1.
A Critique of "Possible Health & Safety Problems in the Use
of Novel Plant Essential Oils and Extracts in Aromatherapy" The Journal
of the Royal Society for the Promotion of Health 1999, 119(4), 240-243. By
Tony Burfield and Janina Sorensen.1.
A Critique of "Possible Health & Safety Problems in the Use
of Novel Plant Essential Oils and Extracts in Aromatherapy" The Journal
of the Royal Society for the Promotion of Health 1999, 119(4), 240-243. By
Tony Burfield and Janina Sorensen.Introduction
Since its publication, the above-cited article has received a measure
of criticism on Internet newsgroups and at a meeting of a Professional
Aromatherapy organisation. It was felt that the article was scientifically
inaccurate and misleading in its’ content and conclusions, as well as
presenting a prejudiced picture of mainstream aromatherapy. In several
instances, citations have been found to be incorrect. Information, as will be
seen, can easily become distorted by serial recitement without consulting the
original papers. Further, a valid scientific report should always include a
systematic and critical review of the original data cited as references. The
above named authors have therefore agreed to critically review the article
content.
The title of the article includes the word "Aromatherapy" (AT), but the first paragraph gives a definition more suited to Aromachology than Aromatherapy. Lis-Balchin defines Aromatherapy as "treatment using odours, the inhalation of which can have beneficial effects on their clients through the action on the limbic system in the brain and the usual hormonal pathways to their target sites". Aromatherapy and Aromachology are, however, two different principles. The modern definition of aromatherapy is the investigation of the physiological and pharmacological activities of natural aromatic substances, whereas Aromachology is defined as the study of relations between psychology and fragrances (Buchbauer,1998). Going back further, the term Aroma-chology dates back to 1982 and was describes as "dedicated to the study of the inter-relationship between psychology and fragrance technology to elicit a variety of specific feelings and emotions – relaxation, exhilaration, sensuality, happiness and well-being – through odours via stimulation of olfactory pathways in the limbic system (Fragrance Research Fund, Jellinek, 1995). This definition is clearly more suited to the definition above by Lis-Balchin.
The same confusion is revisited later on page 243 "aromatherapy should be practiced using essential oils from plants whenever possible, but otherwise composed of synthetic components if necessary". Clearly the title of the paper should have been modified to Aromachology, because no Aromatherapist would knowingly use non-natural ingredients: it may invalidate their contract for membership of professional Aromatherapy bodies and may also compromise their professional insurance cover. Most organisations such as NAHA (National Association of Holistic Aroma therapists) have a suitable definition of an essential oil which requires an (implied) 100% derivation from the named source. The aversion of the Aromatherapy profession to all that is synthetic, chemical, unhealthy and mechanistic is discussed by Vickers,1996.
In the second paragraph, Lis-Balchin goes on to talk about Aromatherapists desire to use pure plant essences, and then comments on commercial essential oils which are adulterated with components from plants (this is taken to mean adulteration with cheaper natural materials) and synthetics. Whilst it is true that adulteration is common in the essential oil industry, many Aromatherapists buy oils from small distillers who supply AT grade oils. The buyers personally know many of these distillers, and there is a thriving distiller-client relationship which suits the small-scale "cottage industries" feel of the profession. Larger perfume-based or aroma-chemical based companies are treated with some suspicion, as they are considered to have materials close at hand to adulterate oils, and so their "commercial essential oils" sometimes have a bad connotation.
Toxicity data for several botanicals are listed in a table entitled "Novel and potentially toxic essential oils, phytols or infused oils used in aromatherapy" on page 242. The listed plants and their stated toxicity data are Arnica montana (poisonous, internal intake), Calendula officinalis (possibly abortogenic), Centella asiatica (photosensitiser, pruritic), Symphytum officinale (hepatotoxic, carcinogenic), Harpagophytum procumbens (hypoglycaemic, oxytocic), Croton perdicipes (dermal sensitiser), Echinacea purpurea (dermatitic, hepatotoxic), Trigonella foenum graecum (hypoglycaemic, oxytocic), Kunzea ericoides (toxicity not studied, therefore caution), Tilia spp. (cardiotoxic), Leptospermum scoparium (toxicity not studied, therefore caution), Filipendula ulmaria (gastric irritant, bronchospastic), Schinus molle (dermal irritant, sensitiser), Minthostachys spicata (toxicity not studied, see Mentha pulegium), Satureja boliviana (irritant, hepatoxic), Mentha pulegium (fatal in high doses), Myroxylon pereira (dermal sensitisation), Hypericum perforatum (photodermatitis) and Artemisia absinthum (neurotoxin, renal failure).
In the table's title as well as in the text, these botanicals are referred to as essential or herbal oils. Many of these herbs are not used in mainstream Aromatherapy, but find their main application in herbalism, in flavourings, or as insect repellents. Even finding evidence of fringe use of these oils in Aromatherapy may be difficult: Centella essential oil (from Centella asiatica) is available but the annual production from suppliers figures is thought to be only a few hundred grams per annum. Most of the listed herbs are used as alcoholic or aqueous extracts, not infused oils, and their (often questionable) toxicology data refer to internal application of these extracts, which seldom are relevant to their external use. Neither their activity profile, nor their toxicology data can be extrapolated to their use as herbal oils in aromatherapy or massage. Many of the listed phytopharmaceuticals contain no essential oil. Regarding toxicity, only Arnica montana is listed correctly in the table as "poisonous, internal intake". This leaves the reader to wrongly assume that the remaining toxicity data refers to external (topical) use, as mentioned in the title of the table.
Returning to Centella asiatica, - it seems that the use of an infused oil has well documented activity, which fits well in the scope of medical herbalism (one hesitates to say Aromatherapy). Looking at this further, the listing of Centella as 'pruritic' in the article is described thus in the cited reference (Newall C.A. et al 1996): 'ingestion of hydrocotyle is stated to have produced pruritis over the whole body'. The entry 'phototoxicity', referenced to the BHP (British Herbal Pharmacopoeia), is mentioned in that book as one study from 1958 (!) entitled : "Centella may produce photosensitation in tropical areas" (Chopra et al. 1958).
Lis-Balchin mentions 'hydrocotyle (Centella asiatica) has been used to treat leprosy, which does not make it necessarily ideal for normal Aromatherapy'. In fact, documented clinical and animal data support the herbal use of Centella as a dermatological agent, successful in the treatment of psoriasis, cutaneous ulcers and warrants further research into its use in wound management. The triterpenoids (non-volatile compounds) are regarded as the active principles, and are reported to have a stimulating effect on the epidermis and to promote keratinisation applied topically. Traditional use includes topical application for several skin disorders. Interesting, one of the triterpenoid constituents, asiaticoside, was reported to improve the general ability and behavioural pattern and increased vital capacity of 30 retarded children (Newall C.A. et al 1996). This profile makes an infused oil of centella an ideal candidate for medicinal herbal treatments, and does not exclude its potential use in aromatherapy per se.
Devil's claw (Harpagophytum procumbens) is used internally in the treatment of degenerative disorders of the locomotor system (Blumenthal et al. 1998), even though the article indication might suggest external application. There is no record of herbal oil from this plant. The active principles are believed to be iridoid glycosides (harpagoside), and extraction is achieved by polar solvents, alcohols of various concentrations (Bruneton J. 1995). Extracts, taken orally, showed considerable effectiveness in the treatment of chronic back pain in a randomized, placebo controlled double-blind study, and there was no evidence for related side effects. (Chrubasik S. et al. 1999).
Under the heading 'toxicity', hypoglycaemic activity is listed for H. procumbens. Apart from the fact that hypoglycaemic activity is not documented for this plant (it is suspected to have hyperglycaemic activity in fact), in the text, the author states: 'Devil's claw contains iridoids, which may contribute to a hypoglycaemic effect, therefor diabetics are excluded'. Hypoglycaemic drugs are not contraindicated in diabetes; on the contrary, they are used to treat diabetes mellitus. This citation, however, is based on the original mistake in the authors' reference (Newall C.A. et al 1996 p 99): 'Devil's claw is stated to be contraindicated in diabetes (hypoglycaemic action)'. Furthermore, Newall C.A. et al. 1996 mention that 'no scientific data were located to support this statement'. Why should such a statement be made at all, when no reference can be given and nor scientific data located, and secondly, on what grounds should M. Lis-Balchin cite it in the article in reference to AT use?
However, the hypoglycaemic activity of Fenugreek (Trigonella foenum-graecum) is well documented in the scientific literature. Again, reports refer to the aqueous and alcoholic extract (Abdel-Barry J.A. et al 1997); there is no record of an infused oil. The active principle of the seeds is reported to be 4-hydroxyisoleucine, a polar, non-charged original amino acid, which is able to stimulate insulin release and is a potential antidiabetic agent of interest (Broca Ch. et al. 2000). Debitterized fenugreek powder did not produce any significant acute and cumulative toxicity in rats and mice who failed to show any signs of toxicity or mortality up to a maximum dosage of 2 and 5g/Kg bodyweight respectively (Muralidhara et al. 1999). In phytomedicine, hot decoctions of the seeds are used locally as a poultice to draw abscesses (Weiss, 1991). It is difficult to conceive that Fenugreek would be used at all in Aromatherapy practice: its overwhelming curry-like odour is persistent on the skin and clings to clothing. The social consequences of its use therefore have not lead to any notable deployment in mainstream Aromatherapy.
Meadowsweet herb (Filipendula ulmaria) contains a very small amount of essential oil (0.05%), composed mainly of salicaldehyde (up to 75%) methyl salicylate, ethyl salicylate, methoxybenzaldehyde etc (Lindeman A. et al 1982. Other components include anisaldehyde, benzaldehyde, and b -phenylethyl alcohol). According to a principle essential oil supplier to the aromatherapy trade, essential oil of Filipendula ulmaria in a vegetable oil base is commercially available, but annual sales volumes are minute. An infused herbal oil might be available, however, in herbalism the decoction and extract which contain chiefly flavonol glycosides are traditionally used for fever, flu-like symptoms and to enhance renal and digestive elimination functions (Bruneton as above p. 221-2). Local administration of the decoction has shown interesting results in cervical dysplasia and cancer treatment (Peresun'ko A.P. et al 1993).The mode of administration is the comminuted herb or other galenic preparations for infusions, side effects are not reported. However, because of the salicylate content, the drug should not be used internally where a salicylate sensitivity exists (Blumenthal et al, as above, p.169). It is difficult as well to correlate the listing under 'gastric irritant' to the external application of an infused oil.
Progressing further through the listed oils, It is very difficult to see what connection many of these entries have to do with Aromatherapy. The roots of Echinacea purpurea can be distilled to obtain an essential oil high in sesquiterpenes, but the distilled oil is not commercially available to the authors' knowledge. In the whole plant, the listed un-documented hepatotoxicity is an assumption based on the fact that the plant contains a small amount of pyrrolizidine alkaloids. However, these are unlikely to cause any liver damage, since they are lacking the 1,2 unsaturated necine ring system required for hepatotoxicity (Wichtl, 1994). Its traditional external use in herbalism as a wound-healing agent, however, seems to be based on immune-stimulatory activity, ascribed to the hydrophilic polysaccharides. Topical application is in the form of an aqueous extract. Even though Nevall et al are cited as reference for the 'hepatoxic action' of Echinacea listed by Lis-Balchin, such a statement is non existent in the cited reference, as the authors state clearly that the alkaloids present in Echinacea possess a 'saturated pyrrolizidine nucleus and are not thought to be toxic' (Newall C. A. et al., 1996 p. 103).
There is further confusion regarding the listing for Peru Balsam (Myroxylon pereira). Since the title of the table includes the words "Novel and potentially toxic essential oils…." one should assume that the entry refers to Peru Balsam oil, which is very high vacuum dry distilled or molecularly distilled from Peru Balsam. This results in a skin safe oil, which is approved IFRA (International Fragrance Research Association) for perfumery use. It may be that the author is referring to Peru Balsam itself, which does not now have much use in perfumery, following its restriction by IFRA in Dec. 1995 because of its 'sensitising potential'.
More relevant are entries for Manuka (Leptospermum scoparium) and Kanuka (Kunzea ericoides) oils. The author states that these are variable oils, which is true. However many individual companies concerned with selling these oils from source sell specific chemically defined chemotypes, and one major supplier accompanies sales with full GC and safety data. The concern may be that this information is not in the public domain but only available to purchasers of the essential oils. We understand that the view of the company is that they have paid out large sums to obtain the data, and are not about to hand it over to competing companies. Whilst the authors take no stand on the issues involved here, the fact is that safety data is therefore potentially available to prospective purchasers. Toxicity studies using normal human epidermal keratinocytes (NHEK) have also been carried out for a specific chemotype of Manuka oil, and a reference to this has been posted by the commercial company "Essentially Oils" on their website www.essentiallyoils.com .
Ravensara is noted as a novel oil in Aromatherapy according to Lis-Balchin …"introduced by a French aromatherapist, in an unrefereed book (Franchomme P. and Penoel D. "Aromatherapie Exactement" 1990). Michel Vanhove provided the following information (Internet newsgroup communication 7.5.00): "….It is not true to state that Ravensara has been introduced by Pierre Franchomme or Dr. Penoel"…, and he goes on to show that in fact it has long history of use. Its first citation was in 1913 by Ferraud & Bonnafous and was mentioned later in 1929 by E. Gildmeister & Fr. Hoffman (also predating the coining of the word "Aromatherapy" by R.M. Gattefosse in 1937: see Belaiche P. (1991)). Further, introduction of Ravensara oil into French aromatherapy practice was well before the publication of Aromatherapie Exactement in 1990. Thus, Ravensara aromatica is mentioned by several French Aromatherapy authors: Viaud 1983, Balz 1986, and Roulier 1990. Vanhove further comments that he believes similar prejudices against the oils Manuka, Kanuka and Lemon Scented Myrtle may result from the fact that there were no English translations available of French Aromatherapy source books.
The entry regarding comfrey (Symphytum officinale) as hepatotoxic and carcinogenic (because of their pyrrolizidine alkaloids with a 1,2 unsaturated necine ring system!) as well refers to the internal use of plant preparations. This detail seems here to have caught the authors' attention, since in the text a mother is mentioned who 'drank some comfrey tea'. Comfrey is indicated for external application only, and when used as such, side effects are not reported. However, duration of administration is recommended for not more than 4-6 weeks per year (Blumenthal, as above, p.115).
Marigold (Calendula officinalis), listed as 'possibly abortogenic' by the author with reference to Newall C.A. et al., is a distortion from the original. Newall C.A. et al. 1996 cited no side effects, toxicity or contraindications for this herb, but mentioned that an 'uterotonic effect (in-vitro rabbit and guinea pig preparations) has been reported'. Their reference again is a single study (Shipochliev T. 1981). on extracts of medicinal plants enhancing uterine tonus. Locating this original reference, we learn that water extracts from several plants were used in a series of experiments.
In a final concentration of 1 to 2 mg crude drug per 1 cm3 isolated uterine horn tissue, chamomile extract ranked first, marigold next as anti-inflammatory and a good uterotonic enhancer. It seems very far-fetched to extrapolate the statement 'abortogenic' from this single study.
Tilia spp., listed as 'cardiotoxic' in the article is referenced to Tyler (1993) and referred to by Newall et al (1996) p. 181 as well. However, these latter authors state 'excessive use of lime flower tea may result in cardiac toxicity. However, the author does not include the rationale for this statement'. Moving on, 'Satureya boliviana', mentioned as 'irritant and hepatotoxic' in the listing and referenced to Newall et al. (1996) but could not be found in the cited book at all.
For the listed pennyroyal oil (Mentha pulegium), the quote 'fatal in high doses' is stated with reference to Tisserand and Balacs (1995). Again, this refers to internal use. The authors document 3 fatal cases due to ingestion of the essential oil, in two of these cases about 30ml essential oil was ingested - another case which is a "far cry" from typical Aromatherapy use.
It is interesting to note that the author correctly mentions that "allergic contact dermatitis is a growing concern even for (essential oils)", but then states, that "in comparison to CO2 extracts "essential oils, obtained by steam distillation, do not contain any of the more water soluble plant components, including alkaloids, chlorophylls, carotenoids etc, which could cause dermatitic effects". Dermatitis effects, of course, can be caused by many essential oil components as well. Finally, since CO2 extracts are not essential oils - essential oils are defined as the volatile products of steam distillation, or of mechanical pressing of the rinds of citrus fruits - CO2 extracts should therefore not be used in true Aromatherapy.
Additionally, because of their relatively recent appearance on the market, these extracts, unlike many essential oils, have neither decades of documented use nor yet have undergone formal safety testing for skin application or for use in flavorings, and should be used only for environmental fragrancing as yet. Aromatherapists should be aware of these issues through newsgroups, the advice of their professional bodies, and from the statements of specialists in essential oil safety.
As can be seen from all these cases, the toxicity data referred to throughout the article has little, if anything to do with the external use as essential or infused herbal oils in Aromatherapy. Neither is the presented data well researched and evaluated. A recent critical article is relevant here. It addresses the new field of herbal research, which is stuck in an intellectual twilight zone, and results in a rapid accumulation of published research data on herbs, much of it largely meaningless. After evaluating several papers, the author argues that a scientific publication should be precise, otherwise it is not science, and that sloppy research only perpetuates the continued dissemination of misinformation or meaningless information. This results in wasted resources in transporting meaningless data back and forth and also in correcting misinformation, or debunking meaningless research. The author proposes an establishment of a uniform set of criteria and guidelines for herbal research for authors, reviewers and journal editors to avoid contaminated information (Leung, 2000)
We are in full agreement with the author that the possible toxicity of essential oils used in Aromatherapy applications, as far as external uses are concerned, still needs more valid investigations to be carried out. However, valid research should be conducted with essential or herbal oils used in AT, in realistic concentrations and by external application. As was clearly demonstrated above, available safety data on botanicals cannot be extrapolated to their use in mainstream aromatherapy.
The conclusion of M. Lis-Balchin that "this paper has given ample evidence…that.... (the use of) .. various novel essential oils and various herbal oils…is potentially dangerous" cannot be founded on the scientific data presented. The statement that "herbal remedies are obviously not safer than conventional drugs" lacks any refereed foundation in this article, and in general for that matter. A recent review of data showed, that fatal conventional drug reactions might be the fourth-leading cause of death in the US, after heart disease, cancer and stroke. These findings are especially discouraging because they do not include human error interventions in the form of incorrect administered drug, wrong dose nor reactions occurring outside of hospitals, as they are based on the incidence of adverse reactions in hospitalised patients only. If those were to be included, it is entirely possible that adverse effects would be the third leading cause of death:; (Lazarou J. et al. (1998), Kohn L.T et al, 1999).
Again, in full agreement with the author, allergic reactions, irritation and sensitisation potentials have to be investigated for the oils used in Aromatherapy. Certain data is available from RIFM, where the essential oils are used in the perfumery industry, and from other sources. However, it has to be kept in mind, that allergic reactions as well as sensitisation reactions can be individual responses as well. This finds a parallel in our current inability to determine the allergic or sensitising potential of common foods in particular instances. This paints a picture where allergic contact dermatitis is an ever-growing phenomenon, and far from restricted to essential oils or other natural substances. To get into the possible causes of this increasing intolerance among the Western population exceeds the scope of this paper.
Reviewing the available literature on Aromatherapy, it springs to attention that there is a primary emphasis on the purity and authenticity of the essential oils used. The statement by M. Lis-Balchin that "there is a firm conviction among aromatherapists that all essential oils are pure products, healthy and harmless", has no foundation in the literature. Quite on the contrary, quality issues are widely discussed in articles, at conferences and in newsgroups, and criteria for obtaining genuine, unadulterated oils are frequently discussed. Many textbooks on AT devote some chapters to quality issues and include a definition of genuine oils, as well as to methods of production and botanical origins.
We hope in this short account some errors and misconceptions in the original paper have been righted.
References-
Abdel-Barry J.A. et al (1997), "Hypoglycaemic and anti-hyperglycemic effects of Trigonella foenum-graecum leaf in normal and alloxan induced diabetic rats." J Ethnopharmacol 58: 3149-55
Balz, Rodolphe 1986 "Les huilles essentielles et comment les utiliser" Balz, Rudolph.
Belaiche P. "Historique et Prescription en Aromathérapie" in Encyclopédie des Medicines naturelles publiée sous la direction du Professeur P. Cornillot. Editions Terchniques, Paris C-19-1991,
Blumenthal et al. The Complete German Commission E Monographs (1998) American Botanical Council, p.120
Blumenthal et al., as above, p.169
Blumenthal et al, as above, p. 115
Broca Ch. et al. (2000), Hydroxyisoleucine: effects of synthetic and natural analogues on insulin secretion. Eu J Pharmacol 390 (3): 339-345
Bruneton J. Pharmacognosy, Phytochemistry, Medicinal Plants (1995). Lavoisier, France p. 479
Bruneton J, as above, p.221-2
Broca Ch. et al. (2000), Hydroxyisoleucine: effects of synthetic and natural analogues on insulin secretion. Eu J Pharmacol 390 (3): 339-345
Buchbauer G. (1998). "Aromatherapie, naturwissenschaftlich betrachtet." Z. Phytother 19: 209-212).
Chopra R.N. et al. (1958). "Centella may produce photosensitation in tropical areas. in: Indigenous Drugs of India". U.N. Dhur & Sons Ltd, Calcutta p. 554
Chrubasik S. et al. (1999) "Effectiveness of Harpagophytum Extract WS 1531 in the treatment of exacerbation of low back pain: a randomized, placebo controlled, double-blind study." Eur Anaestet 16 (2): 118-29
Ferraud & Bonnafous (1913) Bull Sci Pharmacol 20, 403.
Fragrance Research Fund (1992) "Living with your sense of Smell, New York: p21.
Kohn L.T, Corrigan J.M. and Donaldson M.S. (eds). To Err is Human: building a safer health system. Washington, D.C., National Academy Press, 1999, pp 23, 27-29
Lazarou J. et al. (1998). "Incidence of adverse drug reactions in hospitalised patients: a meta analysis of prospective studies." JAMA 279: 1200- 05
Gattefosse R.M. (1937) "Aromatherapie" Giradot, Paris.
Gildemeister E., &. Hoffman Fr. "Die Ätherischen Öle" 1928 Schimmel & Companie, Jellinek, J.Stepan "Aromachology-A Status Review Part 1, Dragoco Report 1/1995 p5.
Leung A (2000). "Scientific Studies and Reports in the Herbal Literature: What are we studying and reporting?" Herbalgram 48:63-4
Lindeman A. et al (1982) Lebens Wiss. Und Technol. 15(5) 286.
Muralidhara et al. (1999) "Acute and subchronic toxicity assessment of debitterized fenugreek powder in the mouse and rat." Food Chem Toxicol 37: 8831-8
Newall C.A., Anderson L. and Phillipson J.D. (1996). Herbal Medicines, The Pharmaceutical Press, London
Peresun'ko A.P. et al. (1993) "Clinico-experimental study of using plant preparations from the flowers of Filipendula ulmaria (L.) Maxim for the treatment of precancerous changes and prevention of uterine cervical cancer." Vopr Onkol 39: 291-5
Roulier Guy (1990) Les huilles essentielles pour votre santé, Edtions Dangles
Shipochliev T. (1981). "Uterotonic action of extracts from a group of medicinal plants." Vet Med Nauki; 18: 494-8
Tisserand R. and Balacs T. (1995). Essential Oil Safety. Churchill Livingstone, p. 159
Weiss R.F. Lehrbuch der Phytotherapie (7th Ed. 1991). Hippokrates Verlag, Stuttgart, Germany
Wichtl M. (1994) "Herbal Drugs and Phytopharmaceuticals" (1994) (and references therein). A handbook for practise on a scientific basis. CRC Press Boca Raton, Ann Harbour, London, Tokyo, p. 183
Viaud H. (1983) Huiles essentielles, hydrolats, Viaud Henry, distillateur pub. Editions Presence, Sisteron
Vickers A. (1996) Massage & Aromatherapy – a Guide for Health Professionals pub. Staney Thornes UK p44.
BACK NEXT BACK TO ARTICLES INDEX
Copyright © 2000 by Tony Burfield. All Rights Reserved.