Summary and conclusions

Kortlægning og sundhedsmæssig vurdering af kemiske stoffer i æteriske olier og duftolier

As part of the Danish Environmental Protection Agency's programme for surveying chemical substances in consumer products, a survey of the use of essential oils and fragrance oils in Denmark was undertaken in order to examine possible health effects by inhalation of the chemical substances released from the oils.

The extension of the use of essential oils and fragrance oils

A large number of essential oils and fragrance oils are marketed for use in different kinds of aromatherapy. The focus of this investigation has been on the use of essential oils and fragrance oils for dispersion of fragrances in the home by use of candle diffusers and electrical fan diffusers. The use of candle diffusers is the most widely applied method for dispersion of these types of fragrances in the home and these diffusers are sold by many health food shops, materialists, hardware dealers, stores for life style and on the Internet.

Data on chemical substances were received from producers who submitted Material Safety Data Sheets and supplementary data sheets with information on the content of 26 substances on EU's list of fragrance allergens.

The survey identified more that 50 essential oils and 40 fragrance oils specifically stated as useful for this application, but several hundred different essential oils are marketed, which in principle may be used by the non-professional user. The total sale in Denmark in 2006 is estimated to be in the order of 300,000-700,000 flasks of essential oils and 100,000-300,000 flasks of fragrance oils. The flask size can be 1½, 5, 10 and 100 ml, respectively, with 10 ml as the most common. The total consumption is increasing.

By proper aromatherapy the oils are used specifically to obtain a particular therapeutic effect, but many users primarily use the oils to disperse a pleasant fragrance in the room in order to obtain a particular atmosphere. The possible beneficial effects of aromatherapy have not been in investigated.

Content of fragrance allergens
Data on the content of the 26 fragrance skinallergens on EU's list in 90 products were collected. Practically all of the essential oils and all fragrance oils contain one or more of the substances on the list, and in many products substances from the list constitute more than half of the product. Skin sensitizing is not covered by this investigation, but it is worth mentioning that the oils in some types of aromatherapy is in contact with the skin also; however, always heavily diluted.

Constituents of essential oils
Essential oils are natural oils extracted from plants, and each oil typically contains more than 100 different chemical substances. On the basis of Material Safety Data Sheets (MSDS) for the essential oils identified, a list of 52 constituents was compiled. These are either included in the list of dangerous substances (6 substances) or classified as dangerous by the producers' self classification.

By the producers' self classification none of the oils have been assigned the sentences R23 (Toxic by inhalation) or R42 (May cause sensitization by inhalation). The most frequent R-phrases for the oils are R38 (Irritating to skin), R43 (May cause sensitization by skin contact) and R65 (Harmful: may cause lung damage if swallowed). Tea tree oil and wintergreen oil have been assigned the sentence R22 (Harmful if swallowed). One oil (nutmeg oil) has been assigned the sentence R45 (May cause cancer) due to the presence of 5-allyl-1,3-benzodioxol, whereas. Products labelled R45 are forbidden to sell to the generel public in Denmark. The Chemical Inspection Service has dealth with the infringement of the regulation and the product is no longer at the Danish Market.

Constituents in fragrance oils
Fragrance oils typically consist of a mixture of essential oils and so-called "nature-identical" constituents (i.e. synthetic substances with a chemical structure similar to the structure of constituents of natural oils). Some of the fragrance oils, however, consisted solely of synthetic substances, of which some cannot be designated "nature-identical". On the basis of MSDSs a list of 92 constituents, which are either included in the list of dangerous substances (6 substances) or classified as dangerous by the producers' self classification, was compiled. None of the substances were assigned the sentences R23 (Toxic by inhalation) or R42 (May cause sensitization by inhalation).

Labelling of the products
For the applications concerned the products must be considered chemical products regulated by the Statutory Order on the classification, packaging, labelling, sale and storage of chemical substances and products. According to the statutory order at least eight of the ten purchased products should have been assigned the sentence R43 (May cause sensitization by skin contact), as they contain more that 1% d-limonene or citral, which are both classified with R43. It should appear from the label that the products are classified with R 43, and which constituent gives rise to the classification. None of the products were labelled correctly. The statutory order includes an exemption where the package contains no more than 125 ml, but this exemption does not apply to products classified with R43. However, most of the products were, provided with instructions for avoiding contact with eyes, mucous membranes and skin because of the risk of irritation/allergy. The Chemical Inspection Service is dealing with the infringement of the regulation according to the labbeling.

Analysis of constituents of selected products
Based on the lists of constituents of the oils 15 constituents were selected for analysis. Already at this stage of the investigation it became clear that only limited information on the effects caused by inhalation of the substances would be available, and the list of selected substances consequently had a predominance of substances, which are also released from building materials, as there has been some attention to the effect of these substances in the indoor environment. A number of substances released from wood - including a range of terpenes - also constitute a significant part of many of the essential oils.

The results of the quantitative analysis of constituents of the oils were in accordance with the information on constituents from the MSDSs considering some natural variation in the composition of the oils. For some of the oils the analyses confirmed that classified substances constituted more than half of the products, e.g. more than 70% d-limonene in lemon oil and more than 70% citral in lemongrass oil were discovered.

Climate chamber tests
Five climate chamber tests were set up in order to investigate to what extent the substances are released to the room where the user stays. In three of the tests a candle diffuser, in which the oil is released from a little bowl with water heated by a tea light, was used. In two tests an electrical fan diffuser of the brand name Aroma Stream was used. In all tests 10 drops of oils were added. This corresponds to the upper limit for the recommended quantities; many suppliers recommend 1-5 drops only.

Results of the measurements
The results of the tests are difficult to interpret as the measured emission rates calculated on the basis of the measurements of the substances in the air evacuated from the chamber apparently were much lower than the rates, which could be expected from the quantities of the substances that disappeared from the diffusers. The explanation can be that the substances are decomposed or oxidized from when they were added until they were measured, but a contributing factor could also be that lower numbers are measured because of high atmospheric humidity in the chambers. It is a well-known fact that e.g. terpenes relatively rapidly oxidize in the air, and that the oxidation products contribute to the respiratory effects observed by exposure to the substances. Irrespective of the explanation and to be on the safe side it was decided to undertake two exposure assessments: One assessment based on the actually measured emission rates and a "worst case scenario" assessment, in which it is assumed that half of the added quantity of each substances is released to the room in a chemical state in which it may have an effect similar to the pure substance.

As the investigated oils are not necessarily the oils with the highest content of the substances, the "worst case scenario" was extended with a scenario in which it was assumed that the applied oils are those having the highest content of each substance. For d-limonene "worst case" would e.g. be to apply lemon oil with 72.5% d-limonene, for alpha-pinene it would be pine needle oil in which the substance constitutes 50%, and for diethyl phthalate it would be one of the fragrance oils in which diethyl phthalate constitutes 50%.

Exposure scenario
In the applied exposure scenario it is assumed that the fragrances are released at a constant emission rate for two hours to a standard room of 17.4 m³ with an air exchange rate of 0.5 times per hour. The person stays in the room for 4 hours, and based on the emission rates, the average concentration to which the person is exposed during the 4 hours is calculated. This average concentration is compared to the concentration levels of possible effects.

Toxicity of the substances
Health data were collected for six of the substances with particular focus on information on health effects by inhalation. The six substances are d-limonene, alpha pinene, benzyl alcohol, p-cymene, citral and camphor. For two of the substances, d-limonene and alpha-pinene, systematic reviews of human inhalation experiments are available. For the other four substances the information on health effects on humans by inhalation is very scarce or lacking. While a large number of investigations of the effects of the substances by dermal contact exist, in relation to the use of the substances in cosmetics, investigations of the effects of fragrances by inhalation in general are very rare.

For more of the substances irritative effects on humans and animals are described, but only at relatively high concentrations. D-limonene and alpha-pinene are known as potent airway irritants after oxidation e.g. by ozone. By this process ultrafine particles are formed.

Health assessment
The health assessment applies to a large extent of so-called LCI values and limit values of the Danish Working Environment Authority. The LCI value is defined as the lowest concentration of a certain substance (Lowest Concentration of Interest), which according to our present knowledge at permanent exposure to the indoor air would not imply a risk of hazardous effects to humans. LCI values are used in Denmark and Germany for studies of effects in relation to releases of volatile chemical substances from wood and wood-based materials. For most of the applied LCI values irritation was the decisive effect. As the typical user of aromatic oils does not use them constantly, one should be on the safe side, if the calculated room concentrations during use of the aromatic oils are below the estimated LCI values. Limit values of the Danish Working Environment Authority are generally much higher that the LCI values, as they are applied in relation to shorter time of exposure. LCI values have been identified for four of the six substances closely assessed and for nine of the other substances. In addition to the six investigated substances, comparisons between exposure levels and LCI values or limit values were made for 10 additional substances. It should be noted that several of the LCI values are based on relatively limited data; by way of example the same value has been established for a number of terpenes by analogous considerations.

Health assessment on the basis of emission measurements
For the scenario where the concentration in the model room is estimated on the basis of the actual measurements, the calculated exposure levels, for all substances are below the LCI values and far below the limit values of the Danish Working Environment Authority. It is consequently estimated that the use of the oils in the way assumed in the exposure scenario does not result in any health risk from the evaluated substances.

Health assessment on the basis of worst case scenarios
By the worst case scenario based on the oils with the highest concentration of the substances, the exposure level is estimated to be substantially above the LCI values for many of the substances. This concerns e.g. d-limonene, alpha pinene, camphene, p-mentha-1,4-diene, p-mentha-1,3-diene, beta-pinene and 3-carene. For diethyl phthalate in one of the fragrance oils the exposure level is close to the limit values of the Danish Working Environment Authority which is not appropriate by uses at home.

The results of this scenario indicate that it may not be appropriate to use the oils in these quantities (10 drops in a small room) frequently. The possibility of long term health effects, like chronic bronchitis or pneumonia, caused by repeated, short-term inhalations over longer periods, cannot with any certainty be excluded. At present, however, this assumption rests on no safe foundation according to the literature and experts consulted.

On the basis of the existing data it is not possible to estimate to what extent this worst case scenario is realistic, but the results indicate that there may be a need for further investigations of the problem; if necessary in a more scientific context also examining in which chemical form the substances are present in the air.

Particularly sensitive persons
The health assessment does not concern particularly sensitive persons. The literature describes a particular group of people experiencing symptoms in their upper and lower airways when they breathe perfumes and fragrances containing products. These persons react with airway contraction when tested with a standard substance. This airway reaction is not associated with an allergic reaction, such as asthma. It is not known to what extent essential oils and fragrance oils contain substances that may elicit such reactions in sensitive persons.

Conclusion
The results of the investigation do not indicate any health risks by occational use of aromatic oils,subject to the fact that only a limited number of substances have been assessed. On the other hand, on the basis of the results of the investigation it cannot be rejected that daily use of 5-10 drops of oils in smaller rooms for a prolonged period in the long term may result in airway irritation.

Recommendations
It is recommended, until more certain findings are available, to use small quantities only, to ventilate to room carefully when the fragrance effect is no longer desired, and not to make the use of fragrance diffusers an everyday phenomenon.

It is recommend to further carry out research on the possible effects of long term effects of inhalation of fragrances at concentrations relevant for the indoor environment and on the effects of simultaneous exposure to a number of chemical substances.

It is recommended that the consumers only use oils recommended by the producers specifically for this purpose and follow the instructions on the packaging. It is further recommended, before beginning to use candle diffusers or other ways to disperse the fragrances, to read the safety instructions provided by producers and suppliers on their web-sites or provided in books about the subject.