10 The nervous system comprises the central nervous system (CNS) and the peripheral nervous system (PNS). Nervous tissue, especially in the CNS, is particularly vulnerable to toxic substances because it has a very limited capacity to regenerate, and if nerves are damaged they may not recover. Although the CNS receives a significant proportion of the cardiac output, it is protected from many blood-borne xenobiotics by the blood–brain barrier,1 which permits access to only the more lipophilic substances. However, most essential oil constituents are moderately lipophilic and some will be neurotoxic, at least in high doses. A neurotoxic substance causes temporary (reversible) or permanent (irreversible) disruption by interfering with the structure and/or function of neural pathways, circuits and systems. Reversible behavioral changes may include cognitive function deficits, effects on mood or sleep (CNS), and muscular effects such as weakness, numbness and alterations in motor coordination (CNS and PNS). Developmental neurotoxicity (by definition caused during pregnancy or lactation) may disrupt behavioral, learning, or other developmental patterns (see Ch. 11, p. 158). Cumulative, or even short-term exposure to a neurotoxin may lead to irreversible damage and chronic abnormalities. Known neurotoxins include industrial solvents such as toluene and trichloroethylene, metals such as mercury and lead, and organophosphates, often used in pesticides and nerve gases. Reversible neurotoxicity has been recorded for the minor essential oil constituent p-cresol.2 Styrene, which can occur at up to 0.1% in cassia oil, is generally regarded as neurotoxic (Gagnaire et al 2006). Hydrocyanic acid, found in unrectified bitter almond oil, is a potent neurotoxin (Reynolds 1993). Wormseed oil is powerfully neurotoxic, and oral ingestion has caused visible edema of the brain and meninges. In most cases of wormseed oil poisoning, considerable CNS damage is seen at autopsy, particularly in the brain (Van Lookeren Campagne 1939; Opdyke 1976 p. 713–715). Because of its high toxicity, this oil should be avoided altogether by any route of administration. Wintergreen oil is a less potent neurotoxin. In overdose, it causes death through cardiopulmonary arrest and respiratory failure. In a review of fatal salicylate poisoning cases, 18% had nervous system abnormalities on autopsy (McGuigan 1987). Structural alterations to the CNS have been seen in repeated dose animal tests using some essential oil constituents. Given orally to rats for 28 days, (−)-menthone produced cerebellar lesions at 400 and 800 mg/kg/day, but not at 200 mg/kg/day (Madsen et al 1986). Also in 28-day studies, dietary (1R)-(+)-β-pulegone and peppermint oil were reported to produce microscopic lesions in the white matter of the rat brain and the cerebellum, respectively (Thorup et al 1983a; Olsen & Thorup 1984; Spindler & Madsen 1992), although other studies have failed to reproduce these lesions using peppermint oil (Mengs & Stotzem 1989; Mølck et al 1998). When male rats were exposed to 300 ppm turpentine vapor for 6 hours daily, 5 days per week for 8 weeks, α-pinene accumulated in brain tissue (Savolainen & Pfäffli 1978). The structural effects seen in these studies are reversible, but long-term exposure at such levels may lead to irreversible damage. Refsum’s disease is a rare recessive familial disorder characterized by peripheral neuropathy, cerebellar ataxia, retinitis pigmentosa, as well as bone and skin changes (Berkow & Fletcher 1992). Its neurological component involves malformations of myelin sheaths around nerve cells, and is caused by an accumulation of phytanic acid in the plasma and tissues due to a deficiency of phytanic acid hydroxylase, an enzyme which metabolizes phytanic acid (Steinberg et al 1966). Phytanic acid is a metabolite of phytol, a constituent of jasmine absolute (and also many foods). Phytol is only found at significant levels in jasmine absolute, and this is not taken orally. Even externally it is not normally used at more than 1%. A whole body massage using 30 mL of oil with jasmine absolute at 1% would entail a maximum of 0.006 mL of phytol being applied to the skin, not all of which is absorbed. Therefore very much less phytol would be absorbed than the dietary 0.5% found to cause no adverse effect in rats over 15 months (Steinberg et al 1966). However, some caution may be appropriate in people with Refsum’s disease. CNS stimulation or depression resulting from essential oil use is generally reversible at moderate doses, with only minor disruption of function. There is one report of dizziness, incoordination, confusion and fatigue in mice exposed to the vapors of commercial fragrances for one hour (Anderson & Anderson 1998). Sensory irritation (see Ch. 6, p. 100) is also a form of neurotoxicity, since it involves an alteration in neurological function. It is interesting to note that in isolated cell or animal models, some essential oils and constituents protect nervous tissue from damage, often through antioxidant mechanisms.3 Both CNS stimulation and CNS depression are manifestations of functional neurotoxicity, since they can alter normal modes/states of cognition, alertness and coordination. CNS stimulants include many ketones, which reduce sleep duration and can also cause seizures (see Table 10.3). Not all CNS stimulants are convulsants, but this is the most important safety issue related to CNS stimulation by essential oils. Table 10.3 CNS stimulant effects in albino mice on subcutaneous administration aLD50 values were assessed for those compounds that were lethal at the convulsant dose. Modified from Wenzel & Ross (1957) • an absence (petit mal) seizure is characterized by a sudden, temporary change in behavior such as a loss of conscious activity or rapid blinking, and may not be noticed by others. These seizures typically occur in people under 20 years of age and last only a few seconds, followed by a full and rapid recovery • a clonic seizure is characterized by rapidly alternating contraction and relaxation of muscles • a tonic seizure is characterized by sustained muscle contraction or tension • a tonic-clonic (grand mal) seizure includes characteristics of both tonic and clonic seizures. During the tonic phase, a sufferer may cry out, become incontinent, lose consciousness and fall to the ground. Muscle contraction may be extreme, and respiration may cease. The clonic phase involves generalized rhythmic jerking. When chronic and recurrent, unprovoked seizures are classified as epileptic and may require medical intervention. Seizures may be precipitated by a number of mechanisms. These include a diminution of GABAergic transmission (as occurs with thujone and pinocamphone) and a drastic reduction in blood supply to the brain (as may happen in some LD50 tests). The immature nervous system is more susceptible to toxic substances because the blood–brain barrier is less effective than that of the adult brain (Saunders et al 2000). There is also a greater susceptibility to seizures because the developing brain exhibits features that enhance neuronal excitability (Jensen 1999, Veliskova et al 1994). Changes that take place in the brain at a relatively young age reduce this excitability and the consequent susceptibility to seizures (Sperber et al 1999). Because of sex-determined differences in brain development, male children have a higher incidence of unprovoked seizures than do females (Veliskova et al 2004). The ketones pinocamphone, (±)-camphor and α- and β-thujone can all cause convulsions (see below) at non-lethal doses in mice, while (1R)-(+)-β-pulegone has been implicated in convulsions at non-lethal doses in humans, when ingested as a constituent of pennyroyal oil (see Pulegone, below). Overdoses of methyl salicylate have also induced seizures. Essential oils containing significant amounts of these compounds therefore present a dose-dependent risk, though this may be modified by the co-presence of anticonvulsant constituents. Table 10.1 shows the maximum safe dermal and oral exposure for each of these constituents. The rationale for these amounts is explained by the appropriate constituent profile, and is also based on the data in Table 4.6. Table 10.2 lists the essential oils requiring dose limitation to avoid neurotoxicity. Table 10.1 Dermal and oral maxima for convulsant essential oil constituents Table 10.2 Potentially convulsant essential oils with dermal and oral maximum doses bRestriction also accounts for safrole and methyleugenol content cSee Adjusting safe doses and concentrations (see Ch. 13, p. 189). Pinocamphone is convulsant and lethal to rats above 0.05 mL/kg ip (Millet et al 1981). The only essential oil that contains more than 5% is the pinocamphone CT of hyssop. Both pinocamphone and isopinocamphone act as GABAA receptor antagonists, a characteristic of some convulsants (Höld et al 2002). ‘Hyssop oil’ normally refers to the pinocamphone CT. The convulsant effects of hyssop oil were first researched in the 19th century. Doses of 2.5 mg/kg were injected ip into dogs, producing almost immediate seizures (Cadéac & Meunier 1891). In subsequent tests, injecting hyssop oil at 1–2 mL revealed a biphasic response. During the first phase, blood pressure fell, breathing became rapid, and random clonic movements appeared. The second phase was characterized by hypertension, rapid heartbeat and numerous clonic contractions (Caujolle & Franck 1945b). In rats, convulsions appeared for hyssop oil at single doses of 130 mg/kg ip, and the mean non-toxic dose was 80 mg/kg (Millet et al 1979). When administered to rats at 20 mg/kg ip for 15 days, cortical effects began to appear after 3–4 days, becoming stronger and more frequent thereafter (Millet et al 1980). Diazepam protected mice against the convulsant effects of hyssop oil (Höld et al 2002). The convulsant action of hyssop oil is assumed to be due to its content of pinocamphone (31.2–42.7%) and isopinocamphone (30.9–39.2%). The amounts of hyssop oil used in these animal tests were generally high, but there are three reported cases of low-dose hyssop oil ingestion by humans resulting in convulsions. The first case was of a 6-year-old whose mother frequently gave him 2–3 drops of hyssop oil for his asthma. During one severe attack, he was given ‘half a coffee spoon’ (perhaps ~ 1 mL) shortly after which he suffered a convulsion. He fully recovered after three days in hospital (Arditti et al 1978). In the second case, an 18-year-old girl ingested 30 drops of hyssop oil to treat a cold. One hour later she lost consciousness for 10 minutes, during which she suffered generalized contractions and bit her tongue (Arditti et al 1978). In the third case a 26-year-old woman took 10 drops of hyssop oil on each of two consecutive days, and had a seizure on the second day (Millet et al 1981). There is clearly a risk of seizures from ingested hyssop oil, although it should be noted that there is a linalool chemotype with ~ 50% of linalool, and only 0.5–1.0% of pinocamphone. The risk of seizures from the linalool CT is probably negligible, especially since linalool is an anticonvulsant (see Unilateral action below). A mixture of α- and β-thujone was convulsant and neurotoxic when administered ip to rats (Millet et al 1980). Subcutaneous thujone was convulsant but not lethal in mice at 590 mg/kg (Wenzel & Ross 1957) and in rats at 36 mg/kg (Sampson & Fernandez 1939) but convulsant and lethal in rats above 0.2 mL/kg (Millet et al 1981). In rats, the highest ip dose producing no convulsions was 0.02 mL/kg (Sampson & Fernandez 1939). Both α- and β-thujone inhibit GABAA receptor-mediated responses (Hall et al 2004). A number of essential oils contain substantial concentrations of thujones (Table 10.2). For example, boldo oil (21.5% total thujone) caused convulsions in rats at an oral dose of 70 mg/kg (Opdyke & Letizia 1982 p. 643–644). Other essential oils for which there are animal data or recorded cases of convulsions are outlined below. Commercial Dalmatian sage oil typically consists of 13.1–48.5% α-thujone and 3.9-19.1% β-thujone. When administered ip to rats, cortical effects began to appear at single doses of 300 mg/kg ip, with convulsions at 500 mg/kg. The maximum non-lethal dose was 3.2 g/kg (Millet et al 1979, 1980). The summary of a Slovenian paper states that Artemisia caerulescens oil, rich in camphor and α-thujone, produced generalized seizures in various test animals (Cvetko et al 1973). Convulsions were induced in a 33-year-old man who accidentally ingested one ‘swallow’ of sage oil (Arditti et al 1978), and in two other adults who ingested 12 drops and at least one swallow (Burkhard et al 1999). In a fatal case, a 44-year-old woman ingested approximately 0.25 oz (~ 7 mL) of Dalmatian sage oil, which was later found to be high in thujone. She suffered several episodes of convulsions in the hours between ingesting the oil and dying (Whitling 1908). She had chronic asthma, and this may have contributed to the fatal outcome. Commercial thuja oil typically consists of 50% α-thujone. There is one report of thuja oil causing convulsions. A 50-year-old woman ingested 20 drops twice a day for 5 days. She was advised to take it by a ‘naturologist’, but did not follow instructions to dilute the oil to 1% before taking the drops. Thirty minutes after her tenth dose she suffered a tonic seizure and fell, fracturing her skull (Millet et al 1981). Commercial wormwood oil typically consists of 33.1–59.9% β-thujone. In perhaps the earliest formally recorded case, a male adult ingested “probably about half an ounce” of wormwood oil. Within minutes he was unconscious, convulsing, foaming at the mouth with his jaw clenched. After prompt medical intervention the man survived, but he could not remember taking the oil (Smith 1862). A 31-year-old man was hospitalized after ingesting 10 mL of wormwood oil. He had been found, by his father, in an agitated, incoherent and disoriented state. Paramedics noted tonic and clonic seizures with decorticate posturing (Weisbord et al 1997). The dose of subcutaneously injected camphor required to produce convulsions in mice was 600 mg/kg (Wenzel & Ross 1957). The camphor used is described as a ‘commercial synthetic product’. ‘Synthetic camphor’ normally denotes (±)-camphor. Humans are more susceptible than rodents to camphor neurotoxicity. Convulsions in children of 2 and 3 years of age were survived following ingestion of 9.5 mL and 700 mg camphor, respectively (Phelan 1976; Gibson et al 1989). In a fatal case, a 16-month-old boy ingested one teaspoon of camphorated oil (vegetable oil with 20% camphor). He experienced frequent fits, constricted pupils, rapid pulse and an extremely high respiratory rate (Smith & Margolis 1954). Two cases of intoxication were survived by a 19-year-old and a 72-year-old. Both ingested 1 oz of camphorated oil (~ 6 g camphor) which caused generalized seizures with no serious consequences (Reid 1979). A 37-year-old man experienced prolonged tonic-clonic seizures after ingesting ~ 90 mL of camphorated oil. He was given hemodialysis and survived (Kopelman et al 1979). A man attempted suicide by ingesting 150 mL of camphorated oil (~ 30 g camphor). He suffered peripheral circulatory shock and severe, prolonged grand mal attacks, but he survived after intensive treatment (Vasey & Karayannoppoulos 1972). This is one of the highest doses of camphor to be survived. Camphor does not have to be ingested to initiate convulsions. A 15-month-old child suffered loss of muscular coordination and seizures after crawling through camphorated oil spilled by his sibling. He recovered fully. This case may represent sensitivity to camphor in a near-epileptic (Skoglund et al 1977). A 9-month-old child had three seizures in the 24 hours after a dressing containing ~ 15 g of camphor was administered to thoracic burns. The level of camphor in the blood was 2.6 mg/L at the time of the seizures. It is assumed that most of the camphor was percutaneously absorbed (Joly et al 1980). Toxic effects may follow a pattern of CNS stimulation (delirium, seizures) followed by depression (lack of coordination, respiratory depression, coma (Budavari 1989)). Neurologic symptoms can include anxiety, depression, confusion, headache, dizziness and hallucinations (Siegel & Wason 1986; Committee on Drugs 1994). Initial symptoms of camphor toxicity may begin within 5 to 15 minutes of ingestion. Camphor presents a clear risk to humans, even from non-oral exposure in the case of infants. Subcutaneously administered pulegone (isomer unspecified) induced seizures in mice at the lethal dose of 1,709 mg/kg (Wenzel & Ross 1957), but it is not clear whether the seizures and deaths were related. Since (1R)-(+)-β-pulegone is the major constituent of pennyroyal oil (61–87%), it was probably responsible for the seizures suffered in four cases of attempted abortion through pennyroyal oil ingestion (Wingate 1889; Kimball 1898; Holland 1902; Early 1961). The amounts taken were substantial, including one teaspoon (~ 5 mL) in one case, and 30 mL in another, but there were no fatalities. The risk of seizures from pennyroyal oil is likely to be limited to oral overdose. McPherson (1925) commented that seizures were an unusual but possible symptom of eucalyptus poisoning. Mack (1988) similarly stated that seizures were possible, and were more common in children than adults. Seizures are in fact very rare with eucalyptus, even after ingestion of large quantities. In a case reported by Witthauer (1922), a 39-year-old male had tonic-clonic seizures after swallowing 26 mL of eucalyptus oil. However, the report expresses reservations as to whether the liquid ingested was in fact eucalyptus oil, as there were symptoms unusual in eucalyptus oil poisoning, including dilated and fixed pupils, and cyanosis. The author commented that, in the previous 30 years, there had not been a reported case of eucalyptus oil causing convulsions. In eleven early case reports, none had seizures. Five of these were adults (Myott 1906; Kirkness 1910; Winterbotham 1914; Gibbon 1927), four were children aged 6–16 years (Neale 1893; Benjamin 1906; Foggie 1911; Sewell 1925), and the remaining two were 20 months and 33 months (Orr & Edin 1906; Allan 1910). In four later cases, all in children of 3–7 years, none had seizures (Craig 1953; Patel & Wiggins 1980). Similarly, there were no seizures among nine cases of eucalyptus nose drop instillation in children aged 1–36 months, 42 cases of eucalyptus oil ingestion in children under 14 years, and 109 children aged from 2 weeks to 9 years (Melis et al 1989; Webb & Pitt 1993; Tibballs 1995,). Out of 14 further cases, including nine children, one suffered seizures with a fatal outcome: an 8-month-old child who ingested 30 mL of eucalyptus oil (Spoerke et al 1989). No description of the seizures is given. We could find only three further reports involving convulsions. In the first, an 11-month-old boy had 10–15 mL of eucalyptus oil spilled onto his face and into his mouth. On arrival at hospital he was given oxygen, and shortly after experienced a ‘short-lived, generalized convulsion’. He eventually recovered (Hindle 1994). In the second, a 4-year-old girl, with no previous history of seizures, suffered a grand mal convulsion lasting less than 1 minute. Earlier that day her mother administered, for the first time, 40 mL of an OTC head lice treatment containing 11% eucalyptus oil, which was washed out after 10 minutes, as directed. A hair conditioner with 2.5% eucalyptus oil was then applied and left on the head. Three hours later the girl felt nauseated and lethargic, and the convulsion followed. She recovered rapidly after the conditioner was washed off (Waldman 2011). Burkhard et al (1999) present the case of a healthy 12-month-old girl who was given five prolonged baths containing an unknown quantity of eucalyptus, pine and thyme oils over a 4-day period. Shortly after the last bath she had a tonic convulsion lasting for 1 minute, and two similar episodes occurred the same day. Over the following days the number of episodes increased to a maximum of 133 in 24 hours. After 4 weeks, seizure activity ceased while she was being treated with phenobarbital and phenytoin, but there were further episodes several months apart. If a large amount of essential oil was used, the frequency of the baths and the susceptible age of the child could possibly explain the seizure activity. It is not possible to determine what caused the seizures, but a strong predisposition seems likely. In many of the above cases, very large quantities of eucalyptus oil were taken (up to 45 mL) and there were several fatalities. In spite of this, seizures were reported in only four of 192 cases (excluding the doubtful Witthauer case). All four were children of less than 5 years, and large amounts (10–15 mL and 30 mL) were ingested in two cases. Therefore, we might conclude that seizures can occur in 2% of young children after intensive exposure to eucalyptus oil. The Waldman (2011) case is atypical, but confirms that eucalyptus oil can cause CNS problems in young children. If eucalyptus oil was convulsant, suspicion would naturally fall on its major constituent, 1,8-cineole. In in vitro studies, hyssop, sage, camphor, and thuja oils (Steinmetz et al 1985) as well as calamus oil (Dhalla et al 1961) and 1,8-cineole (Steinmetz et al 1987) were all shown to modulate the cellular respiration (calcium and potassium levels) of rat brain slices, 1,8-cineole slightly more so than camphor. This action is thought to possibly correlate with the potential to induce seizures. According to Burkhard et al (1999), the Steinmetz et al results demonstrate that both camphor and 1,8-cineole share the same potential to affect brain cell calcium and potassium levels as does the known convulsant, pentylenetetrazol (PTZ). However, in cat brain, during PTZ-induced seizures, potassium levels increased (Heinemann & Louvel 1983), while in the Steinmetz study they did not. In addition, the pathophysiology of PTZ-induced seizures involves many more mechanisms than calcium and potassium gradients (Ahmed et al 2005) as does the pathophysiology of seizures in general (Kovacs et al 2005). Therefore the Steinmetz et al (1987) report is not evidence that 1,8-cineole is a convulsant. It is notable that, while CNS depression and coma are seen in LD50 tests in rats given 1,8-cineole, convulsions are absent (Jenner et al 1964). There is one report of fennel oil inducing an epileptic seizure. A 38-year-old woman developed a typical generalized tonic-clonic seizure lasting 45 minutes, 2 hours after eating five or six cakes containing an unknown quantity of fennel oil. She was an epileptic patient, and took 300 mg/day of lamictal (lamotrigine) to control her seizures (Skalli & Bencheikh 2011). Sweet fennel oil contains 0.2–8.0% of (+)-fenchone, and Wenzel & Ross (1957) reported that subcutaneously injected fenchone (isomer unspecified) produced clonic convulsions in mice at 1,133 mg/kg. This is equivalent to a human sc injection of 79.3 g of fenchone, or > 991 g of sweet fennel oil. The rat acute oral LD50 for (+)-fenchone was 6,160 mg/kg (equivalent human dose 431 g); even at this level, it did not cause seizures (Jenner et al 1964). This dose of fenchone is equivalent to human ingestion of at least 5.4 kg fennel oil, or 10,000 times the recommended maximum oral dose (Table 4.7). Sweet fennel oil also contains (E)-anethole, at 58.1–91.8%. In toxicity testing, no convulsions occurred in rodents given single, lethal doses of 3.2 or 5 g/kg, or ip doses of 300 mg/kg/day for 7 days (Newberne et al 1999). Similarly, there were no convulsions in rats given single oral doses of up to 1.5 mg/kg sweet fennel oil (Ostad et al 2001). Since fennel oil appears to be devoid of convulsant activity, a likely cause of the seizures in the above case is drug interaction. Lamictal is metabolized by UDP-glucuronosyltransferase (UGT) enzymes, and (E)-anethole significantly enhances the activity of UGT (Rompelberg et al 1993). This could have the effect of metabolizing the drug and clearing it from the system too quickly, leaving the patient vulnerable to her seizures. Craig (1953) reported 16 cases of turpentine oil ingestion, all in children of 5 years or under. There were convulsions in two of these cases, infants of 13 and 14 months, the older child having ingested 4 oz (~ 102 mL), and the younger child an unknown quantity. A fatal case involving convulsions was reported in a child of 11 months, who had been given two teaspoons of spirits of turpentine by her grandmother, who said she thought the baby had worms. The child was distressed before the turpentine was given, she had a temperature of 103 °F, and findings on post-mortem included an enlarged thymus and acute bronchitis. It is therefore possible that the turpentine was not the principal cause of either the convulsions or the fatal outcome (Harbeson 1936). Convulsions are a frequent consequence of wintergreen oil or methyl salicylate poisoning, and there is a general CNS excitation, causing very rapid breathing and heartbeat (Adams et al 1957). In LD50 tests for methyl salicylate, convulsions have been seen in guinea pigs, but not in rats (Opdyke 1978 p. 821–825). Anise oil is reported to have caused seizures in a 12-day-old infant, who had been given ‘multiple doses’ by the parents as a treatment for colic. After admission to hospital the infant recovered rapidly, and had no further seizures (Tuckler et al 2002).4 There are no recorded cases of seizures from peppermint oil, although it has produced convulsions in acute LD50 assays: at 3–5 mg/kg and 2–8 mL/kg (oral/rat), and at 0.5–2.0 mL/kg (ip/rat and ip/mouse) (Eickholt & Box 1965; Mengs & Stotzem 1989). However, there were no convulsions in rats dosed orally with up to 100 mg/kg/day for 90 days, or up to 500 mg/kg/day for 35 days (Mengs & Stotzem 1989; Spindler & Madsen 1992). The maximum recommended human oral dose of peppermint oil is 1.2 mL (1.1 g), which is 116 × less than the minimal dose causing convulsions in rats (assumed to be 2 mL/kg, though it could be greater, as the convulsant threshold in this study was not determined) and 32 × less than the non-convulsant daily dose in rats given for 35 days. Since a peppermint oil with 1.1% (1R)-(+)-β-pulegone and 25% menthone had a 90-day oral NOAEL of 40 mg/kg/day in rats (Spindler & Madsen 1992), it is unlikely that peppermint oil will produce toxic effects from therapeutic use. In peppermint oil, the neurotoxic potential of pulegone and menthofuran may be mitigated by (–)-menthol, which potentiates GABAA receptor-mediated responses (Hall et al 2004). It has been said that ketones in general are highly stimulant to the CNS, and therefore a risk to vulnerable groups (Franchomme & Pénöel 1990). Table 10.3 summarizes seizure data from Wenzel & Ross (1957), which may be the basis of this assertion. The constituents in the second (LD50) column were convulsant, but were also lethal at about the same dose. Wenzel & Ross suggest that lethality in a group of terpenoid ketones correlates inversely with ease of conversion to less toxic alcohols and their glucuronides. However, this model would need to be refined if predictions were to be made about other ketones, and there are doubts about some of these findings. Since this report concerned subcutaneous administration in mice, the convulsant doses and the relative risk of each compound cannot be extrapolated with confidence to human oral or inhalation use. For example, in oral LD50 testing, no convulsions occurred in rats administered lethal doses of carvone (isomer unspecified) (LD50 1,640 mg/kg), piperitone (isomer unspecified) (LD50 3,550 mg/kg), or α- + β-ionone (LD50 4,590 mg/kg) (Jenner et al 1964, Opdyke 1978 p. 863–864). Therefore, oral administration of these compounds does not appear to cause seizures. Similarly, De Sousa et al (2007) found neither isomer of carvone to be convulsant in mice by ip administration, and (S)-(+)-carvone was anticonvulsant. While there is a known human convulsant potential for various isomers of pinocamphone, thujone, camphor and pulegone, we could find nothing to substantiate a convulsant effect for the other ketones listed in Table 10.3. In other research, thymoquinone protected mice against PTZ-induced seizures through a GABA-mediated mechanism (Hosseinzadeh & Parvardeh 2004), (Z)-jasmone potentiated GABAA receptor-mediated responses, and therefore is likely to be anticonvulsant (Hossain et al 2004), and valeranone is a CNS depressant, being sedative, hypotensive and hypothermic (Houghton 1988). In almost all the recorded cases where seizures have been induced by essential oils, most of them apparently in non-epileptics, the oils were taken orally, though in a few cases very young children have proved vulnerable to non-oral exposure. A review of olfactory stimulation of seizures in epileptics concluded that, ‘Reflex epilepsy caused by scent in the narrowest sense of the word is very uncommon. It depends on the degree of oversensitivity of the individual to olfactory stimuli’ (our translation from the original German) (Nedbal 1967). In addition to those who are known epileptics, there is a potential risk to those with a low convulsive threshold, i.e., a potential for epilepsy that has not yet declared itself. Essential oils with a convulsant potency requiring dose limitation are listed in Table 10.2. Most of these are also listed in Table 11.1, which addresses pregnancy, breast-feeding and infants. In summarizing convulsant essential oils and the constituents responsible, Burkhard et al (1999) list 11 essential oils as ‘powerful convulsants’. Six of these (hyssop, rosemary, sage, tansy, thuja and wormwood) contain camphor, pinocamphone and/or thujone, and so may present some risk. Pennyroyal oil is a risk if taken in overdose, as is wintergreen oil (not mentioned by Burkard et al). However, Burkhard et al make no reference to dose or mode of administration, and no consideration is given to chemotypes of, for example, hyssop or rosemary oils. Although some evidence is presented for listing the other four oils (eucalyptus, fennel, savin and turpentine) this is often tenuous. For example, Spinner (1920) is cited as evidence of convulsions for both eucalyptus and savin oils, but this is a review paper, and no original cases are reported. Similarly, no hard evidence is presented for fennel oil. Turpentine and eucalyptus oils have already been discussed and only present a limited risk. It is noteworthy that anticonvulsant activity (against PTZ-induced seizures) has been documented for two essential oils with 1,8-cineole and α-pinene as major constituents: Psidium guyanensis (40.5% 1,8-cineole, 13.9% α-pinene) and Laurus nobilis, (typically ~ 40.8% 1,8-cineole, ~ 11.5% α-pinene) (Neto et al 1994; Santos et al 1997; Sayyah et al 2002a). Many of the essential oils listed in Table 10.2 have not been tested for convulsant activity, but it is likely that, for example, Western red cedar oil presents a convulsant risk, since it contains 69–99% of α- + β-thujone. On the other hand, lavandin and spike lavender oils contain greater quantities of (anticonvulsant) linalool than (convulsant) camphor, so risk here may be negligible. It is noteworthy that a methanol extract of Lavandula stoechas protected mice from PTZ-induced convulsions (Gilani et al 2000). This extract would contain the essential oil constituents camphor, fenchone, 1,8-cineole, and α-pinene (Ristorcelli et al 1998). Lavandula stoechas is listed as a ‘well-known anticonvulsive drug’ in Iranian medicine, along with clove oil, valerian oil, anise, fennel seed, caraway, calamus, rue, melissa and lovage (Gorji & Khaleghi Ghadiri 2001). As in all areas of toxicology, there is an element of uncertainty in extrapolating the data for a single constituent to an essential oil containing it, since other constituents may interact in a positive or negative way. In one report, four eugenol-rich Ocimum gratissimum oils, one from each season, were tested against MES-induced seizures. The ‘Spring’ oil was the most effective, and this correlated with a higher content of each of the following sesquiterpene minor constituents: β-selinene, β-caryophyllene, α-selinene, β-elemene, germacrene A and germacrene D. However, there was no correlation with eugenol content (Freire et al 2006). Conversely, in a similar report on four Salvia libanotica oils, there was a strong correlation between degree of seizure activity and content of camphor and thujones (Farhat et al 2001).
The nervous system
Neurotoxicity
Structural changes
Phytol
Functional effects
CNS stimulant activity
Convulsants
Constituent
Dermal
Oral
Camphor isomers
4.8%
2 mg/kg
Methyl salicylate
5%
5 mg/kg
Pinocamphone isomers
0.25%
0.1 mg/kg
β-Pulegone
1%
0.5 mg/kg
Thujone isomers
0.25%
0.1 mg/kg
Pinocamphone
Hyssop oil
Thujone
Sage oil
Thuja oil
Wormwood oil
Camphor
Pulegone
Eucalyptus oil
Fennel oil
Turpentine oil
Other reported cases
Animal data
Discussion
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
Basicmedical Key
Fastest Basicmedical Insight Engine
