tea tree may even present anti-estrogenic effects (Myers, et al, 2015).
However, the case is far from settled as at least three additional case reports have been published, providing additional evidence that further research regarding lavender oil exposure, as it relates to a specific possible underlying factor, should be conducted (Diaz et al, 2016). In these additional cases, two of the three young boys were exposed to lavender oil through repeated exposure to a cologne over several years (the third child being exposed by use of a lavender oil-containing shampoo) and laboratory tests did confirm the presence of lavender essential oil in the cologne. Additionally, as with the first case series, the symptoms resolved after the exposure to lavender oil was discontinued.
Perhaps in the future we may discover that certain individuals have yet to be identified factors predisposing them to adverse outcomes from specific types of lavender oil, explaining some cases of prepubertal gynecomastia which are currently labeled as idiopathic. Or perhaps these cases do exist as mere coincidences and causation will be explained through some other yet-to-be-identified factor. Ultimately, such clarity can only be obtained through the epidemiological studies the 2007 authors recommend. Such studies are specifically designed to identify rare outcomes among exposures which are widely found within a population.
Until that time, the practical application of the original report remains: if someone you love experiences prepubertal gynecomastia while being exposed to lavender essential oil and/or tea tree essential oil, and you have already ruled out the extremely large number of environmental toxins which are known to exhibit estrogen-like responses, removing them is worth a try. Otherwise, as is well documented in the literature, responsible use of lavender oil for both skincare applications and medicinal effects is clearly established as being safe for both sexes of all ages.
Case Report: Henley, D., Lipson, N., Korach, K., & Bloch, C. (2007). Prepubertal gynecomastia linked to lavender and tea tree oils. New England Journal of Medicine. 356(5):479-485.
Bloch, C. A., & Korach, K. S. (2007). Prepubertal gynecomastia linked to lavender and tea tree oils. The New England Journal of Medicine, 356(24), 2543-2544.
Demirbilek, H., Bacak, G., Baran, R. T., Avcı, Y., Baran, A., Keleş, A., … & Hussain, K. (2014). Prepubertal Unilateral Gynecomastia: Report of 2 Cases. Journal of clinical research in pediatric endocrinology, 6(4), 250.
Diaz, A., Luque, L., Badar, Z., Kornic, S., & Danon, M. (2016). Prepubertal gynecomastia and chronic lavender exposure: report of three cases. Journal of Pediatric Endocrinology and Metabolism, 29(1), 103-107.
Myers, S. L., Yang, C. Z., Bittner, G. D., Witt, K. L., Tice, R. R., & Baird, D. D. (2015). Estrogenic and anti-estrogenic activity of off-the-shelf hair and skin care products. Journal of Exposure Science and Environmental Epidemiology, 25(3), 271-277.
Politano, V. T., McGinty, D., Lewis, E. M., Hoberman, A. M., Christian, M. S., Diener, R. M., & Api, A. M. (2013). Uterotrophic assay of percutaneous lavender oil in immature female rats. International journal of toxicology, 32(2), 123-129.