The respiratory system

CHAPTER CONTENTS

Due to their large surface area, rich blood supply, and the thin membrane separating air and blood, the lungs are very efficient organs for the absorption of gases and volatile substances from the air. The mucous membranes of the nasal cavity and pharynx make a contribution to gas exchange, but much less so. Conversely, exhalation may contribute to the excretion of blood-borne volatile compounds that have reached the bloodstream by any route. The lungs are therefore exposed to airborne toxic vapors as well as xenobiotic chemicals carried in the bloodstream. In addition to carcinogens and other toxins entering the body, the respiratory system is susceptible to irritation and allergic reactions. The significance of the lungs as ports of entry to the body is reflected in the presence of important local detoxifying enzymes.

Many factors affect air quality. The major contributors to the indoor aerial environment are bio-odorants (including human body odor, molds, and animal-derived materials), tobacco smoke, volatile household materials (found in cleaning agents, new carpet and paint) and fragranced products such as deodorants, perfumes and air fresheners (Cone & Shusterman 1991). Outdoor sensitivities can be triggered by vehicle exhaust, pesticides, dust storms and plants (Bener et al 1996; Baldwin et al 1997, 1999). Smoke from fires can also cause problems, whether indoors or outdoors.

Essential oils may be inhaled incidentally because they are present in the ambient air from plant emissions, personal care products, household products, etc., or because they are intentionally vaporized (in candles, essential oil vaporizers, aromatic baths or steam inhalations). Reasons for intentional use include relaxation, masking unpleasant odors, and reducing airborne molds, bacteria or viruses. In order to alleviate symptoms of respiratory disease, essential oils may be used in environmental or steam inhalation, or applied to the chest in the form of a ‘rubbing ointment’.

Many essential oils and essential oil constituents have significant antimicrobial activities, and hence offer potential benefits in respiratory infections. For example, the vapors of cinnamon, thyme, and oregano essential oils, and their constituents, cinnamaldehyde, thymol, and carvacrol, were effective against Gram-negative bacteria (Escherichia coli, Yersinia enterocolitica, Pseudomonas aeruginosa, and Salmonella choleraesuis), Gram-positive bacteria (Listeria monocytogenes, Staphylococcus aureus, Bacillus cereus, and Enterococcus faecalis), molds (Penicillium islandicum and Aspergillus flavus), and a yeast (Candida albicans) (López et al 2007).

Inhaled essential oils might also be harmful, and fundamental safety questions that should be addressed include:

• can essential oils contribute to poor air quality?

• can essential oils cause respiratory disease?

• can essential oils exacerbate the symptoms of pre-existing respiratory disease?

• what are the NOAELs for inhaled essential oil vapors?

• under what circumstances is risk increased or reduced?

Inhalation toxicity is an emerging field of study, and even for recognized respiratory diseases, causal relationships with inhaled substances have not been fully elucidated. In addition, there is enormous scope for new compounds to be formed in the atmosphere, especially indoors. This combination of incomplete understanding and chemical complexity make it difficult to draw clear conclusions and establish safety guidelines. Adding to this complexity, there is sometimes uncertainty about whether a patient has asthma, chronic obstructive pulmonary disease (COPD), multiple chemical sensitivity (MCS), or some other respiratory complaint.

Most of the data in this chapter concern inhalation, but other routes of administration may also be pertinent to the respiratory system (see Pulmonary toxicity). Neurological effects from inhalation are covered in Chapter 10.

Volatile organic compounds

A volatile organic compound (VOC) is defined by the US Environmental Protection Agency (EPA) as ‘any compound of carbon, excluding carbon monoxide, carbon dioxide, carbonic acid, metallic carbides or carbonates, and ammonium carbonate, which participates in atmospheric photochemical reactions’ (www.epa.gov/iaq/voc2.html accessed April 28th 2012). There are hundreds of identified VOCs, including toxic substances such as chloroform, formaldehyde, acetaldehyde, benzene, toluene, xylene and styrene,¹ as well as some essential oil constituents. VOCs form photochemical oxidants, including ozone, that have adverse effects on health, materials, crops and forests. Common sources of VOCs in the home include hair sprays and air fresheners, which may contain volatile propellants in addition to a high proportion of low molecular weight fragrance chemicals.

Much has been published concerning the adverse respiratory and neurological effects of VOCs in general. The regular use of air fresheners and aerosols in the home has been strongly correlated with an increased incidence of diarrhea in infants and headache in new mothers. Total VOC concentration was estimated from measurements of toluene (Farrow et al 2003). If exposure is high enough, VOCs can cause sensory irritation. VOCs that have been associated with reduced respiratory function and/or asthma include formaldehyde, benzene, ethylbenzene and 1,4-dichlorobenzene (Rumchev et al 2002, 2004; Elliott et al 2006). Although few essential oil constituents have been implicated, long-term exposure to moderate concentrations of mixtures containing terpenes entails possible health risks, and the term ‘air freshener’ may often be a misnomer (Anderson & Anderson 1997a).

Adverse effects of airborne substances

The two principal adverse effects of ambient essential oils and other inhaled substances are bronchial hyper-reactivity (BHR), and sensory irritation (SI), which includes both eye and airway irritation. Sensory irritation is non-allergic, although it can exacerbate allergic conditions; BHR can be allergic or non-allergic in origin, and may be associated with respiratory disease such as asthma, allergic rhinitis, or COPD.

A controversial, asthma-like syndrome, known as sensory hyper-reactivity (SHR) is not an allergic reaction and has no known etiology, but may be associated with inhalation of fragrant substances. SHR can be viewed as both an adverse reaction and a respiratory disorder—we have included it under Respiratory disease.

When an odorous substance is sufficiently pungent, trigeminal or vagal afferent C fibers in the nose, mouth and eyes are activated (see Transient receptor potential channels below). This sensory system, called the common chemical sense (CCS), may have evolved as a warning system for potentially hazardous chemicals. It is distinct from olfaction (i.e., the sense of smell), and also reacts to chemicals with no detectable odor, such as tear gases and capsaicin (an alkaloid from Cayenne pepper fruit). It evokes sensations such as irritation, tickling, burning, warming, cooling, and stinging in the nasal and oral cavities and in the cornea, via pain receptors and the trigeminal nerve. Eye irritation may be a good assay for CCS sensitivity, as smell does not interfere with it (Millqvist 2006). Effects may be exacerbated by low humidity.

When 66 healthy males were exposed for 2.75 hr to a mixture of 22 VOCs, including α-pinene, at 0 and 25 mg/m³, a significant majority associated the VOC exposure with increasing headache, degrading air quality, and general discomfort (Otto et al 1990). There is evidence that younger people and allergic rhinitis sufferers are more sensitive than others to VOC sensory irritation (Shusterman et al 2003). A survey of 643 young adults found self-reported illness from inhaling air that carried chemicals from paint, new carpet, perfume (all sources of VOCs), vehicle exhaust or pesticides. A total of 66% reported feeling unwell from one or more of these sources, and 15% to four or more. From a possible list of 11 symptoms, the most commonly reported were daytime sleepiness, daytime fatigue, difficulty concentrating, headache, irritability, and joint or muscle pain (Bell et al 1993). This picture is similar to that of MCS (see Sensory hyper-reactivity).

Sensory irritation may be over-reported, since subjective impressions may be biased by expectation, and researchers have found separating odor perception from true irritation to be a challenge. When two groups of people inhaled methyl salicylate, perceived nose, throat and eye irritation was 5–10 times more intense in the group told it was an industrial solvent, than in those informed it was a natural extract (Dalton 2003). However, objective evidence of nasal mucous membrane irritation (measured as increased neutrophil influx) in humans was found after inhalation of 25 mg/m³ of mixed VOCs for 4 hours (Koren et al 1992).

The essential oil constituents most commonly cited as sensory irritants are listed in Table 6.1. When inhaled by mice, δ-3-carene caused a dose-dependent decrease in breathing rate due to SI, but there was no deterioration of lung tissue (Kasanen et al 1999). When δ-3-carene was inhaled by humans, there was an increase in pulmonary irritation when the concentration increased from 225 ppm to 450 ppm (Falk et al 1991a). Sulfur compounds are also irritants. Inhalation of mustard oil, for example, greatly irritates the eyes and the upper airway mucous membranes (Von Skramlik 1959).

Table 6.1

RD₅₀ concentrations of some VOCs in male mice ^a

Substance	RD₅₀ (ppm)	Calculated irritation threshold (0.03 × RD₅₀)
Chlorine^b	3.5	0.1 ppm
Formaldehyde^b	4	0.1 ppm
1-Octen-3-ol^c	35	1 ppm
3-Octanol^c	256	7 ppm
Benzaldehyde	363	11 ppm
(+)-α-Pinene	1,053	31 ppm
(+)-Limonene	1,076	32 ppm
(+)-β-Pinene	1,279	38 ppm
(+)-δ-3-Carene	1,345	40 ppm
Menthol	1,653	50 ppm
3-Octanone^c	3,360	101 ppm
(−)-β-Pinene	4,663	158 ppm
(−)-α-Pinene	Inactive	No limit

RD₅₀ is the concentration that depresses the respiratory rate by 50%

^a All values from tests in OF1 species, except for formaldehyde and (+)-limonene, which were BALB/CA species

^b Chlorine and formaldehyde, not essential oil constituents, are shown for comparison

^c Minor constituent (up to 5%) of a small number of essential oils

Benzaldehyde data are from Steinhagen & Barrow (1984) and menthol from Luan et al 2006. All other data are from Nielsen et al (2007a), and include data cited therein.

The respiratory tract is intimately connected with the ears and eyes by a mucous membrane, and thus exogenous irritants may affect tissues at one or more of these different anatomical sites. Eye irritation can be provoked by perfume in patients with a history of BHR to chemical triggers (Millqvist et al 1999; Elberling et al 2005), and eye reactions were confirmed in 8 of 21 such susceptible patients in a double-blind study (Elberling et al 2006).

Not all essential oil constituents are sensory irritants. Coumarin, 2-phenylethanol, octanoic acid and vanillin were virtually undetectable by nasal pungency in anosmics or via nasal localization in normosmics at room temperature (Cometto-Muñiz 2005a). These compounds are regarded as being ‘nontrigeminal’, and are therefore unlikely to pose any risk of SI. The exposure of rats or hamsters to finished fragrance in the ambient air at up to 50 mg/m³ for 4 hours/day, 5 days/week for 13 weeks resulted in no toxicologically significant effects. The fragrances included, as major components, coumarin, 2-phenylethanol and benzyl acetate (Fukayama et al 1999).

Terpene oxidation

The irritant and sensitizing actions of many, often aged, essential oils have been ascribed to oxidation products of various constituents, especially unsaturated compounds (see Chapter 2). This could be due to oxygen (O₂) as well as ozone (O₃) and other oxidants. Experiments in mice suggest that respiratory irritation can be caused by oxidation products of α-pinene and of (+)-limonene, two common constituents of indoor air (Wolkoff et al 1999; Clausen et al 2001). Ozone is a powerful oxidizing agent. It reacts readily with alkenes to form unstable ozonides, which go on to form aldehydes, ketones or carboxylic acids that are more reactive and irritating than their alkene precursors. Short-lived oxidizing radicals such as hydroxyl may also be formed, especially during daylight hours, and these also react with alkenes (Calogirou et al 1999).

Ground-level ozone increases during very hot, sunny weather, especially in cities, since it is photochemically formed in the presence of nitrogen dioxide (from vehicle exhaust) and man-made VOCs.² High ozone levels have been epidemiologically correlated with respiratory hospital admission in young children and the elderly, and with asthma morbidity in children (Gielen et al 1997, Yang et al 2003). An increase in outdoor ozone will lead to indoor ozone increases.

In mice, ozone irritancy was marked by shallow breathing followed by a decrease in the respiratory rate. The NOEL of ozone was estimated to be about 1 ppm during 30 minutes exposure. No major effect occurred in resting humans at about 0.4 ppm (Nielsen et al 1999). At 0.5 ppm, ozone did not augment the irritancy of 50 ppm (+)-limonene in mice, suggesting that the NOAEL of ozone is approximately 0.5 ppm (Wilkins et al 2003).

Respiratory parameters were measured in mice for 60 minutes during inhalation exposure to a mixture of 3.4 ppm ozone, and 47 ppm (+)-α-pinene, 51 ppm (+)-limonene, or 465 ppm isoprene. Due to reaction, the ozone concentration at the point of exposure fell to < 0.35 ppm. Upper airway irritation was a prominent effect, as was the development of airflow limitation that persisted for at least 45 minutes after exposure. The effects were reversible within six hours, suggesting that terpene oxidation products may have adverse effects of moderate duration on the airways (Rohr et al 2002). In 10 healthy humans, blink frequency increased by 17% following exposure to the vapors of a 10-minute-old mixture of 92 ppb (+)-limonene and 101 ppb ozone. This increase was accompanied by weak eye irritation, presumed to be caused by oxidation products of limonene (Nøjgaard et al 2005). (+)-Limonene itself is not a strong eye irritant.

Of concern in the context of essential oils is the potential of ozone to react with airborne unsaturated terpenes well below their NOEL concentrations to form irritant compounds and sub-micron particles (Wainman et al 2000; Wolkoff et al 2000). This has been demonstrated for α-pinene, β-pinene, camphene, p-cymene and (+)-limonene (Lamorena et al 2007). Increases in the indoor concentrations of sub-micron particles of up to tenfold over controls were observed for (+)-limonene in the presence of ambient ozone (Weschler & Shields 1999). The extent to which ultrafine particles contribute to airway irritation is, however, uncertain. When mice were exposed to a mixture of 0.05 ppm ozone and 35 ppm (+)-limonene for 30 minutes, the respiratory rate decreased by more than 30%; 75% of the SI here was explained by formaldehyde and residual (+)-limonene (Wolkoff et al 2008).

The relevance of these animal data to human exposure to essential oils in the presence of ozone is questionable. The concentration of 0.05 ppm ozone used in the Wolkoff et al (2008) study is in accordance with the classification of ‘good air quality’ as defined by the Pima County Department of Environmental Quality/US EPA (http://www.airinfonow.com/html/ed_ozone.html accessed April 28th 2012). The 0.8 ppm concentration of ozone used in the Sunil et al (2007) study, however, is greater than the range described as ‘very unhealthy’ in the same guidelines. At this concentration, ozone alone might be expected to cause significant respiratory distress including difficulty in breathing, cough, phlegm, and reduce peak expiratory flow rate in children with mild asthma, and possibly lower airway inflammation.

Apart from its presence in the atmosphere, ozone is formed in vivo as part of the inflammatory response, being released by neutrophils and other white cells. Although terpene oxidation products may be irritating to bronchial tissues, they represent much less of an irritant threat than ozone. It therefore appears reasonable that inhaled alkenes such as (+)-limonene might afford protection against the damaging effects of endogenous ozone, and may have a beneficial scavenging role in asthma. Inhaled (+)-limonene significantly prevented bronchial obstruction in rats constantly exposed to 125 ppm of (+)-limonene for 7 days by means of an electric scented oil warmer. This was therefore a close approximation to the circumstances of essential oil vaporization (Keinan et al 2005).

In rats, age differences have been observed in responses to (+)-limonene oxidation products. When female rats were exposed to a mixture of 6 ppm (+)-limonene and 0.8 ppm ozone for 3 hours, 2 month-old animals showed endothelial cell hypertrophy, perivascular and pleural edema and thickening of alveolar septal walls. In 18-month-old animals, only patchy accumulation of fluid within septal walls in alveolar sacs and subtle pleural edema were noted (Sunil et al 2007). This shows that the pulmonary inflammatory response is significantly attenuated in older, compared to younger rats.

We caution against the use of monoterpene-rich essential oils in high-ozone environments (Tamas et al 2006). It would be sensible to minimize indoor concentrations of ozone in aromatherapy treatment rooms by removing sources such as photocopying machines, laser printers and ozone-generating air cleaners.³ As noted above, ozone is itself a respiratory irritant, and low levels are often present in offices. Both high humidity and increased ventilation reduce the irritancy of mixtures of limonene and ozone (Weschler & Shields 2000; Wilkins et al 2003).

Thresholds

Sensory irritation has been found to slow respiration, and the airborne concentration of a chemical required to depress the respiratory rate in animals by 50% (RD₅₀) is accepted as a standard measure of sensory irritation by the American Society for Testing and Materials (ASTM E981-04).⁴ Mouse log RD₅₀ values correlate well with log (LOAEL) values in humans, suggesting that RD₅₀ is a useful predictor of safe public exposure levels (Kuwabara et al 2007). RD₅₀ values for various VOCs in mice are shown in Table 6.1.⁵ In humans, RD₅₀ concentrations of these substances are expected to produce intolerable irritation, and it has been suggested that 0.03 × RD₅₀ might be a close approximation to the tolerable threshold value in humans (Schaper et al 1993).

The reported airborne sensory irritation threshold for inhaled (+)-limonene in humans is above 80 ppm, while the NOAEL was estimated to be 100 ppm in mice (Larsen ST et al 2000). These data are consistent with the calculated RD₅₀ × 0.03 of 32 ppm for (+)-limonene (see Table 6.1). The American Industrial Hygiene Association has set a limit of 30 ppm for (+)-limonene (0.003%, or 167 mg/m³) for workers continually exposed to it. The Swedish occupational exposure limit for inhaled terpenes is 150 mg/m³ (Eriksson et al 1997). This is approximately equivalent to the vapors of 7 drops of essential oil per cubic meter (150 mg (+)-limonene = 0.238 mL (density is 0.842 g/mL). If there are 30 drops per mL (Table 4.6) this makes 7.14 drops).

(+)-Limonene is not a strong eye irritant. The threshold for eye irritation assessed in 12 healthy individuals was 1,250 mg/m³ (224 ppm). δ-3-Carene had a similar threshold, and α-pinene and α-terpineol were less irritating (Molhave et al 2000).

Four parameters have received much attention as measures of the detectable actions of airborne VOCs. These are the odor threshold (measured in normosmic subjects), nasal pungency threshold (measured in anosmic subjects), eye irritation threshold and nasal localization (left versus right) (measured in both normosmic and anosmic subjects). All four parameters were determined for δ-3-carene, p-cymene, linalool, 1,8-cineole, geraniol and cumene vapors. The three trigeminally mediated thresholds were about three orders of magnitude higher than the corresponding odor thresholds, which ranged between 0.1 and 1.7 ppm. Trigeminal chemosensitivity was comparable between the nose and the eyes, and between normosmics and anosmics. Only 1,8-cineole and δ-3-carene consistently caused eye irritation, at 235 and 1,636 ppm, respectively (Cometto-Muñiz et al 1998a, 1998b).

Good quantitative correlations have been reported between both the human nasal pungency and eye irritation thresholds of various nonreactive VOCs, including esters, aldehydes, ketones, alcohols, carboxylic acids, aromatic hydrocarbons and pyridine, and various calculated physicochemical parameters. These equations have been interpreted as implying that potency is determined by the ease of transport of the VOCs from air into the receptive tissues (Abraham et al 1996, 1998, 2001). An equation for odor thresholds has been derived by including an extra parameter to describe molecular length (Abraham et al 2001).⁶

An essentially similar approach has been taken by Hau et al (1999), who related the nasal pungency of VOCs to measured partition coefficients between octanol and water, and between air and water. A common irritation receptor site was proposed for all VOCs. More recent statistical models for predicting the sensory irritation (logRD₅₀) of reactive and nonreactive VOCs using various calculated constitutional, topological, electronic, and other descriptors have been proposed by Luan et al (2006).

Bronchial hyper-reactivity

In people with BHR, also known as bronchial hyper-responsiveness, there is an exaggerated tendency for the smooth muscle of the tracheobronchial tree to contract in response to a given stimulus, compared to normal individuals. (For testing, either methacholine or histamine is commonly used to provoke a response.) The most prominent manifestation of this contraction is a decrease in airway caliber that can be readily measured with a spirometer. BHR is a diagnostic feature of asthma, and is also a concern in other respiratory diseases. In sensitive individuals, BHR may be elicited by dust mites, pollen or other triggers. It is more prevalent in atopic individuals, and in the young and the elderly.

A correlation was found between BHR and the concentration of (+)-limonene in the home in a Swedish epidemiological study including 47 people with asthmatic symptoms and 41 without. Of those with symptoms, 72% were women. Spirometric testing showed correlations between variability of peak expiratory flow and concentrations of α-pinene and δ-3-carene. Terpene concentration correlated with nocturnal breathlessness or tightness of the chest. The maximum concentration of terpenes was very low, 1.0 mg/m³, with an average of 64% (+)-limonene, 21% α-pinene and 15% δ-3-carene (Norbäck et al 1995). High average concentrations of CO₂ were also noted, presumably due to poor ventilation. None of the correlations reached a level of significance needed to support causation. Respiratory symptoms were more common in dwellings with dust mites, or dampness and microbial growth.

Also in Sweden, sawmill workers exposed to aerosol wood, including α-pinene, β-pinene and δ-3-carene, experienced slightly reduced lung function measured by spirometry (FVC, FEV₁ and MMF, see Table 6.2) although this did not deteriorate upon further acute exposure (Hedenstierna et al 1983).

Table 6.2

Some pulmonary function parameters measurable with a spirometer

FVC	Forced vital capacity	The total amount of air that can be forcibly exhaled after full inspiration
FEV₁	Forced expiratory volume in one second	The amount of air forcibly exhaled in one second
FEV₃	Forced expiratory volume in three seconds	The amount of air forcibly exhaled in three seconds
MEFR	Maximum expiratory flow rate	Maximum forced flow rate during full expiration
MMF	Maximum mid-expiratory flow	The mean forced expiratory flow during the middle half of the FVC
FEF	Forced expiratory flow	Same as MMF
FEF_25%	Forced expiratory flow 25%	The mean forced expiratory flow during the first 25% of the FVC
VC	Vital capacity	The maximum volume of air that can be exhaled after a maximal inspiration

These reports suggest that some airborne terpenes could induce low-to-moderate BHR, though there is no evidence of allergy. Låstbom et al (1998, 2000, 2003) found that prior skin sensitization to δ-3-carene increased BHR to inhaled δ-3-carene in isolated guinea pig lungs, suggesting an allergic mechanism. However, we could find no other evidence for such an effect, and a review of the literature by Nielsen et al (2007b) found little support for the allergy-promoting effects of indoor VOCs. Neither eugenol nor isoeugenol sensitized the respiratory tract in a mouse immunoglobulin E (IgE) test, even though both compounds elicited positive responses in skin sensitization tests (Hilton et al 1996). IgE is associated with atopy, allergic asthma and allergic rhinitis. However, although it may play a role in eczema, it is not associated with contact dermatitis and type IV skin allergy. In a study of BHR to perfume and fragranced products, no significant association was found with atopy, suggesting that the BHR was not IgE-mediated (Elberling et al 2005). In 11 patients with contact allergy to isoeugenol, there were no adverse respiratory effects from inhaling 1 mg/m³ of isoeugenol for 60 minutes (Schnuch et al 2011).

Transient receptor potential channels

An important mechanism in both normal sensory irritation and hyper-reactivity is the interaction of inhaled substances with transient receptor potential (TRP) ion channels located in the airways (Table 6.3). These channels are divided into several subfamilies, including the TRPV (vanilloid), the TRPA (ankyrin) and the TRPM (melastatin) groups. The TRPV subfamily includes channels that are critically involved in nociception and thermo-sensing (Vennekens et al 2008).

Table 6.3

Examples of substances that interact with transient receptor potential channels

TRPV1 and TRPA1 especially, are activated by pungent compounds in spices (Iwasaki et al 2008), and such activation may contribute to MCS, asthma, chronic cough, and other conditions exacerbated by airway reactivity. TRPA1 is activated by chlorine, reactive oxygen species (ROS), and noxious constituents of smoke and smog, initiating irritation and airway reflex responses (Bessac & Jordt 2008). Ovalbumin is used to induce experimental allergic airway hyper-reactivity in mice. However, in TRPA1-deficient mice, there was an almost complete lack of BHR, inflammatory cytokine activity and other signs characteristic of asthma following ovalbumin injection (Caceres et al 2009).

TRPV1 is predominantly expressed in the cell membranes of afferent sensory neurons. On stimulation, they release neuropeptides that trigger effects such as airway inflammation, bronchoconstriction and cough (Adcock 2009; Takemura et al 2008). Evidence in experimental animals and patients with airway disease indicates a marked hypersensitivity to cough induced by TRPV1 agonists (Materazzi et al 2009). Patients with allergic rhinitis have an increased itch response to TRPV1 stimulation (Alenmyr et al 2009). Therefore, TRPV1 receptor antagonists have been proposed as therapeutic candidates (Takemura et al 2008).

(−)-Menthol is a TRPV1 receptor antagonist (Mandadi et al 2009) and so may be therapeutic for cough. Inhaled at 30 μg/L, (−)-menthol reduced evoked cough in guinea pigs by 56% (Laude et al 1994). An inhaled mixture of 75% (−)-menthol and 25% 1,8-cineole significantly reduced cough evoked by citric acid compared to pine oil or plain air in healthy individuals (Morice et al 1994).