Toxicity Myths:
The Actual Risks of Essential Oil Use

By Ron Guba

In the wide variety of Aromatherapy books and periodicals available today, we find many recommendations regarding the safe, therapeutic use of essential oils, often contradictory and seldom supported by either references, research or actual clinical experience. In this presentation, I will explore the range of recommendations made and address the validity of each, especially addressing the underlying assumptions and reasons for these statements.

I have personally been involved in the both the practice and the business of Aromatherapy since arriving in Australia in 1986. Having always approached the therapeutic use of essential oils from the "radical" French "Aromatic Medicine" perspective, I have long noted the many incongruous and exaggerated statements regarding essential oil toxicity.

Over these past twelve years, through my involvement with various government and industry bodies, I have specifically focused on this topic of essential oil toxicity as one area of study, given the potential "poisons scheduling" of various essential oils by the Australian National Drugs and Poisons Scheduling Committee.

Why is there such a diversity of opinion regarding essential oil toxicity?

Three reasons appear to me outstanding - that of "philosophical" differences, the lack of knowledge amongst practitioners and authors and the fear of public misuse.

Philosophical Differences

Utilising Dr. Daniel Penoel's concept of the "Aromatic Tryptic"(1), we can characterise "Holistic" Aromatherapy as fundamentally "energetic" in nature. Encouraged by the work of Maugerite Maury in France during the 1930's (2), this approach has become the dominant form of Aromatherapy practiced in English-speaking countries.

Employing relatively low dosages of essential oils (generally 2.5% or less in massage applications), the majority of therapeutic effects noted appear to be primarily of a secondary "energetic" or "terrain" nature, as in the case of acupuncture or homeopathy, for example, as well as working via the olfactory sphere. Actual physical doses of essential are 'ingested' via Aromatherapy massage treatments, but the dosages are quite small, perhaps on average from 50 to 75mg.

"Holistic" Aromatherapy originally developed primarily in the domain of beauty therapy. Practitioner training, even up to the present day, has tended to concentrate more on massage and other application methods, than on an in-depth understanding of essential oils from both the chemical/pharmacological viewpoint and their full history of use in traditional medicine.

M. Maury also stated her own preference to avoid the more "medical" applications of essential oils, including internal use. Such applications, she felt were best left to medical practitioners. (3) Following from M. Maury, the growth of "Holistic" Aromatherapy continued primarily in England by those influenced by her, such as Marceline Arcier and Daniele Ryman.

Developing from the domain of beauty therapy, we can see a particular "dogma" has evolved, one that is "gentle" and oriented from an "energetic" perspective towards both "low-dose" applications and the avoidance of internal and other "high-dose" applications. As such, I suggest that this particular bias has served as the "philosophical base" on which many of the common statements regarding essential oil toxicity are based.

In contrast, we can say the French "Aromatic Medicine" approach that has developed most strongly amongst French medical practitioners (as well as naturopathic and herbal medicine practitioners) since R. M. Gattefosse's work in the 1930's, is more of a "physical" approach.
This "French" approach often utilises comparatively high doses of essential oils both topically and internally, to realise dose-dependent pharmacological effects.

This discipline relies on a greater understanding of the chemical structure and the pharmacological/toxicological effects of essential oils, to suggest safe dosage levels and contra-indications for use. I can therefore suggest that such dosage recommendations represent a more realistic view of the safe uses and potential toxicity of essential oils for all practitioners.

Limited Knowledge

As I have mentioned above, "Holistic" Aromatherapy training has not generally taken into account any in-depth training in either the chemistry or known pharmacology of essential oil compounds. As such, we can notice that many of the dosage recommendations and contraindications mentioned in Aromatherapy literature are based on an incomplete or limited understanding of the issues involved. What can be noted in many publications are statements that are based on the attitude:

"If I do not know about the possible negative effects of an essential oil, then, if any possible negative effect might be noted, I'm going to recommend that people not use such oils at all or else in very tiny amounts".

To err on the side on caution may be considered laudable. However, we can notice that such exaggerated statements has led to a common perception that the therapeutic use of essential oils can be an extremely risky proposition, even amongst those who are purported to be highly qualified practitioners.
It is my premise, that those who would call themselves Aromatherapists should be the most qualified in the actual uses and the potential toxicity of essential oils, as we would expect those with either medical training (with pharmaceutical drugs) or medical herbalists (with herbal preparations) to have with their common prescriptions.

Public Misuse

The vast majority of Aromatherapy books are written for the lay public. In this regard, care is taken to recommend dosages and essential oils that will neither create negative reactions or lawsuits. Hence, dosages are kept extremely low and any essential oil that might be construed to have any possible negative effect, such as during pregnancy, is routinely advised to be best left alone.

If we inspect such books, we also find that these publications, easily accessible to the public, are often used as "textbooks" in Aromatherapy practitioner training. If we observe further, we also find that many publications offered for practitioners and health professionals make many of the same recommendations.

Why is this? I suggest that "Aromatherapy" still needs to go beyond being just a "good feeling", fad therapy. As with the standards that have developed relative to the training and practice of medical herbalism, Aromatherapy demands a level of practitioner training that is comprehensive in it's scope and knowledgeable in all the effects of essential oils - both positive and potentially negative.

Essential Oils and Toxicity Issues

As with most medicinal drugs, whether of a "synthetic" or a "natural" origin, the compounds present in essential oils have the potential to create serious, even fatal toxic effects, if ingested in overly large quantities.

There are numerous cases reported in toxicological literature regarding both serious (non-fatal) and fatal outcomes of essential oil ingestion in both children and adults.

These cases are generally due to accidental ingestion by young children, attempts at creating abortions in past years and the use of essential oils for suicide attempts. There are more rare cases of toxic effects due to overly large doses of specific essential oils being "self-prescribed", "prescribed" to children by parents or prescribed to clients by ill-informed therapists.

Most essential oil compounds have a "non-specific" toxic effect, whereby the absorption of these lipophilic compounds into cellular membranes can eventually lead to disruption of membrane permeability. The primary toxic outcome is that of the disruption of ion channel function in nerve cells, first affecting the heart and central nervous system, leading to cardiac and respiratory depression. To create such effects, however, require huge dosages, in the order of 300mL and beyond.

Certain aromatic compounds, most notably 1,8 cineole (as in many Eucalyptus species), camphor (borneone) (as an isolated compound or as in Rosmarinus officinalis CT camphor and Lavandula latifolia) and methyl salicylate (as a synthetically derived compound or as in Gaultheria procumbens) have specific toxic effects at much lower doses. These compounds make up the bulk of both serious and fatal poisonings in children and adults, due not just to their toxicity, but to the common availability of products containing these compounds and their reputed beneficial properties.

Given the rapid and almost complete absorption of essential oils ingested orally, this route of administration has the highest potential for toxic effects.
First aid measures for ingestion of significant amounts of particularly toxic essential oils (such as more than 2mL of high-cineole Eucalyptus oils in young children) is straightforward: take the child to the nearest hospital emergency room or at least call or a Poisons Information Centre for instructions. The vast majority of accidental essential oil ingestion in children results in few, if any symptoms and resolve safely with no medical intervention.

It is often difficult to determine just how much of an essential oil (or any product) a young child has ingested. If toxic symptoms do begin to develop, gastric lavage, hemodialysis and other supportive medical measures may well be necessary. To attempt to either dilute the stomach contents by giving burnt toast (or activated charcoal), milk or other foods or to try to induce vomiting is not recommended. Either approach, if vomiting occurs, has the potential to allow these volatile compounds to enter the lungs, potentially creating aspiration pneumonia.

"Aromatic Medicine", or the use of essential oils as ingested herbal medicines by trained physicians and complementary therapists, has not been responsible for any severe cases of toxicity. As with any "drug", if an appropriate dose is used (with essential oils, this is often in the range of only 100 to 300 mg per day), toxicity is not an issue.

In the case of the more common practices in Aromatherapy, we are speaking of topical applications - in the form of essential oil preparations used in massage treatments, in baths, etc. or in the form of "low dose" inhalations. Such applications have not created any acute or chronic systemic toxicity - the amounts absorbed into the body and the dosages used are far too low.

'Lethal Dose' Tests and Essential Oils

The most common test of potential human toxicity is that of the "LD50" test or the "median lethal dose". This test is routinely applied to laboratory animals (humans do not usually volunteer) in the testing of compounds used in pharmaceuticals, agricultural chemicals, flavours, fragrances and cosmetics, to name a few.

In this testing procedure, laboratory animal are given measured doses of compounds until approximately half of the test population die. The "median dosages" are then generally given in the ratio of milligrams or grams of test compound per kilogram of bodyweight.

Hence, a LD50 rating of 1.0 represents that 50% of the test animals died on a dosage of 1 gram per kilogram of body weight. If we consider ourselves to be large rodents, this would translate to 60 grams of a particular compound would be the likely lethal dose to an adult weighing 60 kilograms.

We should consider (outside of ethical considerations - but no effective substitute has yet to be found) that such tests generally are based on either acute oral (by mouth) or injected lethal doses. This means that the LD50 dose represents the median toxic dose taken all at one time, either by ingesting or by direct injection of the test compound.

There are also LDLo values used. This figure represents the lowest dose of a given compound that has caused death. LDLo values are often assigned when the number of test animals used is small. As well, since there have been human fatalities caused by the ingestion of various essential oils or isolated aromatic compounds, we can sometimes approximate a LDLo value for humans, if a reasonable idea of how much of a particular compound was ingested is known.

Chronic (long term) toxic doses and dermal (high-dose topical applications) have also been studied with laboratory animals. Toxic chronic doses are generally less than the corresponding acute dose. Long term studies are often carried out for 90 days of exposure, to detail whether or not a particular test substance has a cumulative toxicity. If it does have a cumulative toxicity, this will be seen by a lower daily dose being required to demonstrate toxic effects or death, than the acute LD50 dose required in one application. There are few results available for chronic toxicity of essential oils, but studies from 1959 suggest a dose range from 5 to 10 times less than an acute toxic dose.

Dermal studies performed on rabbits have produced conflicting results that do not appear to be terribly relevant to human exposure. One reason appears to be the greater permeability of rabbit skin to the passage of different compounds, as compared to human skin. There are some reports of lasting skin damage in humans, such as due to the spilling of undiluted Clove oil onto the skin and to the chronic application of a concentrated methyl salicylate preparation (covered with a heating pad). However, I have seen no reports to date of any fatalities via topical application of essential oils; fatal outcomes are due to excessive ingestion of specific essential oils.

The Mistakes in Applying LD50 Values to Aromatherapy Applications

In terms of the most common uses of essential oils in Aromatherapy, it is the acute LD50 dose that is most relevant in this consideration. This is considering therapists using essential oils on their clients and preparing formulations for the client to use outside of visits. Animal LD50 values can be a useful guide to potential essential oil toxicity when we are considering the acute toxicity of essential oils, such as Wintergreen (mostly methyl salicylate) or Eucalyptus species (those with a high 1,8 cineole content).

For example, an essential oil, such as Thuja (Thuja occidentalis), with an animal LD50 rating of 0.83, would translate to approximately 50 grams being a median lethal dose for an adult weighing 60 kilograms. Of course, this would be a huge dose and severe toxic effects would still be seen (and have been) at lower doses like 10 grams. The point to make is again, the values we are considering here are based on acute oral toxicity, that is, the lethal dose that would be ingested all at one time.

There are two areas where mistakes relative to Aromatherapy "toxicity statements" are made: Dosages: Essential oil dosages, such as applied in preparations for massage, in baths or for inhalations (or simply to fragrance an environment) are generally of a minute fraction of the acute toxic dose (and of the chronic dose as well).
Let us take Wintergreen oil as an example. The acute oral rat LD50 is 1.2. In humans, however, methyl salicylate does appear to be more toxic. Given the numbers of fatalities in years past, with the amount ingested being known in a number of cases, we can estimate a human LD50 of 0.3. For a 60kg adult, this would translate to the ingestion of about 18 grams. (11)
Now, let us say that we want to apply a 2.5% dilution of Wintergreen oil to our sore lower back. We then apply 1 mL of this preparation…

1mL x 2.5% = approximately 0.025 grams of methyl salicylate.

0.025 gms ÷ 18 gms (LD50 dose) = 0.00139 or 0.139%.

Hence, the applied dose is only 0.139% of the median lethal dose.
This is more than 700 times less. Of course, if we increase the amount applied of the 2.5% formula, we increase the dosage received. Hence, if we applied 10mL of the formula all at once, the dose would now be 0.25 grams or 250 milligrams. Putting this into perspective, even if the methyl salicylate was totally absorbed, this dose would represent the same amount of salicylate compounds as found in one tablet of aspirin

Wintergreen and Sweet Birch essential oils are routinely mentioned as oils to avoid in Aromatherapy, even for trained practitioners. Members of the International Federation of Aromatherapists take a "vow" not to use Wintergreen essential oil. (12) Yet, we have the strange contradiction of many methyl salicylate-containing topical products (containing from 10% to 30% methyl salicylate) being readily available to the untrained public - with few negative side-effects reported (methyl salicylate, even used topically, is reasonably contraindicated in people taking the anti-coagulant drug, warfarin). (13)
Even with this relatively toxic compound (as I would suggest that any essential oil with an LD50 of less than 1.0 is), an effective anti-inflammatory preparation can be used with no acute toxic effects.

Method of Application

Not only should we consider the dosage given, but also account for how the essential oil is applied.
We can say that when an essential oil is orally ingested, it is fully absorbed into the portal blood circulation. However, other types of applications do not represent the same level of absorption and dosage. The following chart details the potential toxicity of each method of application. This accounts for both the amount of absorption as well as the amount of the typical doses given.

Mode of application - Comments

In this light, we can then understand why the relatively toxic essential oil of Pennyroyal, could be used as a safe addition as a mucolytic used in an inhalation. With inhalations, absorption is quite high, but the typical dose is small. With topical applications, we cannot assume full absorption of applied essential oils. If we do not occlude (or cover) the site of application, as is generally the case with topical Aromatherapy applications, the dose is significantly lessened by evaporation.

One American study found that 75% of an applied dose of various fragrance compounds were absorbed through human skin when the skin was covered after application. When the skin was left uncovered, the total amount absorbed dropped to only 4.0%. (14) It is clear that topically applied essential oils will penetrate the epidermis of the skin. However, it is an area that still requires further research to understand how a variety of different factors (such as the type of essential oil compounds, the excipient or "carrier" base used, temperature, age of the skin and on) affect the amount absorbed through the skin.

Available studies suggest a wide range of absorption amounts. d-limonene, the major constituent of most citrus oils, was demonstrated to only have an absorption rate of 2.0% when applied to human tissue samples. (15) A 2.0% dilution of True Lavender (Lavandula angustifolia) oil applied to the abdomen of a volunteer, showed that approximately 10% of the Lavender oil was absorbed into the general blood circulation, showing a relatively rapid absorption rate that peaked 20 minutes after application. After 90 minutes, both linalool and linalyl acetate (the compounds tested for) had dropped almost to zero, showing almost complete biotransformation and excretion. (16)

Some compounds found in essential oils have relatively high absorption rates, even on unoccluded skin, such as coumarin (46% absorption ) and cinnamic aldehyde (24%). (17) Lastly, a study testing percutaneous absorption with rhesus monkeys, tested three compounds, benzyl alcohol, benzyl acetate and benzyl benzoate (all naturally occurring in Ylang Ylang essential oil, for example). When applied in a moisturising lotion base, with the skin uncovered, the total absorption rate varied from approximately 20% for benzyl acetate, up to 70% for benzyl benzoate. (18) The assumption here is that essential oils and like compounds are more easily absorbed through the hair follicles than just the stratum corneum or "horny layer" of the skin. Hence, it can be theorised that monkey skin, covered in hair follicles, would more efficiently absorb essential oils.
Taking the available research into account it would be fair and conservative to state the following when figuring the absorbed dose of an essential oil applied to unbroken skin in some form of an excipient or "carrier" (vegetable oil, cream, gel, etc.) and left uncovered:

Only up to 50% of a topically applied dose is absorbed.

Hence, in the Wintergreen oil example given above, instead of the low amount of 0.025 grams being absorbed, the amount can be figured at half that value, or 0.0125 grams - less than 1400 times the median toxic oral dose. This covers both the typical applications applied in Aromatherapy treatments, as with massage, as well as topical OTC preparations, such as methyl salicylate containing "liniment" products. However, in the case of broken skin, where the stratum corneum is compromised or not present (as in wounds, burns and various forms of dermatitis), it would be more prudent to figure a 100% absorption of applied essential oils. (19)

"It's all dose-related"

Therefore, we can look at a number of the essential oils mentioned in Aromatherapy books "never to be used in therapy", such as Hyssop, Pennyroyal, Tansy, Thuja, Wintergreen and Wormwood for example. (16) We can understand, however, that such essential oils can be used safely, if one respects the dose given and the method of application used. There is certainly an argument against having essential oils with a high, specific toxicity easily available to the public. However, we are considering here the discriminative use of essential oils by therapists that should be trained in the details of appropriate dosages and applications.

Essential Oils and Cancer

Also found in many books are statements regarding the non-use of essential oils containing compounds that have been shown to promote carcinogenesis in animal studies and various cell assays. I have written a separate paper on this topic (Toxicity Myths - Essential Oils and Cancer) and will present a summary here.

Two compounds in particular, A- and b-asarone as found in various cultivars of Acorus calamus and in Guatteria gaumeri and safrole (the major constituent of Sassafras albidum and Oceotea pretiosa essential oils), have already been effectively banned (only tiny dose levels are permitted in foods) by various governmental 'food and drug' agencies around the world, due to the demonstrated carcinogenic effects in laboratory animals. Hence, these are not essential oils would ever be widely used by therapists nor will OTC products for therapeutic use be made.

However, using Calamus as an example: Dr. Rudolph Weiss, a respected German physician who taught herbal medicine up until the 1980's, states in his book, Herbal Medicine regarding Calamus, "… At the same time it should be noted that Calamus has been widely used since antiquity and is still much used today, and there have been no reports of cancer developing after it was taken. It is important not to attach excessive importance to such experimental studies. …Where Calamus is concerned, there clearly is no need for concern if the drug is used for medicinal purposes and for a limited period only." (20)

A- and b-asarone have been found to be carcinogenic in rodents at elevated levels. As is the common case, we find that large doses are fed to lab animals to elicit hepatic cancers. Dosages as high as 50mg/kg per day (about 3 grams of Calamus oil for an average adult) in rats showed no toxicity or genotoxicity. 25 Dosages as high as 5mg/kg of a-asarone showed no embryotoxicity or teratogenecity (birth defects) in the embryos of pregnant mice, and doses as high as 20mg/kg, did not alter either sperm count in male mice nor create germinal mutations in male or female mice. (21, 22)

These are much larger doses than would be had by therapeutic ingestion of Calamus herbal preparations or topical application of Calamus essential oil. At the same time, asarone and the herbs containing it, are being studied for a strong cholesterol-lowering effect, as well as a diverse range of other effects - tranquillising, sedative, anti-ulcer, anti-spasmodic, anti-carcinogenic, for the treatment of gallstones and more. Calamus has a sterling reputation as a powerful stomach tonic, being an excellent appetite stimulant, as in the case of treating anorexia. (23, 24, 25, 26, 27, 28) One can reasonably suggest that the effective 'banning' of a herb such as Calamus, is indeed a case of 'overkill' by regulatory agencies.

Two other compounds found in essential oils, estragole (or methylchavicol) and methyleugenol have been demonstrated to have weak genotoxic and carcinogenic potential. At this time, no governmental body has taken any step to limit or ban these compounds in foodstuffs or consumer products, given the low potential for harm. Nevertheless, this has led to various Aromatherapy authors giving their opinion that therefore the essential oils of 'Exotic' Basil and Tarragon (rich in estragole) and Melaleuca bracteata (rich in methyleugenol) should not be used at all in Aromatherapy practice. (29, 30)

--From Toxicity Myths - Essential Oils and Cancer