4 • the essential oil(s) to administer • the frequency of administration • the medium or formulation in which to administer the essential oil • additional factors such as timing of administration in relation to meals, baths or other activities. In making a decision, the following factors may need to be taken into account: • the desired therapeutic action(s) • the intended site(s) of action • the optimum concentration(s) of pharmacologically active constituent(s) desired at the target tissue or organ • the maximum concentration(s) of toxic constituent(s) tolerated in any tissue or organ. For many essential oil constituents, reliable quantitative information about the latter two points is not currently available. However, informed estimates can be made based on the limited amount of existing data on the toxicity and fate of essential oils and different classes of pure compounds after they have been administered by different routes. Since each constituent in a mixture is processed by the body in different ways, the metabolic fate of an essential oil must be understood in terms of its individual constituents. This argument should also apply to compounds that exist as mixtures of isomers, unless proof of identical properties has been demonstrated. To add another level of complexity, the actions of constituents studied in isolation often differ from their actions when given as part of a mixture. Many examples have been quoted in the literature of the benefits of administering compounds as mixtures, often obtained from plant sources, including reduced toxicity. For example, the unexpected lack of genotoxicity and carcinogenicity in essential oils or extracts containing safrole (Ishidate et al 1984; Bhide et al 1991a; Choudhary & Kale 2002), and the reduced acute toxicity in essential oils containing thymol (Karpouhtsis et al 1998). Such benefits may occur as a result of interactions between constituents, phenomena that were discussed in Chapter 3. The passage of molecules inwards through the skin, unlike sweating which is an active, energy-requiring process, occurs mainly by passive diffusion (Grandjean 1990). We know that the rate of passive diffusion of a substance through the skin, as with any cell membrane, depends on its concentration gradient (i.e., the difference between its concentrations on either side), and is given by Fick’s first law.1 It is also influenced by solubility in aqueous and lipid media (i.e., hydrophilicity and lipophilicity, respectively) and its molecular size. These aspects are discussed below. The bioavailability of some essential oil constituents administered by different routes has been reviewed (Kohlert et al 2000). In most studies, these compounds are reported to be quickly absorbed after oral, dermal and inhalational administration, though it is concluded that there is a general lack of good quality pharmacokinetic data in humans. Table 4.1 gives the peak serum concentrations of some constituents administered by different routes. Although the first three compounds in this table seem to have an extremely low bioavailability from dermal application, the study that most closely approximates to an aromatherapy massage is the one by Jäger et al (1992a), which involved an abdominal massage with 2% lavender oil in peanut oil for 10 minutes (also see Figure 4.2). This may be because the essential oil was being inhaled as well as being dermally absorbed. Bioavailability is subject to biological variation. No two individuals will handle the same substance in the same way, and even the same individual will handle the same substance in different ways at different times, because of factors such as health status, nutritional status, age, integrity of skin, and metabolism (see Pharmacogenetics below). For example, in atopic dermatitis, the skin is likely to be broken so its function as a barrier to exogenous substances is compromised (see Percutaneous absorption below). This has obvious implications not only for the choice and dose of an oil for its therapeutic effects, but also for minimizing its toxicity. It is important to know how quickly, and to what extent components of essential oils and vegetable oils penetrate human skin and find their way into the circulation. It is also pertinent to understand the factors that aid or hinder dermal penetration. Surprisingly, these questions have received little attention in the literature, apart from some toxicological studies.2 For many years, biologists believed that the skin formed an impervious barrier to the outside world. We now know that this is not the case and that many substances are dermally absorbed to some degree.3 Volatile chemicals have been detected in human breath following dermal exposure (Thrall et al 2000). Nevertheless, the skin is still an important protective barrier, limiting the rate at which potentially harmful substances enter the body, as well as preventing the loss of body fluids. Apart from its outer, horny layer the skin has other means of protection, including sweating, detoxifying enzymes and certain immune mechanisms (Hotchkiss 1994). Several studies have shown that human skin is less permeable than that of most experimental animals such as rodents (Bartek et al 1972; Garnett et al 1994; Beckley-Kartey et al 1997), although there are exceptions (Yourick & Bronaugh 1997). In the latter study, coumarin passed more easily through human skin than rat skin. Therefore, caution is needed in extrapolating animal data to humans, and in this text we have ignored most existing animal data. The skin is the largest organ in the human body. It is essentially a water-resistant barrier about 3 mm thick, consisting of an outer epidermis (F) and a deeper dermis (G) (Figure 4.1). The outer layer of the epidermis, the stratum corneum (B), is the main physical barrier to free access by external chemicals, and incorporates dead epidermal cells embedded in a lipid matrix. The main lipids are ceramides (41%), cholesterol (27%), cholesterol esters (10%), fatty acids (9%) and cholesterol sulfate (2%) (Tanojo et al 1998). This matrix, where lipids form a highly convoluted mass, is primarily responsible for the very low permeability of the stratum corneum to water (Potts & Francoeur 1991). Below the stratum corneum lies the remainder of the epidermis (the ‘viable’ epidermis) consisting of living cells arising in the deep epidermis, which become flatter as they rise to the surface and replace exfoliated dead cells. Below the epidermis is the dermoepidermal junction, and then the much thicker dermis, containing nerves, sweat glands, sebaceous glands, hair follicles, blood vessels and lymph vessels. Beneath the dermis lies subcutaneous tissue, primarily fat (H). The absorption characteristics of skin differ between individuals and between different areas of the body. The apparent permeability coefficients (Papp) of 13 of the most abundant constituents of Damask rose oil were estimated following application of the oil to human abdominal, breast and upper arm skin in vitro. Significant differences were observed for the permeation of some constituents. For example, eugenol was found to permeate breast and abdominal skin with similar efficacy, but failed to penetrate upper arm skin, while β-pinene was found to permeate upper arm skin significantly better than abdominal skin. It was argued that these differences were related to chemical structure, but overall, no application site was considered preferable to the others (Schmitt et al 2010). There are a few structural differences between black and white skins. These include a higher total lipid content but a lower amount of ceramides in black skin, and fewer cell layers (although equal thickness) in white stratum corneum (Berardesca & Maibach 1996). However, there is no convincing evidence of important differences in percutaneous absorption between different colored skins. Since the cells of the stratum corneum are not living, they are incapable of registering a physiological response to toxic chemicals. Therefore, before an essential oil constituent can cause a toxic response in the skin, or indeed anywhere else in the body, it must first cross the stratum corneum. Two pathways are theoretically available for this: the intercellular route (between the skin cells) and the transcellular route (through them) (Michaels et al 1975). The intercellular regions are full of lipids structured in multi-lamellar arrays, through which more than one route of molecular diffusion can be envisaged, and there are data to support this view (Albery & Hadgraft 1979; Bunge et al 1999). A third possible route of entry, through the hair follicles, bypasses the stratum corneum altogether (Scheuplein & Blank 1971; Meidan et al 2005). Lipophilic substances might diffuse preferentially through hair follicles and sebaceous ducts using sebum, a lipophilic secretion, as a transport medium. This might explain why, for example, the flux of coumarin across human scalp skin is higher than for human abdominal skin (Ritschel et al 1989). Terpinen-4-ol has been found in the hair follicles of cattle udder skin after application of 5% tea tree oil in a variety of vehicles. Its concentration in sebum was 0.16–0.43%, depending on the vehicle used (Biju et al 2005). There is good evidence that many essential oil constituents travel from the skin surface to the stratum corneum, then to the dermis and the blood circulation. In a study using isolated human skin, methyl salicylate penetrated both the stratum corneum and the dermis, and was also detected in subcutaneous tissues.4 This was due to direct tissue penetration, and not redistribution by the systemic blood supply (Cross et al 1998). Once a constituent has been absorbed, the epidermis may act as a reservoir, retaining a proportion for up to 72 hours before it crosses the dermoepidermal junction, enters the dermis, and then the blood capillaries (Chidgey & Caldwell 1986; Beckley-Kartey et al 1997; Hotchkiss 1998). The majority is absorbed within 24 hours. In an ex vivo dermal absorption study with tea tree oil, 2.75% of the applied dose, mainly terpinen-4-ol, crossed the epidermis, and 0.3% was retained in it after 24 hours (Cross et al 2008). Following dermal application of a cosmetic tanning lotion containing bergamot oil to the forearm skin of 11 healthy volunteers, bergapten was detected in the dermis after 220 minutes (Makki et al 1991). In a similar study, bergapten was found in the dermis, and also in plasma after repeated applications (Moysan et al 1993). In a further study, bergamot oil, emulsified jojoba oil containing bergapten, and cleansing foam with bergamot oil were each applied to the forearms of volunteers. The skin was left uncovered. Bergapten was detected in the stratum corneum after 120 minutes, and in the volunteers’ blood after 240 minutes (Wang & Tso 2002). Since bergapten is not volatile, there is no possibility of inhalatory absorption; it could only have reached the bloodstream via the skin. An analgesic ointment containing methyl salicylate was applied topically to the thighs of 12 human volunteers twice daily for four days. Blood concentrations of 0.31–0.91 mg/L of methyl salicylate were detected within 1 hour of application, rising to a maximum of 2–6 mg/L after the seventh application. Urinary recovery of methyl salicylate and metabolites during the first 24 hours averaged 175.2 mg (Morra et al 1996). In a study of camphor, menthol and methyl salicylate applied undiluted to human volunteers by skin patch for eight hours, peak blood levels were 16.8, 19.0 and 26.8 ng/mL, respectively, for four patches, and 29.5, 31.9 and 41.0 ng/mL, respectively, for eight patches. Systemic exposure was said to be low, in spite of the long period of time and large number of patches (Martin et al 2004). There is one report that closely approximates to what happens in an aromatherapy massage. Peak plasma concentrations of two lavender oil constituents were detected 20 minutes after the oil had been applied by massage; after 90 minutes, concentrations had fallen close to zero (Figure 4.2). In this study, 1.5 g of 2% lavender oil in peanut oil was massaged over the abdomen for 10 minutes, and blood samples were drawn from the arm (left cubital vein) 0, 5, 10, 20, 30, 45, 60, 75, and 90 minutes after finishing the massage. The two constituents measured were linalool (24.8% of the oil) and linalyl acetate (29.6% of the oil). Linalool peaked at 120 ng/mL plasma after 20 minutes (Jäger et al 1992a). Figure 4.2• Blood levels of linalool and linalyl acetate after the application of lavender oil by massage. (Reproduced with permission from Jäger et al 1992a Percutaneous absorption of lavender oil from a massage oil. Journal of the Society of Cosmetic Chemists 43:49-54. © The Society of Cosmetic Chemists.) This study did not control for inhalation and pulmonary absorption. However, in an earlier report by Schuster et al (1986) subjects inhaled ‘clean’ air, and plasma concentrations of α-pinene, β-pinene, camphor, δ-3-carene and (+)-limonene were determined over a period of three hours. In 12 human volunteers, 2 g of Pinimenthol ointment (containing eucalyptus oil, pine needle oil, menthol, camphor) was applied over a 400 cm2 area of skin, and plasma levels peaked at between 1 ng/mL (camphor) and 10 ng/mL (α-pinene) after 10 minutes. After application to uncovered excised human skin, varying amounts of essential oil constituents have been absorbed (Table 4.2). A proportion of the volatile compounds evaporate, but occlusion reduces loss and increases skin permeation. The maximum percentages of any essential oil constituent absorbed dermally over 24 and 72 hours were 33% and 50%, respectively, for coumarin (Beckley-Kartey et al 1997). However, coumarin is rarely found in essential oils, and the 24 hour maximum of 5.9% for benzyl alcohol is more typical. If we assume that a further 50% (of the 5.9%) is absorbed over the next 48 hours, this would extrapolate to about 8.8%. In Table 4.3 we have rounded this up to 10%, and this is the maximum we have assumed for almost all essential oil constituents in setting safety limits for dermal application. Using this parameter, the maximum quantity absorbed from dermal application in 24 hours (0.15 mL) is therefore less than that of oral dosage (0.22–0.66 mL, as shown in Table 4.7). The high levels of skin absorption for methyl salicylate reported by Moody et al (2007) may, in part, be due to the use of acetone as a vehicle. Acetone disrupts the barrier function of skin in mice mainly by removing corneocytes and extracting nonpolar lipids (Rissmann et al 2009; Kamo et al 2011). Table 4.2 The percentage absorption of essential oil constituents over a 24 hour period through uncovered, excised human skin a1% Tween, 0.9% NaCl, 98.1% water c52% coumarin was absorbed over 72 hours. None of the other studies measured absorption for this long Table 4.7 Daily oral doses of essential oil-based medicines Daily totals range from 50 mg (cinnamon bark oil) to 1,200 mg (GeloMyrtol forte), or 0.05–1.3 mL. The majority of daily doses range from 200–600 mg (0.22–0.66 mL) aMain active ingredient by volume is 1,8-cineole, and also contains (+)-limonene and α-pinene bContains peppermint, cinnamon, clove, lavender and thyme oils. No datum is currently available for the percentage skin penetration of α-thujone or methyleugenol, but values can be estimated from their permeability coefficients, which are directly related.5 The in vitro skin permeability coefficient (Ps) of a substance can be measured by applying it in a suitable vehicle to human epidermis in a thermostatic diffusion cell. Ps values for (−)-camphor, (−)-carvone, 1,8-cineole, (−)-linalool, (−)-menthol, α-thujone, (−)-menthone and (E)-anethole ranged from 1.51 × 10− 3 cm/h to 0.14 × 10− 3 cm/h (Gabbanini et al 2009). α-Thujone has a similar Ps value to (−)-linalool (0.62 × 10− 3 cm/h and 0.82 × 10− 3 cm/h, respectively) using reconstructed human epidermis of unstated origin (Gabbanini et al 2009). Methyleugenol has a similar reported apparent permeability coefficient (Papp) to linalool and geraniol, i.e., 5.23 × 10− 5, 3.87 × 10− 5 and 3.22 × 10− 5 cm/s, respectively, for human abdominal skin, and 6.29 × 10− 5, 4.12 × 10− 5 and 4.11 × 10− 5 cm/s for human breast skin (Schmitt et al 2010). Therefore, it is reasonable to assume that the dermal absorption of methyleugenol and α-thujone is within the range of 2–5%. The stratum corneum has both hydrophilic and lipophilic regions. Highly water-soluble molecules, such as glucose, have difficulty passing through lipid-rich regions, while highly lipid-soluble substances, such as cholesterol, have a low probability of crossing the aqueous regions. In general, lipophilic substances cross the dermal barrier more readily and more extensively than hydrophilic ones (Wester & Maibach 2000). However, some water solubility is important too, to facilitate the passage of a substance from the dermis into the bloodstream.6 We would therefore expect substances that pass most readily from the surface of the skin into the bloodstream to have a favorable balance between water and lipid solubility (Hansch and Fujita 1964, Wepierre et al 1968).7 Attempts to derive mathematical models for the dermal penetration and absorption of topically applied substances are still being refined (Williams & Riviere 1995). A convenient measure of the relative solubilities of a substance in lipid and aqueous media is its partition coefficient (P), usually expressed in its logarithmic form, log10 P.8 For the dermal absorption of essential oil constituents, an optimum log P value of 2–4 (for the n-octanol/water system) has been proposed (Cal 2006b), and in an in vitro study of human skin penetration, terpinen-4-ol penetrated the epidermis and dermis more rapidly than 1,8-cineole, α-pinene and β-pinene (Cal et al 2006). However, no clear relationship with log P is apparent here. Log P (octanol/water) values for a range of essential oil constituents are given in Table 4.4. Table 4.4 Log10 P (octanol/water) values for some essential oil constituents Constituents with log P values within the range of 2–4 are thought to penetrate skin more readily than those with log P values either higher or lower (Cal 2006b). Lower values indicate greater water solubility, and higher values indicate greater lipid solubility aThis book, calculated for the octanol/water system using the CLogP algorithm in ChemDraw Ultra 6.0 The uptake into, and the elimination of (±)-β-citronellol, (±)-linalool and linalyl acetate from human skin were studied in vitro. All compounds easily penetrated all layers of the skin, and (±)-linalool promoted its own uptake. No clear relationship between absorption or elimination and log P values was seen, and the most lipophilic compound, linalyl acetate, was the least well absorbed (Cal & Sznitowska 2003). However, in a study comparing a wide range of compounds, including some essential oil constituents, a highly significant linear relationship between percutaneous absorption across excised human skin in vitro and octanol/water log P values, and an inverse relationship with molecular mass, was found (Cronin et al 1999). Many essential oil constituents appear to enhance their own dermal uptake and that of other substances. Some, such as methyl salicylate, may do this in part by acting as rubefacients,9 increasing local capillary blood flow (Cross et al 1999). Others temporarily alter the transport properties of the stratum corneum, through interacting with the intercellular lipids (Williams & Barry 1991). They can either be applied as a skin pre-treatment, or may be formulated with the vehicle. For example, carveol, α-terpineol and terpinen-4-ol significantly boosted permeation of water and ethanol in isolated human epidermis after 4 hours (Magnusson et al 1997). Similarly, (+)-limonene enhanced the permeation of citronellol and eugenol, and both α-pinene and β-myrcene enhanced phenylethanol permeation (Schmitt et al 2009). When 1–2 mL of undiluted terpenes were applied to a small area of uncovered human skin for 12 hours, 8.9% of the applied dose of (+)-limonene, 26.2% of 1,8-cineole and 39.6% of nerolidol were detected in the stratum corneum. These compounds appear to enhance their own accumulation by disrupting the integrity of intercellular lipid bilayers (Cornwell et al 1996). This could be a problem with topical drugs. 5-Fluorouracil (5-FU) is a prescription pharmaceutical used in a cream base at 5% as a treatment for sun-damaged skin cells. In a study using excised human skin, pre-treatment with four essential oils enhanced the dermal absorption of 5-FU, from 2.8-fold for anise oil, to 34-fold for eucalyptus oil (Williams & Barry 1989). The transdermal absorption of other drugs, not normally applied to the skin, is enhanced by essential oils and constituents, for example aspirin ((+)-limonene, in vivo, human), indomethacin (cardamom oil, in vivo, rabbit), and nicrorandil (carvone, in vivo, human) (McAdam et al 1996; Huang et al 1999; Krishnaiah et al 2006). For medications, controlled dosage is important for safe and effective treatment, and any enhancement from coincidental aromatherapy could have adverse effects. Therefore essential oils should not be applied to skin to which any medication is applied, or on which drug patches are being used (e.g., hormone or nicotine patches). In various animal studies, both eucalyptus oil and camphor enhanced nicotine absorption (Nuwayser et al 1988). A recent study attempted to rationalize the permeability-enhancing ability of a group of 49 terpenes (Kang et al 2007). The permeability coefficient of haloperidol through excised human skin appeared to be related to lipophilicity, molecular weight, boiling point, terpene type and the presence or absence of certain functional groups. However, the applicability of this relationship to other drug types is not known. When administered in a vehicle, the tendency of a substance to leave that vehicle and diffuse through the skin is related to the difference in lipophilicity between substance and vehicle. Being predominantly lipophilic, essential oil constituents tend to move from aqueous to lipid environments. Thus, when applied to the skin diluted in a vegetable oil, its constituents will diffuse into the skin more slowly than when they are dispersed in a semi-aqueous medium (Florence & Attwood 1998). This has been shown for benzyl alcohol, isoeugenol and methoxsalen using human epidermis (Gazith et al 1978; Jimbo et al 1983). When (±)-linalool or terpinen-4-ol were applied to human skin in vitro at 5% in three different vehicles, the absorption rates of the terpenes into the stratum corneum decreased from hydrogel to grapeseed oil to a mineral oil-in-water emulsion (Cal 2006a). Compounds with molecular masses greater than 500 usually have more difficulty passing through skin than smaller ones (Cronin et al 1999). These include triacylglycerols (fatty acid esters), which are found in vegetable oils. Being highly lipophilic, triacylglycerols are likely to penetrate the stratum corneum which contains fatty acids, but may not penetrate any further. Free fatty acids, however, combine lipophilicity with molecular masses well below 500, and can therefore be absorbed through skin. The transdermal absorption of fatty acids may be further promoted by some essential oil constituents. A preparation containing various concentrations of 1,8-cineole, ethanol and borage oil containing ~ 25% γ-linolenic acid (GLA) enhanced the in vitro permeation of GLA through full-thickness pig skin (Ho et al 2004). Conversely, the permeation of several drugs through mouse skin was greatly enhanced by the addition of unsaturated fatty acids in propylene glycol, the most effective being oleic acid. Saturated fatty acids were mainly less effective (Oh et al 2001; Gwak & Chun 2002). Using full-thickness human skin, the permeation of tamoxifen was enhanced either by the addition of borage oil or GLA. These effects were explained by the formation of solvation complexes with greater penetrant properties (Karia et al 2004; Heard et al 2005). Permeation enhancement has also been reported using oleic, palmitoleic and linoleic acids in combination with benzyl alcohol (Nanayakkara et al 2005). Although an oil is not an ideal vehicle for promoting the absorption of lipophilic substances (Bowman & Rand 1980), the rate of passage across the skin will increase with concentration according to Fick’s first law.1 For this reason, we would expect the constituents of undiluted essential oils may diffuse into the skin more rapidly than when they are diluted in a vehicle. For example, when linalyl acetate was applied undiluted to human skin in vitro, absorption into the stratum corneum, viable epidermis and dermis were all noted, but when diluted at 0.75% in hydrogel, grapeseed oil or emulsion, no absorption was seen (Cal & Sznitowska 2003; Cal 2006b). In regard to safety, the application of undiluted essential oils to the skin is controversial. It is proscribed by most aromatherapy schools and practitioner associations. Trade organizations such as the UK’s Aromatherapy Trade Council require their members to include a safety statement on bottles of essential oils to the effect that the oil should not be used undiluted on the skin. On the other hand, the ‘raindrop technique’ involves the application of undiluted essential oils, specifically: oregano (2–4 drops), thyme (3–5 drops), basil (6–10 drops), cypress (6–10 drops), wintergreen (6–10 drops), marjoram (6–10 drops) and peppermint (6–10 drops), making a total of 35–59 drops (Essential Science Publishing 2004).10 There are reasons for avoiding this practice, especially in vulnerable groups such as infants, children or the elderly. First, the risk of skin reactions increases with essential oil concentration (see Ch. 5, p. 72) and the widespread use of raindrop technique could lead to an escalation of skin allergy to essential oils. Undiluted thyme and oregano oils, for example, pose a risk of skin irritation. Second, when essential oils are applied undiluted to the skin, percutaneous absorption may lead to relatively high constituent concentrations in the bloodstream, which increases the risk of systemic toxicity. Wintergreen oil, for example, is moderately-to-severely toxic, and many basil oils are potentially carcinogenic, with recommended dermal use levels of below 2% (see Chapter 13, Basil profiles). Finally, the risk of drug interactions is increased. Topically applied methyl salicylate can increase the anticoagulant effect of warfarin, causing side effects such as internal hemorrhage (Le Bourhis & Soenen 1973), and wintergreen oil contains 98% methyl salicylate. There may be scenarios in which undiluted essential oils can be safely and profitably applied to the skin, possibly in the medical treatment of localized infections, and where benefits outweigh risks this may make sense. However, most practitioners should avoid using essential oils in this way, and encouraging untrained people to apply concentrated essential oils to themselves or others is unwise and unsafe. Table 4.5 gives recommended maximum concentrations for general massage purposes, and these also apply to any application to large areas of skin. For local application, concentrations greater than 5% may be appropriate. For wound healing (cuts, burns, sores and ulcers), concentrations between 4% and 12% can be both safe and effective (Guba 1998/1999; Jandera et al 2000; Hartman & Coetzee 2002; Kerr, 2002; Dryden et al 2004). Hydration of the stratum corneum, such as occurs during a bath or shower, facilitates essential oil absorption (Bowman & Rand 1980). For example, the dermal penetration of terpenes increased when essential oils were put in hot bath water, although inhalation was not controlled for (Römmelt et al 1974). By measuring the amount of salicylate excreted in urine, an early, very detailed study concluded that methyl salicylate permeation through the skin of the hand is greatly enhanced by prolonged prior immersion in hot water. It was also found that massage of the hand increased dermal absorption by 34–158%, depending on the experimental parameters (Brown & Scott 1934). This may be due to the stimulating effect of massage on blood flow, though in a study using excised human skin, pressure alone increased the transcutaneous absorption of caffeine by up to 1.8 times (Treffel et al 1993). A rise in temperature of 10 °C in the oil or oil suspension in which volunteers’ hands are immersed increases the rate of percutaneous absorption several-fold, presumably because of enhanced capillary circulation in the area (Hotchkiss et al 1992). The use of an aqueous medium does not mean an increase in skin reactivity. In a small study of hand dermatitis patients allergic to hydroxycitronellal, there were no reactions when the fingers were immersed in water containing 0.025% of the substance (Heydorn et al 2003c). When applied to warm, unoccluded (uncovered) skin, in general, the smaller constituents are more volatile than the larger ones. The more volatile essential oil constituents partially evaporate, reducing the amounts available for absorption. This may partly explain why only a small proportion (2.61%) of benzyl alcohol (liquid, boiling point (bp) = 108 °C) is absorbed even though it readily partitions through the skin (Jimbo et al 1983), while a much higher proportion (33.0%) of coumarin (solid, bp = 301°C) is absorbed (Beckley-Kartey et al 1997). Both substances were applied, diluted in ethanol, to unoccluded, excised human abdominal skin (benzyl alcohol) or breast skin (coumarin) and assessed after 24 hours. If the skin is occluded with a non-permeable material after application of an undiluted essential oil, absorption into the bloodstream is greatly increased (Wester & Maibach 1983; Bronaugh et al 1990). Using epidermal tissue from six donors and 70/30 v/v ethanol/water as a vehicle, the total amount of linalool absorbed after 24 hours was 3.57% for unoccluded skin, and 14.1% for occluded skin (Lapczynski et al 2008g). Occlusion changes the temperature and hydration of the skin, as well as minimizing evaporation, and these physical factors affect absorption (Wester & Maibach 1983). When the skin is damaged or diseased, the rate of absorption of applied substances can be significantly faster. For example, the severity of atopic and seborrheic dermatitis in children correlates with the percutaneous absorption of hydrocortisone (Turpeinen 1988). Similarly, increasing human skin barrier damage correlates with increased dermal penetration of salicylic acid (Benfeldt et al 1999). A larger concentration of bergapten was found in the skin of psoriatic patients than in volunteers with healthy skin following application of a topical gel (Colombo et al 2003). In inflammatory skin diseases, such as psoriasis and atopic dermatitis, there is decreased barrier function, allowing easier dermal penetration (Madison 2003). Barrier disruption produces a cytokine response and an increase in epidermal Langerhans cell density, which may promote inflammation (Ghadially 1998). In atopic dermatitis, there is a significant decrease in the production of certain skin lipids, which in part explains the barrier disruption (Schafer & Kragballe 1991). The increased absorption results in a greater risk of skin reactions, creating a negative cycle (Wester & Noonan 1980). Surprisingly, even non-diseased areas of skin can show barrier dysfunction, whether on parts of the body not affected by skin disease, or in people whose condition has only recently resolved (Berardesca et al 1990). Factors that can negatively affect skin barrier function include psychological stress and chronic alcoholism (Garg et al 2001; Brand & Jendrzejewski 2008). In any type of skin disease, essential oils should always be applied with caution. Neonatal skin is much thinner than adult skin (Lund et al 1999). The skin of pre-term infants is approximately 2.5 times more permeable than adult skin, and before 30 weeks gestation it is 100–1,000 times more permeable (Fischer 1985; Barker et al 1987). Before 30 weeks gestation, the epidermis is thin, has only a few cell layers, and a poorly formed stratum corneum, but by 34 weeks it has largely matured, and by 37 weeks drug absorption and trans-epidermal water loss (TEWL) has considerably reduced (Harpin & Rutter 1983; Evans & Rutter 1986). Epidermal barrier properties undergo a number of significant changes during the first 4 weeks of life, including a decrease in surface pH and an increase in surface hydration (Visscher et al 2000). These progressive adaptation processes continue until 12 weeks (Hoeger & Enzmann 2002). Therefore children up to three months are at increased risk of skin damage from topically applied agents. In the elderly, a number of radical changes take place in the skin (Roskos & Maibach 1992). The corneocytes become less adherent to one another, there is a flattening of the dermoepidermal interface, and the number of melanocytes and Langerhans cells decreases (Fenske & Lober 1986). There is an overall thinning of the epidermis, and TEWL increases significantly, the result being greater permeability, which is further accentuated in photo-aged skin (Lavker et al 1986; Wilhelm et al 1991). This process is due to a reduction in stratum corneum lipids and a profound abnormality in cholesterol synthesis (Elias & Ghadially 2002). Barrier recovery from dermal injury is approximately three times slower in people over 80 than in those of 20–30 years of age (Ghadially et al 1995). In elderly people, the risk of sensitization may decrease due to a reduction in the number of Langerhans cells, although the reduction in barrier function may more than compensate for this. In a comparison of 41 healthy volunteers with a mean age of 24 years and 82 volunteers with a mean age of 75 years, 37% of the older group reacted to at least one of 22 allergens on patch testing, compared with 15% of the younger group (Mangelsdorf et al 1996). • the total quantity of oil(s) applied • the dilution of the essential oil in the vehicle • the vehicle in which the essential oil(s) is/are dispersed • the total area of skin to which the oil is applied • the health and integrity of the skin • the temperature and moisture content of the skin • the extent to which the skin is covered after the massage • how soon the skin is washed following the massage Using the first five variables, it is possible to make an approximate estimate of the range of quantities that will be absorbed. The percentage dilution used for massage over a large area of skin is commonly between 2% and 3%, but with a minimum of 1% and maximum of 5%. The usual vehicle is a vegetable oil such as sweet almond oil, and the total quantity of essential oil applied ranges between a minimum of 5 mL and a maximum of 30 mL for a full-body massage (Harding, Harris, Sade, private communications, 2005). The maximum number of full-body applications in 24 hours is one. The smallest quantity of total essential oil likely to be applied in practice is therefore 0.05 mL, and the largest quantity is 1.5 mL. Table 4.6 is a key to calculating various dilutions of essential oil. Using Imperial units, a simple approach is as follows: Table 4.6 Calculating essential oil concentrations The numbers in the table refer to drops of essential oil. Although we have given precise figures, fractions of drops may be rounded up or down to the nearest whole drop. These figures assume that 30 drops of essential oil = 1 mL. For instance, a 0.4% dilution of bergamot oil will be obtained by mixing three drops of bergamot oil in 25 mL of vehicle, or 6 drops in 50 mL. These figures are averages, as the number of drops per mL can vary from 20 to 40, according to the type of dropper used (Svoboda et al 2001). 0.1% = 1 drop of essential oil per ounce of excipient 1% = 10 drops of essential oil per ounce of excipient Since 1 mL of essential oil is equivalent to 20–40 drops, depending on the type of dropper used (Svoboda et al 2001), 0.15 mL is equivalent to 3–6 drops of essential oil. Great caution is necessary for infants. Since neonatal skin does not mature until three months of age, it is more sensitive and more permeable to essential oils. A newborn is also less equipped to deal with any adverse effects than an adult because of a lower metabolic capacity, i.e., enzymes present in lower concentrations.11 These cautions apply even more to premature babies, and here it would be prudent to avoid all use of essential oils. When massaging or applying essential oils to children, the total dose applied will be less than for adults because of their smaller body size. Recommended and maximum % concentrations of essential oils for children are given in Table 4.5. Inhaled substances pass down the trachea into the bronchi, and from there into finer and finer bronchioles, ending at the microscopic, sac-like alveoli of the lungs, where gaseous exchange with the blood mainly takes place. The alveoli are extremely efficient at transporting small molecules, such as essential oil constituents, into the blood. This efficiency increases with the rate of blood flow through the lungs, the rate and depth of breathing, and with the fat-solubility of the molecules (Breuninger et al 1970; Römmelt et al 1987). The olfactory epithelium, though small, also acts as an absorptive membrane and a high proportion of the molecules that come into contact with the nasal mucosa are absorbed into the general circulation (Gilman et al 1980). Essential oil constituents absorbed via inhalation may enter the bloodstream and reach the central nervous system (CNS) with relative ease. Easy access to the CNS may have safety implications, especially if potentially neurotoxic compounds are being inhaled. There might be particular risks for people with CNS pathologies, such as epilepsy. When eight male volunteers were exposed to air concentrations of 10, 225 or 450 mg/m3 of 97% pure (+)-limonene during light physical exercise, mean respective capillary blood concentrations were 1.5, 11.0 and 21.0 μmol/L after one hour, and 1.5, 12.5 and 23 μmol/L after two hours. Up to 70% of the higher two doses was absorbed into the blood. The authors suggested that it might take three days for the highest dose to be eliminated entirely. The doses of (+)-limonene used were equivalent to evaporating 1–40 g in a 100 m3 room (2 m × 5 m × 10 m). The subjects experienced no irritation or CNS-related symptoms, or any significant changes in lung function variables (Falk-Filipsson et al 1993). In two similarly constructed studies, the human pulmonary uptake was ~ 60% for inhaled α-pinene, and ~ 70% for δ-3-carene at the two higher doses. Total uptake increased linearly with increasing exposure, and the total blood clearance was high. In both reports there were no changes in lung function at the higher levels, but some airway irritation was observed (Falk et al 1990, 1991a). When two male and two female volunteers inhaled air passed over 4 mL of 1,8-cineole for 20 minutes, peak plasma concentrations of 459–1,135 ng/mL were attained after 14–19 minutes, with average absorption half-lives of 3.4 minutes for males, and 10 minutes for females (Jäger et al 1996). No irritation was reported. In experiments with mice, after one hour of continuous inhalation, the plasma concentrations of various constituents (coumarin, α-terpineol, linalool, linalyl acetate) were in the range of 2–10 ng/mL (Jirovetz et al 1991, 1992). Similarly, one hour after 0.5 mL of rosemary oil was evaporated in sealed cages, the air concentration of 1,8-cineole (39% of the oil) was 13.7–15.6 nL/mL. After 60 minutes of inhalation, the mean plasma concentration of 1,8-cineole (in five mice) increased linearly from 4.5 to 15.5 nL/g blood, depending on the amount of rosemary oil evaporated (0.1–0.6 mL/cage) (Kovar et al 1987). The Swedish occupational exposure limit to inhaled terpenes such as δ-3-carene, α-pinene and β-pinene is 150 mg/m3 (Eriksson et al 1997). The reported sensory irritation threshold for inhaled limonene in humans is above 80 ppm, while the NOAEL was estimated to be 100 ppm in mice (Larsen et al 2000). Advantages of the oral route include that it is convenient for the patient, allows for greater precision in dosing, and the bioavailability of oil constituents is often high. For example, after ingestion of GeloMyrtol forte capsules, a treatment for bronchitis and sinusitis, the bioavailability of 1,8-cineole, the main ingredient, was 95.6% (Zimmermann et al 1995). One disadvantage of oral dosing with essential oils is that some of the constituents might irritate the gastrointestinal mucosa, which is generally more sensitive to insult than skin. Since irritation is concentration-dependent, it is important that the essential oil is efficiently dispersed or dissolved in an appropriate vehicle before being swallowed. Preferred methods would be to administer essential oils either in capsules, dissolved in a lipophilic medium such as a vegetable oil, or in aqueous alcohol. As well as preventing gastric irritation, dispersal aids efficient and steady absorption. Note that high viscosity (such as that of vegetable oils) has been shown to slow absorption from the gastrointestinal tract (Gerarde 1960). Virtually all recorded cases of serious poisoning with essential oils have occurred after the ingestion of large amounts of essential oil (see Chapter 3), although these amounts are generally much higher than therapeutic doses. In various studies, the quantities of essential oils that have been taken orally by adults over a 24 hour period range from 0.05–1.3 mL (Table 4.7). The typical oral dosage range (0.22–0.66 mL) is approximately ten times greater than the amount typically absorbed from massage (0.03–0.06 mL). We have assumed that 100% of any oil administered orally is absorbed. Although this is unlikely in every instance, it is appropriate for a worst-case scenario. The frequency of oral dosing for any therapeutic substance is determined by factors such as optimum plasma concentration, duration of treatment and patient compliance. Most importantly, the elimination half-life tells us how often we need to dose to maintain a certain blood concentration. For conditions where this is important, e.g., in treating infections, essential oils are often given three times per day (Belaiche 1979). As can be seen from Table 4.8, there is considerable variation in the elimination half-lives of constituents. Although three daily doses might be appropriate for some oils, the same regimen for thymol-rich oils, for instance, may lead to adverse effects due to accumulation of thymol. Suppositories are sometimes used as alternatives when oral dosing results in significant breakdown of essential oils in the gastrointestinal tract or by first-pass metabolism in the liver, and where high systemic concentrations are desired. Another advantage is that it is the most efficient way to administer a remedy locally to the lower colon. Being lined with a mucous membrane, the rectum is highly sensitive to irritation, especially if the essential oil is unevenly dispersed. Rectal administration of 1,8-cineole, menthol or thymol resulted in respectively high, moderate and zero elimination via the lungs in rats (Grisk & Fisher 1969). There are very few reports of vaginal or vulval reactions to essential oils. There is one alleging a connection between vulvovaginitis and both tea tree oil and the lavender absolute content of a lavender gel (confusingly, the lavender absolute is also referred to as ‘lavender oil absolute’ and ‘lavender oil’). However, clinical relevance is questionable, especially for lavender absolute, which patch tested positive at 10%, but not at 2% (Varma et al 2000). When 92 women with vulval complaints were patch tested, 35 had positive allergic reactions, 15 of which were considered relevant to their clinical condition, most of these being allergies to topical pharmaceutical products. There were four reactions to the fragrance mix (see Chapter 5), one to 2% oxidized (+)-limonene, and one to 2% isoeugenol, but none of these were clinically relevant (Nardelli et al 2004). In terms of safety, a distinction should be made between fragranced products such as intimate wipes, washes, fragrances, etc., intended for frequent application to the female genitalia, and the therapeutic application of essential oil-based preparations to treat vulvovaginitis. Essential oils such as tea tree and geranium may eliminate the infective cause, and directly reduce inflammation (Blackwell 1991; Maruyama et al 2008). For intimate use products, a safety factor of 20 has been suggested in extrapolating from skin to mucosal exposure (Farage et al 2003). For therapeutic products, essential oil concentrations of 1–5% have been proposed, but clinical data are sparse, and it would appear that adverse reactions are rare. The distribution of substances within the body is largely determined by their solubility in the various aqueous and fatty body compartments. In a similar way to that whereby substances reach the bloodstream from their sites of administration, tissue distribution also depends on the relative lipid and water solubility of the substance in question. However, passage from blood into the tissues varies between different tissues. Lipophilic substances are readily taken up into the liver, while water-soluble compounds tend to remain primarily in the blood or move to other aqueous compartments. Diffusion into the brain requires a substance to be appreciably lipophilic because tight junctions between adjacent endothelial cells lining its blood vessels (the so-called blood–brain barrier) force it to pass by the intracellular route. Essential oil constituents are able to penetrate the blood–brain barrier, and have been observed to interact with various receptor sites in the brain, such as those for GABA and glutamate (Aoshima & Hamamoto 1999; Elisabetsky et al 1999). Being predominantly lipophilic, mono- and sesquiterpenes would be expected to spend a short time in the bloodstream before being redistributed first to muscle, and then over a longer period of time to fat. However, repeated high doses can lead to toxicity due to accumulation. This may explain why a woman who ingested 20 drops of thuja oil twice a day for five days, had a seizure and fell after the tenth dose (Millet et al 1981). Essential oil constituents probably remain in fatty tissues for several hours or days.13 For example, rats and mice eliminated all of an oral dose of citral (up to 1 g/kg) within 72 hours and 120 hours, respectively, and after dermal application of benzyl acetate to rats, virtually all the absorbed dose was excreted within 24 hours (Phillips et al 1976; Chidgey et al 1987). Citral is significantly more lipophilic than benzyl acetate, as suggested by their calculated log P (octanol/water) values of 3.0 and 2.0, respectively.8 In experimental animals, thymol, carvacrol, eugenol and guaiacol (log P (octanol/water) = 3.2, 3.4, 2.4 and 1.3, respectively)8 redistributed rapidly to the blood and kidneys following oral administration (Schröder & Vollmer 1932). Most of an intravenous dose of (E)-anethole (log P = 3.3) given to mice was accumulated by the liver, lungs and brain (Le Bourhis 1968). In rats, citral (log P = 3.0) was rapidly and completely absorbed from the GI tract and then redistributed equally to all the tissues (Phillips et al 1976). Following an intragastric dose of 500 mg/kg linalool (log P = 3.4) in rats, 96% was excreted within 72 hours (Parke et al 1974b). The data in Table 4.9 show that relatively small amounts of constituents are excreted following inhalation. Slower excretion probably indicates a longer time in body tissues, compared to oral administration.
Kinetics and dosing
Absorption
Bioavailability
The substance administered
The mode and route of administration
The recipient
Dermal administration
The structure of the skin
Percutaneous absorption
Molecular size and solubility
Permeability enhancement
The delivery vehicle and concentration
Hydration, temperature and pressure
Volatility and occlusion
Trauma
Age
Dermal dosing
Infants and children
Inhalation
Inhalation dosing
Oral administration
Oral dosing
Frequency of dosing
Rectal administration
Vaginal administration
Distribution
Accumulation in tissues
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
Kinetics and dosing
