Arylcyclohexamines (Ketamine, Phencyclidine, and Analogues)

Abstract

Arylcyclohexamines (also known as arylcyclohexylamines) are a group of compounds that contain a cyclohexamine unit with an aryl moiety, typically a phenyl ring, attached to the same atom to which the amine group is linked (see Fig. 1). They all exhibit dissociative effects due to their antagonism of N-methyl-d-aspartate (NMDA) receptors, and many have been studied as alternatives to traditional anesthetic agents but subsequently abused recreationally in pursuit of these same effects. The most well-characterized arylcyclohexamines are phencyclidine ((1-(1-phencyclohexyl) piperidine; PCP)), ketamine, and, of the novel analogues, methoxetamine.

Arylcyclohexamines (also known as arylcyclohexylamines) are a group of compounds that contain a cyclohexamine unit with an aryl moiety, typically a phenyl ring, attached to the same atom to which the amine group is linked (see Fig. 1). They all exhibit dissociative effects due to their antagonism of N-methyl-d-aspartate (NMDA) receptors, and many have been studied as alternatives to traditional anesthetic agents but subsequently abused recreationally in pursuit of these same effects. The most well-characterized arylcyclohexamines are phencyclidine ((1-(1-phencyclohexyl) piperidine; PCP )), ketamine, and, of the novel analogues, methoxetamine.

Fig. 1

Chemical structures of the primary arylcyclohexamines: phencyclidine , ketamine , and methoxetamine

The main effects of this class of agents are nystagmus, hypertension, and tachycardia in a behaviorally disturbed, and possibly violent, patient. Variation in their potencies, tachyphylaxis in regular users, and the possibility of their misrepresentation as alternative arylcyclohexamines means that prognostication in these patients is difficult. While they can also cause seizures, coma, and death, the majority of patients presenting with acute arylcyclohexamine toxicity will survive with appropriate supportive care.

Historical Perspective

The first arylcyclohexamine to be described was 1-(1-phenylcyclohexyl) amine, in 1907, almost 50 years before the synthesis of PCP [1]. Other analogues reported in the early 1950s included N-ethyl-1-phenylcyclohexylamine (PCE) and 1-(1-phenylcyclohexyl)morpholine (PCMo). While many were not explored further by their initial investigators, chemists at Parke-Davis pursued these and other derivatives in the ensuing years in the hope of developing new anesthetic agents. Compounds such as 1-[1-(thiophen-2-yl)cyclohexyl]piperidine (TCP) were investigated in clinical trials [2, 3]; however, none proved to be effective as an anesthetic due to their prominence of emergence delirium [4].

Phencyclidine , the first well-known arylcyclohexamine, was first synthesized by the Parke-Davis chemist Victor Maddox in 1956. While these experiments were not reported in the literature until nearly 10 years later [5], the pharmacology of phencyclidine had by then already been described [6, 7]. Initially promising human trials appeared to replicate the anesthetic effect observed in earlier animal studies, of potency in anesthesia, and an absence of respiratory depression, and PCP was approved by the US Food and Drug Administration (FDA) in 1957 under the trade name Sernyl [3, 8]. However, its adverse effects – including agitation, psychosis, and catatonia – were soon found to be more frequent and severe than anticipated, and Sernyl was withdrawn from clinical use in 1965 [9], though it was reintroduced to the veterinary market in 1967 as Sernylan [10].

Concurrent research by another Parke-Davis scientist, Calvin Stevens, led to the synthesis in 1962 of 2-chlorophenyl-2-methylamino-cyclohexanone, or ketamine [11–13]. The first human administration of ketamine was conducted in 1964 as part of a trial in 20 prisoner volunteers in the USA; it demonstrated that ketamine was an equipotent anesthetic agent to PCP with a reduced side effect profile, in particular with a reduced incidence of emergence delirium [9, 11]. The study, led by Edward Domino at the University of Michigan, was the first to propose the use of the term “dissociative” in describing ketamine’s anesthetic effects. It was patented and subsequently marketed as Ketalar, having gained approval from the FDA in 1970 [3].

Since then, ketamine has evolved into a diversely utilized drug. Initially gaining widespread use in both human and veterinary anesthesia, its prevalence of emergence phenomena – though less than with PCP – has still largely restricted its human use to specific populations, in particular pediatric anesthesia and field medicine, where its hemodynamic stability and analgesic properties are highly valued. In veterinary medicine, ketamine remains the most widely used anesthetic across all animal species [14]. As further research into this multifaceted drug has progressed, its spectrum of human clinical use has also expanded. Known for decades as a potent analgesic, its use has recently become more common in the treatment of neuropathic pain, complex regional pain syndromes, and as an adjunct in postoperative pain relief [15–17]. While still somewhat experimental, it is also being researched as a treatment for severe depression and for refractory status epilepticus [18–27], and recent observational studies have suggested a potential role for ketamine as a rescue treatment for difficult-to-sedate patients with severe behavioral disturbance [28].

Recreational Use

Phencyclidine and ketamine appeared on the recreational drug scene at a similar time, with the first reports of recreational use of each drug occurring within a few years of each other in the 1960s [29–33]. First reported on the West Coast of the USA, the use of PCP spread rapidly across the USA such that in 1975 its use was described as an “epidemic of drug-induced schizophrenia,” owing to its overdose being responsible for more psychiatric admissions than alcohol abuse and schizophrenia combined [29, 30]. Between 1975 and 1977, the US Drug Abuse Warning Network (DAWN) reported a doubling in the number of PCP-related emergencies and deaths, while 1978 data from the US National Institute of Drug Abuse (NIDA) reported that 13.9 % of young adults (aged 18–25 years old) had used PCP [34]. Complicating its surveillance was its presence as an adulterant in up to 30 different street drugs, PCP being most commonly represented as marijuana [35]. One report of street drugs found on analysis to be PCP demonstrated less than 10 % of samples had actually been sold as PCP [36]; more recently, PCP has been found in tablets sold as methylenedioxymethamphetamine (MDMA or “ecstasy”) [37].

It was not until April 1978, when all manufacture of PCP was prohibited in the USA under the Comprehensive Drug Abuse Prevention and Control Act of 1970, that its use began to decline [10, 38] and the subsequent cocaine epidemic of the 1980s eclipsed PCP use [38, 39]. In 1979, PCP was classed as a Class A drug of the UK Misuse of Drugs Act 1971 (the UK Misuse of Drugs Act 1971 prohibits the activities of Controlled Drugs in relation to their manufacture, supply, and possession; they are graded broadly into three classes, and penalties applied, based on the “harmfulness attributable to a drug when it is misused”) [40]

• Class A (includes): cocaine, heroin, lysergic acid (LSD), methadone, MDMA, morphine, PCP, and class B substances when prepared for injection

• Class B (includes): oral amphetamines, cannabis, codeine, ketamine

• Class C (includes): buprenorphine, benzodiazepines, tramadol, zolpidem

and made a Schedule 2 drug under the UK Misuse of Drugs Regulations 2001 (the UK Misuse of Drugs Regulations 2001 defines the classes of person authorized to supply and possess controlled drugs and defines the conditions under which they may be supplied, possessed, and prescribed [41]),

• Schedule 1: possession and supply are prohibited except in accordance with Home Office authority

• Schedule 2: are subject to full requirements of controlled drugs, including requirements for prescribing, safe custody, and the need to keep registers

• Schedule 3: are subject to prescription and safe custody requirements, however registers do not need to be kept

• Schedule 4: are not subject to prescription or safe custody requirements

• Schedule 5: are exempt from the above controlled drug requirements

as it was thought at the time that it might retain some medical indications; whilst not currently used as a pharmaceutical, PCP remains in Schedule 2 in the UK.

In 1971, the first account in the literature of ketamine’s psychedelic effects was published in the New England Journal of Medicine [42]. Though recreational ketamine use was initially highest among healthcare professionals and others who had easy access to the drug [43], this changed in the 1980s and 1990s as it became increasingly popular among clubbers, and in particular on the rave scene, where it initially appeared as an adulterant to ecstasy tablets [44].

Despite early concern from the FDA about the misuse of ketamine, a request for control of its use from the US Department of Health and Services was rejected by the DEA [45], and it was not until 1995 that ketamine was added to the US “emerging drugs” list and eventually placed into Schedule III of the US Federal Controlled Substances Act (CSA) in 1999. In the UK, it was not until 2006 that ketamine was placed into Schedule 4 and added to Class C of the Misuse of Drugs Act, following evidence of continued abuse [40]; in June 2014, it was made a Class B drug and placed into Schedule 2 [46].

In the USA, increased control did not appear to be associated with a decline in ketamine use; this may, at least in part, relate to the burgeoning Internet trade providing a direct supply, particularly from sources in Hong Kong and in Mexico [47]. Today, most illicit ketamine originates from diverted pharmaceutical production or from illicit production in countries such as Mexico, China, India, and European states [48]. Internet trading sites remain key in its distribution, the now obsolete Silk Road website having hosted numerous ketamine vendors.

First-Generation Arylcyclohexamine Analogues

Between 14 and 30 analogues of PCP have been reported, and while most of these did not reach a wide audience, the use of three became prominent: PCE, TCP, and 1-(1-phenylcyclohexyl)pyrrolidine (PCPy) [3]. PCE and TCP were the first of the PCP analogues to be reported, in street samples from Los Angeles in 1969 and 1972, respectively, with the first report of PCPy from Maryland in 1974 [36, 49, 50]. By 1975, TCP had been reported in 25 US states [30, 36, 51] and was the first of the arylcyclohexamines to be placed into Schedule I of the US CSA (the US Controlled Substances Act encompasses five Schedules of controlled substances (I–V) based on three criteria [52]. In 1975, after it was deemed to be of no medical value and of high abuse potential [45]:

• (a) A drug or substance’s potential for abuse

• (b) A drug or substance’s role in currently accepted medical use in treatment in the USA

• (c) A drug or substance’s potential to lead to psychological or physical dependence

PCE and PCPy followed suit in 1978, both having been associated with increasing use and fatalities [45, 53, 54]. Anecdotal reports suggest that, despite their scheduling, all three analogues remained available throughout the 1980s, with the last reported seizure of PCE in 1991 [3]. The synthesis of TCP, meanwhile, has been described as recently as 2004 on an online drug forum [55]. Though their potencies were thought to be similar, users describe TCP and PCE as more potent than PCP, with TCP having a longer duration of action [3], while PCPy is reported to be roughly equipotent to PCP [4, 56].

The ketamine analogue 2-(ethylamino)-2-thiophen-2-ylcyclohexan-1-one (tiletamine) was developed by Parke-Davis for use in veterinary anesthesia and was patented in 1966 [57]. Despite being placed in Schedule III of the US CSA in 1987, there are reports of tiletamine misuse in the USA and South Korea, including two fatalities from its overdose [58, 59].

Other compounds, such as the PCP precursor 1-piperidinocyclohexanecarbonitrile (PCC), have relevance due to their detection as contaminants. An intermediary in the synthesis of PCP, the presence of PCC indicates an incomplete synthetic process and has been found as a contaminant in up to 33 % of illicit PCP samples [49, 60, 61]; reports of an incorrectly synthesized batch of PCP was reported to have caused abdominal cramps, vomiting, coma, and even death [62]. Two studies in mice have shown conflicting data: one showed PCC to be roughly equipotent to PCP, while the other found PCC to be almost twice as toxic as PCP, with an LD₅₀ of 155 μmol/kg compared to an LD₅₀ of 283 μmol/kg for PCP [63, 64]. This toxicity is thought to be due to the release of hydrogen cyanide [61, 63, 65].

Novel Arylcyclohexamine Analogues

More recently, clandestine chemists have been producing novel analogues of phencyclidine and ketamine. A true example of the flourishing “designer drugs” scene, the theoretical effects of the ketamine analogue methoxetamine (MXE ), or 2-(3-methoxyphenyl)-2-(ethylamino)cyclohexanone (see Fig. 2), were described on online forums in 2006, at least 3 years before its first known synthesis [3]. It was not until May 2010 that the first public report of MXE’s effects was described online, and only a few months later, it was marketed via “research chemical” websites as a “bladder friendly” alternative to ketamine (see below). First reported to the Early Warning System at the European Monitoring Centre for Drugs and Drugs Abuse (EMCDDA) by the UK in November 2010 [48], availability escalated quickly, with up to 58 websites offering MXE by July 2011 [66]. The first report of MXE toxicity in the scientific literature was published in August 2011 [67], with further reports of acute MXE toxicity and MXE-related fatalities subsequently reported [68–74]. In April 2012, MXE became the first drug in the UK to be banned under a Temporary Class Drug Order (TCDO) [75, 76]; later that year, the UK Advisory Council on the Misuse of Drugs (ACMD) formally recommended that MXE was controlled, and in February 2013, it was placed into Schedule 1 as a Class B drug under the UK Misuse of Drugs Regulations 2001 and Misuse of Drugs Act 1971, respectively [75, 77].

Fig. 2

Chemical structures of the arylcyclohexamine analogues

Further methoxylated arylcyclohexamine analogues have emerged in addition to MXE. While they have appeared on the recreational drug scene only recently, their initial descriptions date back to the first description of PCP, with 2- and 4-methoxy-phencyclidine (2-MeO-PCP and 4-MeO-PCP [also methoxydine]) described by Maddox in his 1965 report [5]; 14 years later, 3-methoxy-phencyclidine (3-MeO-PCP) was described by Geneste et al. [78]. Little was heard of these compounds until their effects were described on an online forum by Beagle in 1999 [79, 80]; however, this site was taken offline in 2004. They reemerged again in December 2008, when 4-MeO-PCP appeared for sale online in the UK, with reports suggesting its varied potency to be due to impurities between batches and leading to re-experimentation with 3-MeO-PCP, which was made available online in April 2011 [3]. 4-MeO-PCP was first reported to the EMCDDA in January 2011 by Finland, with its first involvement in a polydrug intoxication reported in August 2011 from Norway; 3-MeO-PCP was first reported to the EMCDDA in March 2012 by the UK [81].

The ban on MXE left a void in the UK designer drug, or “research chemical,” scene which was quickly filled by other aryl-aminocyclohexanone analogues. In May 2012, 2-(2-methoxyphenyl)-2-(methylamino)cyclohexanone, or 2-methoxy-ketamine, came on to the market followed closely by N-ethylnorketamine (N-EK) in August of the same year [3]; they were reported to the EMCDDA in August 2012 by Sweden and in September 2012 by the UK, respectively [82]. Their potency was reported by users to be low, with a wide variation in their effects. Although other arylcyclohexamine analogues exist (see Fig. 2), including some of those mentioned here, most are poorly characterized and hence not discussed further.

Epidemiology of Recreational Use

Today, recreational use of arylcyclohexamines continues throughout the world. Without clear reason, PCP remains most popular in the USA and Canada – indeed, its use is almost exclusive to North America – with 2013 data showing 6.5 million Americans (2.5 %) aged 12 and older to have used it in their lifetime, compared with 2.7 million (1.0 %) for ketamine [83]. Ketamine is more widely used in the UK, however, and overall it retains a much more international presence on the recreational drug market; the 2015 UN World Drug Report reported its use in 58 countries [84].

In both the USA and the UK, the use of arylcyclohexamines is more common among males than females [83, 85, 86]: in the UK, 3.6 % of males report lifetime ketamine use compared with 1.3 % of females; in the USA, this is 1.3 % compared to 0.8 % [83, 87]. Available data from the UK shows that ketamine use is higher in gay or bisexual males (4.2 %) than in gay or bisexual females (1.5 %) and heterosexual adults (0.4 %) [88]. Like many other recreational drugs and novel psychoactive substances, the use of arylcyclohexamines is more common among nightclub attendees [89–91].

Phencyclidine and Ketamine

Despite the decline of PCP use in the 1980s after its ban in 1978, it has remained a consistently available street drug in the USA, though its use reached a nadir in 1984 [92]. Data from the 2013 US National Survey on Drug Use and Health (NSDUH) shows that lifetime use of PCP in the USA has remained relatively constant over the past decade, at around 2.5 % of the population [83]. Of concern, however, is data from the DAWN, which showed a fivefold increase in the number of PCP-related presentations to emergency departments in the USA between 2005 and 2013, from 14,825 to 75,528 [93]. However, NSDUH data showed a decline in PCP use between 2012 and 2013, with past year initiates and prevalence of last year use falling from 90,000 to 32,000 and 172,000 to 90,000 (0.1 % of the population), respectively. In contrast to the diversion of ketamine from overseas suppliers, manufacture of PCP has remained localized, providing one explanation for the sudden decline in use between 2012 and 2013: the discovery and bust of the largest-ever PCP laboratory on US soil in Los Angeles in February 2012 [94]. Of concern, despite the significant fall from 2012 numbers, is that in 2013 more than half (19,000 of 32,000) of past year initiates of PCP use were aged between 12 and 17 years.

Between 2006 and 2013, lifetime ketamine use in the USA has remained steady, at 0.9–1.0 % of the population [83], though there was a significant increase in the lifetime use among those 26 years of age and older (0.7–1.1 %) and a significant decrease in lifetime use among those aged 18–25 years (2.8–1.5 %) and those aged 12–17 years (0.3–0.2 %). Despite the overall lifetime use of ketamine remaining steady, there has been a similar increase in US emergency department visits due to ketamine, from 303 in 2005 to 1,550 in 2011 [93]. Further, while lifetime PCP use remains two-and-a-half times higher than lifetime ketamine use in the USA, the most recent data from NIDA’s Monitoring the Future Study shows a trend among school-aged users toward ketamine rather than PCP, with 1.5 % and 0.8 % of 12th graders reporting last year use of ketamine and PCP, respectively, in 2014 [95].

In the UK, data from the Crime Survey of England and Wales (CSEW) showed last year use of ketamine in those aged 16–59 years increased from 0.4 % to 0.6 % between 2012/13 and 2013/14, with a significant increase in use seen in those aged 16–24 years (0.8–1.8 %) [86]. Lifetime use among 16–59-year-olds increased from 1.3 % to 2.7 % between 2006/07 and 2013/14. Among self-reported substance users in the 2014 Global Drugs Survey, 5.7 % of 78,819 global respondents reported last year use of ketamine; this was highest in the UK (19.8 % of 7,326 respondents), while it did not make the list of top 20 drugs used in the USA among its 6,500 respondents [96]. Data from the European Drug Emergencies Network showed ketamine accounted for 2.0 % of the 5,529 reported presentations of acute recreational drug toxicity to 16 emergency departments in 10 European countries [91]; 0.5 % of self-reported ketamine users in the 2015 Global Drugs Survey had sought emergency medical treatment in the past 12 months [97].

In Hong Kong and China, ketamine has for over a decade been the most commonly used psychotropic drug – and second most common drug of abuse overall, behind heroin – since overtaking ecstasy in 2001 [98]. Among reported drug abusers (0.14 % of the total population), 28.2 % reported ketamine use in 2013, down from a peak of 37.9 % in 2009 [99]. Its use is particularly evident among drug abusers under the age of 21 and among newly reported drug abusers; since 2001, in both groups, ketamine has been the most widely reported drug of abuse: in 2013 representing 52.5 % (from a peak of 85.5 % in 2008) and 46.4 % of users, respectively [98, 99]. Similarly to US and UK data, ketamine use was more than twice as common in males, who represented 73 % of ketamine users.

Arylcyclohexamine Analogues

There is no population level data on the prevalence of use of MXE from surveys such as NSDUH in the USA or CSEW in the UK. Figures regarding the use of MXE are therefore limited to targeted, nonrepresentative surveys among specific subpopulation groups. In 2011, the annual online Global Drugs Survey reported last year use of MXE in 4.2 % of 7,700 UK respondents [100]; in an in situ survey of nightclub attendees in South London, UK, the prevalence of last year use of MXE was 6.4 % [90]. Data from the UK National Poisons and Information Service (NPIS) showed a total of 345 inquiries regarding MXE between April 2010 and August 2012. There was a significant decrease in the number of inquiries to the NPIS in the 3 months following MXE receiving a TCDO in April 2012 compared with the 3 months preceding the TCDO, with the total number of inquiries falling from 166 to 35 [85].

The only published data on the use of other novel arylcyclohexamine analogues are for 3- and 4-MeO-PCP, from the Swedish STRIDA project [101]. Between January 2010 and June 2013, only five inquiries were made to the Swedish Poisons Information Centre relating to “PCP” intoxication (none of these were analytically confirmed). In the 21-month period from July 2013 to March 2015, of a total of 2,687 consultations for suspected intoxication with a new psychoactive substance (NPS), 80 cases (3.0 %) were registered as 3-MeO-PCP, 4-MeO-PCP, or PCP intoxication. Blood and/or urine samples were available for analysis in 30 (37.5 %) of these cases: 27 (90.0 %) were confirmed to involve 3- and/or 4-MeO-PCP; none tested positive for PCP. A total of 1,243 cases during the 21-month study period were analyzed for the presence of an NPS. 3-MeO-PCP and 4-MeO-PCP were detected in 56 (4.5 %) and 11 (0.9 %) of cases, respectively; both substances were found in 8 (0.6 %) of cases [101]. The total of 59 cases in which either 3- and/or 4-MeO-PCP were detected suggests there is the potential that, like PCP and ketamine before them, they may also exist as adulterants to other drugs.

Chemistry and Clinical Pharmacology

The structure of arylcyclohexamines consists of a cyclohexamine unit with an aryl moiety, typically a phenyl ring, attached to the same atom to which the amine group is linked (see Fig. 1). Their characteristics are summarized in Table 1.

Table 1 Characteristics of the major arylcyclohexamines

Chemical Structure

Ketamine and methoxetamine differ from phencyclidine in a number of ways (Tables 2 and 3 and Figs. 1 and 2) [76]. Unlike phencyclidine, ketamine has a chiral center and exists as two optical enantiomers that exhibit different properties. The S(+)-enantiomer has more potent analgesic effects, owing to its fourfold greater affinity for the NMDA receptor, while the R(−)-enantiomer is associated with more agitation due to its more prominent postsynaptic effects [103–105]. Studies of ketamine anesthesia for surgery in rats have also demonstrated the R(−)-enantiomer to be responsible for an increased incidence of emergence reactions than either racemic or S(+)-ketamine [106–108]. Differences in the metabolism of the optical isomers and their metabolites’ subsequent biodisposition have also been reported; however, the significance of this has not been well characterized [103]. Methoxetamine also exists as two optical enantiomers; however, it is not known whether there are significant differences in their pharmacokinetic or pharmacodynamic properties.

Table 2 Structural differences between ketamine and phencyclidine and pharmacodynamic effects

Table 3 Structural differences between methoxetamine and phencyclidine and pharmacodynamic effects [76]

Although, as discussed earlier, there are many other arylcyclohexamine analogues that have been used recreationally, there is limited data available on their pharmacology.

Pharmacokinetics

Phencyclidine is rapidly absorbed from the respiratory and gastrointestinal tracts, with an onset of action that is quickest after intravenous (i.v.) or inhalational routes (2–5 min) and slowest following oral use (30–60 min) [109, 110]. While oral bioavailability of up to 70 % is reported, the typical dose to achieve sedation equivalent to 0.25 mg of i.v. PCP is up to 20 times higher for the oral route (1–5 mg) [38]. Unlike PCP, ketamine is not well absorbed by the oral route and undergoes significant first-pass metabolism, with an oral bioavailability of less than 20 % [111–113].

Arylcyclohexamines are weak bases with a pKa of 7.5–8.5 and molecular weights ranging from 238 to 247 kDa. PCP has a larger volume of distribution than ketamine because it is more lipophilic, and this contributes to its longer half-life and duration of action of between 7 and 46 h, compared with a half-life of 3 h for ketamine [38, 109]. Both PCP and ketamine are highly lipid soluble, enabling efficient distribution to the central nervous system (CNS), and water soluble, allowing them to be administered by a variety of routes. The onset of action of ketamine and MXE, as for PCP, is rapid following i.v. and intramuscular (i.m.) administration, with effects apparent after 1 and 5 min, respectively [114]. In contrast to PCP, the onset of action of ketamine following nasal insufflation is up to 30 min, while users report that for MXE, this is as long as 30–90 min [114]. Duration of effect by different routes of administration is shown in Table 4.

Table 4 Onset and duration of effects (in minutes, unless specified)

All arylcyclohexamines are metabolized by the hepatic microsomal oxidizing system. PCP is metabolized by isoenzymes from the CYP3A and CYP2B subfamilies, primarily by hydroxylation to various inactive metabolites [43, 117, 118]. Metabolism of ketamine is primarily by the CYP2B6 isoenzyme, with contribution from CYP3A4 and CYP2C9 [119].

The main pathway of ketamine metabolism is by N-demethylation to norketamine, its major metabolite and also a noncompetitive NMDA receptor antagonist. Norketamine has one-third the potency of ketamine [38, 120] and, like its parent compound, exhibits chiral properties: its S(+)-enantiomer having greater affinity for the NMDA receptor than its R(−)-enantiomer [121]. Norketamine undergoes hydroxylation at the hexanone ring (this hydroxylation also being a minor metabolic pathway of ketamine itself) and subsequent glucuronidation to more water-soluble metabolites that are excreted in urine. There does not appear to be significant differences in the pharmacokinetics between ketamine’s two enantiomers [104], although some studies have reported a longer elimination half-life for S(+)-ketamine of up to 4–7 h [122].

No human studies have investigated the metabolism of MXE; the little information that is available is based on in vitro studies using liver microsomes and rat models [77, 123]. These showed the predicted metabolism of MXE by CYP 2B6 and CYP3A4, with the main metabolite, normethoxetamine, resulting from N-deethylation and accounting for over 70 % of metabolites; the other major pathway resulted in O-desmethylmethoxetamine and accounted for 14 % of metabolites. As with norketamine, normethoxetamine then undergoes hydroxylation and glucuronidation to form the more water-soluble hydroxynormethoxetamine glucuronide. These metabolites are consistent with those found in urine samples of analytically confirmed MXE ingestions [70, 77, 123]. Like norketamine, normethoxetamine is thought to be pharmacologically active; however, this is unproven.

Known pharmacokinetic properties of the arylcyclohexamines are shown in Table 5.

Table 5 Pharmacokinetic properties of the arylcyclohexamines

Pharmacodynamics

All arylcyclohexamines exhibit dissociative effects due to their noncompetitive antagonism of glutamate receptors of the NMDA subtype [125]. They were initially sought after as an alternative to more traditional anesthetic agents but subsequently abused recreationally in pursuit of these same effects. Ketamine has approximately 10 % of the affinity for the NMDA receptor than PCP [119, 125, 126], while 3-MeO-PCP has been found to have almost three times the affinity of PCP for the NDMA receptor [127]. In addition, the arylcyclohexamines also have effects at many non-NMDA receptors (Table 6).

Table 6 Receptor binding affinity, Ki (nM)

Arylcyclohexamines cause dissociation between the somatosensory cortex and higher brain centers, evoking a cataleptic state in which there is an absence of motor reaction to nociceptive stimuli [128, 129]. Functional magnetic resonance imaging (fMRI) studies have demonstrated a dose-dependent reduction in pain-induced cerebral activation at the secondary somatosensory cortex (S2), the insula, and the anterior cingulate cortex [130, 131]. While cortical effects are predominant, arylcyclohexamines also have significant effects at the spinal level, blocking afferent spinoreticular pathways and enhancing descending inhibitory serotonergic pathways [132–135].

Glutamate-Dependent Mechanisms

Glutamate is the major excitatory neurotransmitter in the mammalian CNS, acting on pre- and postsynaptic receptors located on ion channels [102]. Ionotropic glutamate receptors are classified as either NMDA, activated specifically by N-methyl-d-aspartate, or non-NMDA, activated by other neurotransmitters such as AMPA (α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid) or kainate [136, 137]. NMDA receptors are located on nearly all CNS cells, and in particular on those structures involved with nociception [102]. A specific “PCP binding site” has been identified; of the major arylcyclohexamines, PCP displays the highest affinity for this site, followed by methoxetamine and then ketamine [127] (Table 6).

Activation of NMDA receptors occurs after calcium-dependent exocytosis of glutamate-containing synaptic vesicles [138, 139]. Glutamate is released into the synaptic cleft where it binds postsynaptic NMDA receptors, opening ion channels, and leading to membrane depolarization [140].

The NMDA Receptor

Present on nearly all cells of the CNS, the NMDA receptor is a tetrameric structure consisting of two subunits, of which there are three types: NR1, NR2, and NR3 (alternatively known as GluN1, GluN2, and GluN3, respectively) [102]. Each subunit possesses four hydrophobic segments, M1 to M4, which form the transmembrane domain (TMD); three of these segments (M1, M3, and M4) are truly transmembrane, while the M2 segment faces the cytoplasm and forms the receptor’s ion channel (Fig. 3). As well as the TMD, the NMDA receptor is comprised of two extracellular domains: the N-terminal domain (NTD, or amino terminal domain [ATD]) and the ligand-binding domain (LBD) [102, 139, 141].

Fig. 3

Structure of the tetrameric NMDA receptor. Each of the two subunits (in this case NR1 and NR2) comprise a transmembrane domain (TMD), containing segments 1–4, and two extracellular domains: the ligand-binding domain (LBD) and the amino terminal domain (ATD). The channel-blocking domain (C; site of Mg²⁺ ion voltage-dependent block) and sites of glutamate (Glu) and glycine (Gly) binding are also shown. The PCP binding site is not shown but is located close to the Mg²⁺ binding site (C) within the ion channel [102, 139] (Figure reproduced with kind permission from Prof. David Lodge)

The NMDA receptor is typically comprised of two NR1 and two NR2 subunits, of which seven and four subtypes of each subunit have been identified: NR1 a-g and NR2 A-D. The NR2 subtypes appear to be responsible for the variation in potency between arylcyclohexamines and the pathological processes associated with abnormal NMDA channel function [102, 142–144]. While the NR2A subtype is ubiquitous, the locations of the other subtypes are more localized. The NR2B subtype is found in the anterior parts of the brain, especially the cingulate cortex, hippocampus, amygdala, and olfactory bulb, and is also involved in nociceptive transmission to the thalamus, spinal cord, and extrasynaptic areas such as the primary afferents. The NR2C subtype is prominent in the cerebellum, while the spinal cord is rich in the NR2D subtype. Studies of PCP and ketamine have shown that they are, respectively, four and nine times more potent at NR2B-D receptor subtypes compared with the NR2A receptor subtype [142]. While there are no receptor studies of MXE, there is the potential that, given reports of cerebellar features in patients with acute MXE toxicity [69], it has greater specificity for the NR2C subtype.

NMDA receptors are coupled to calcium ion channels, which are activated by synaptic glutamate binding to the LBD of the NR2 subunit and require co-activation with either glycine (or d-serine) binding to the LBD of the NR1 subunit. The main inhibitory mechanism is the voltage-dependent block of the ion channel by magnesium ions [145]. At resting membrane potential, extracellular magnesium is held within the ion channel by negative electrostatic forces; neuronal depolarization causes magnesium to be released, and in the presence of the co-agonists glutamate and glycine, calcium influx ensues. Predictably, ketamine and magnesium have been shown to have a synergistic antagonistic effect at NMDA receptors [146].

The effects of NMDA receptor antagonism by arylcyclohexamines result in modulation of glutamatergic transmission [147]. Their primary effects, including anesthesia, analgesia, and psychotomimetic properties, relate to this action (see Pathophysiology, below). Other effects thought to be related to the NMDA receptor include antidepressant activity [148] and specific analgesic effects due to action at NMDA receptors in dorsal horn neurons [149].

Non-glutamate/NMDA-Dependent Mechanisms

Arylcyclohexamines also display affinity for a host of other receptors (Table 6). They bind to and inhibit biogenic amine transporters at approximately 10–20 % of their affinity for the NMDA receptor [150], with resultant sympathomimetic and psychomotor effects due to inhibition of catecholamine, serotonin, and dopamine reuptake. It is thought that the effect on dopamine release is especially prominent in the nucleus accumbens [151], with subsequent implications for addiction [152]. In addition to inhibition of amine transporters, arylcyclohexamines have been demonstrated to have varying affinity for D₂ and 5-HT_2A receptors, where they are thought to exert agonist activity [153]. PCP and MXE have greater affinity than does ketamine at the 5-HT_2A receptor, thought to be related to the more pronounced perceptual disorders and hallucinations seen with their use [154].

In overdose, the arylcyclohexamines also stimulate sigma receptors [155], causing an inhibitory effect on cholinergic pathways. It is thought that binding of sigma receptors may play a role in their antidepressant activity [156]. While ketamine has also been shown to bind to opioid receptors, its binding affinity at usual recreational or anesthetic doses is too low to contribute to its analgesic effects [157, 158]. Arylcyclohexamines also display weak affinity for nicotinic and muscarinic cholinergic receptors [157], though their role here is uncertain.

Analgesic Effects

As mentioned, the arylcyclohexamines are weak agonists at opioid receptors; however, at usual anesthetic and recreational doses, this is not thought to contribute to their analgesic effect; further, their analgesic effects are not reduced by the administration of naloxone [160, 161]. Rather, their analgesic properties are modulated by NMDA receptor antagonism in the CNS and in the spinal cord, by decreasing the amplification response to repeated stimulation, or “windup,” that causes CNS sensitization to nociceptive stimuli [102, 122]. Studies of ketamine have also suggested a non-NMDA-dependent role in directly inhibiting nitric oxide synthase, contributing to its analgesic effect [162, 163]. Further, both NMDA- and nitric oxide-dependent mediation of opioid receptors, and in particular the μ-receptor, appear to attenuate analgesic tolerance [164].

The Pharmacology of Tolerance, Dependence, and Withdrawal

The stimulatory effects of the arylcyclohexamines on dopaminergic and serotonergic transmission in the central nervous system, via both inhibition of amine transporters and via direct agonist activity at D₂ and 5-HT_2A receptors, are responsible for their addictive properties [151–153].

While it was initially thought that PCP did not have dependence potential, subsequent animal studies found that monkeys self-administered PCP and appeared to develop withdrawal [165]. These studies have since been replicated, demonstrating the development of dependence and the presence of withdrawal in multiple animal species [166–171]. In a case series of 68 chronic PCP users seeking treatment who had used PCP for a mean of 3.0 years, it was found that both psychological and physical dependence and withdrawal occurred [172]; 25 (36.8 %) had considered themselves to be addicted to PCP. The most commonly reported symptoms were craving (51.5 %), increased need for sleep (48.5 %), poor memory (45.6 %), depression, and laziness (both 44.1 %). Irritability was reported by 30.9 %, increased anxiety by 22.1 %, and headaches and insomnia by 16.2 % and 14.7 %, respectively. However, no details regarding the length or severity of these symptoms were reported.

Ketamine use has been reported to be characterized by binging, often with repeated dosing until exhaustion of a user’s supply [173, 174], and is generally associated with psychological dependence characterized by craving, rather than any physical dependence or withdrawal state [175–177]. In an online survey of 506 mostly Spanish self-reported ketamine users, 40.5 % reported tolerance with ketamine use [178]. The observed tolerance in repeat and chronic ketamine users is thought to be as a result of auto-induction of CYP450 enzymes [177, 179–181].

No studies have examined the dependence or abuse potential of MXE; however, self-reports from users suggest compulsive re-dosing is an issue [48]. Reported withdrawal symptoms of MXE include low mood and depressive thoughts, while one user reported 48 h of insomnia after nasal insufflation of 1,000 mg of MXE [114].

Pathophysiology of Toxic Effects

The effects of the arylcyclohexamines are varied owing to their interactions with numerous receptors, as detailed above. Their predominant effects are on the central nervous system, and their induction of a state akin to catatonia has led to advancements in the pathophysiology of schizophrenia, with NMDA receptor dysfunction as the foundation of the glutamate hypothesis of schizophrenia [147, 182, 183].

Neurological Effects and Pathophysiological Similarities with Schizophrenia

While the clinical effects of the arylcyclohexamines have been apparent from their outset, it has been only recently that their effects on the NMDA receptor have been fully characterized, their schizophrenia-like effects noted, and the glutamate hypothesis of schizophrenia advanced.

In the 1960s and 1970s, observations of dopamine D₂ antagonists, such as chlorpromazine, causing reduced psychotic features led to the development of the dopamine hypothesis of schizophrenia [184]. Characterized by positive symptoms – including delusions, hallucinations, and catatonia – and negative symptoms – of alogia, avolition, and flattened affect – there has been ongoing debate as to the ability of a hypodopaminergic state to account for all these effects [185].

More recently, the hypothesis of glutamate dysfunction as an explicable mechanism for both the positive and negative symptoms of schizophrenia has gained support [182]. While the schizophrenia-like effects of PCP were noted early in its use [11, 29, 186–188], the subsequent elucidation of the mechanisms of the NMDA receptor function has led to further interest in the role of this receptor and its antagonists as a model for studying schizophrenia and possible therapeutic interventions [185, 189]. Studies have shown that arylcyclohexamines produce a dose-dependent transient psychosis-like state closely resembling schizophrenia [147, 183, 190–192], which in some cases is indistinguishable from it [193] and indeed can exacerbate symptoms in those with schizophrenia [191, 194, 195].

These effects appear to be mediated by changes in blood flow seen in the prefrontal, medial frontal, and inferior frontal cortices [196, 197]. Though the fundamental action of arylcyclohexamines in their antagonism of NMDA receptors is reduction in glutamatergic transmission at this receptor site, it has been shown that many of their effects are actually affected by an increase in non-NMDA-mediated glutamatergic activity. This goes some way in explaining the behavioral and neurochemical dysfunctions of arylcyclohexamine use while also providing an explanation for the apparently paradoxical pro-convulsant properties of high doses of PCP despite decreased NMDA-associated glutamatergic activity [198]. That is, the primary antagonism of NMDA receptors leads to decreased stimulation of inhibitory GABAergic interneurons (causing decreased GABA release), with subsequent increase in not only downstream glutamate release but also increased cholinergic and serotonergic activity [199–201]; these effects are seen primarily in the frontal cortex, anterior thalamus, and the superior temporal, posterior cingulate, and parahippocampal gyri. These effects on the pyramidal cells of the posterior cingulate cortex, in particular, may account for the focal neurodegeneration seen with chronic ketamine use due to this over-excitatory effect [199].

Further sequelae of chronic use include dose-dependent loss of gray matter volume and white matter abnormalities in the frontal lobes [202, 203]. Chronic use has also been found to induce regionally selective upregulation of dopamine D₁ receptors in the dorsolateral prefrontal cortex [43]. Further, consistent with the alternate dopamine hypothesis, acute ketamine administration has also been shown to alter the firing of mesocortical and mesolimbic dopaminergic neurons, leading to increased extracellular dopamine in both the striatum and in the prefrontal cortex [204, 205].

Other mechanistic links between the NMDA receptor and schizophrenia have been demonstrated in the signal transduction cascades that mediate cellular growth, differentiation, and survival in proliferating mammalian cells. In particular, the Ras–MAPK pathway has been shown to be involved in NMDA receptor signaling, providing a regulatory role in the synaptic plasticity related to long-lasting changes on memory and addiction [206]. While formal studies of the molecular mechanisms of cognitive deficits due to arylcyclohexamine use are lacking, the similarity between the memory impairment seen in chronic users of ketamine and in schizophrenics suggests that the dysfunction of the MAPK pathway and decreased MAPK1 gene expression in schizophrenic brains [206–208] may also play a role in the cognitive and memory deficits seen with chronic arylcyclohexamine use [209, 210].

Differences Between Acute and Chronic Use

While there have been many reports of the psychotic effects attributable to acute arylcyclohexamine use, differences in the neuropathophysiology in repeated and chronic users have been shown to induce more persisting schizophreniform symptomatology [147].

As mentioned above, increase in blood flow to the frontal cortex is seen after acute ketamine exposure in both healthy volunteers and in schizophrenic patients [196, 197]; in contrast, a reduction in frontal lobe blood flow is seen in long-term abusers of PCP [211, 212], more consistent with the “hypofrontality” observations of cognitive dysfunction associated with reduced blood flow to the frontal cortex in schizophrenia [213–216].

Similarly, there is a difference in the effects on dopaminergic transmission associated with acute and chronic arylcyclohexamine use. Where their acute administration resulted in a dramatic increase in frontal lobe dopaminergic activity [204, 217, 218], long-term administration caused the reverse effect [219, 220], again consistent with effects seen in schizophrenia [221, 222]. The increase in dopamine in the forebrain is responsible for the cognitive dysfunction seen after acute arylcyclohexamine ingestion due to impairment of spatial working memory [223, 224]. Alongside this, the disinhibition of prefrontal glutamatergic transmission caused by NMDA receptor blockade leads to increased downstream glutamatergic stimulation of mesocorticolimbic dopaminergic transmission and of locomotor behavior [225].

The subsequent reduction in dopaminergic activity seen with chronic use of arylcyclohexamines also plays a role in decreased working memory function of the prefrontal cortex [226]. In addition to the cortical effects of NMDA receptor antagonism, studies of subcortical striatal dopamine dysfunction have revealed more about dopamine’s role in the impaired cognitive function seen in schizophrenia and in chronic arylcyclohexamine use. In subcortical areas, it is thought that dopaminergic hyperactivity is in fact associated with chronic use [147]; this has previously been suggested for schizophrenia and would be consistent with the original dopamine hypothesis [227–229].

While the acute administration of arylcyclohexamines increases serotonergic and catecholamine activity via both biogenic amine reuptake inhibition and direct receptor agonism, no changes in their metabolism have been observed after long-term use of PCP [219, 220]. The effects of chronic arylcyclohexamine use on cholinergic pathways are less well studied, though it is thought that a degree of tachyphylaxis occurs [230].

Cerebellar Effects

While the effects of PCP and ketamine on the cerebellum have been demonstrated in vitro in rats, this clinical effect has not been demonstrated in humans [177]. In contrast, cerebellar toxicity was a prominent feature in a three cases of acute MXE toxicity, who presented with incoordination and severe cerebellar signs of ataxia, dysdiadokinesis, and nystagmus [69]. In all of these cases, the route of administration had been by nasal insufflation, with cerebellar toxicity reported within 20–30 min; in two of these, cerebellar features resolved after 16 h; however, in one case, ataxia and nystagmus persisted for 4 days. Toxicological analysis of serum samples of these patients showed methoxetamine concentrations of between 0.16 and 0.45 mg/L; no other drugs, including ethanol, were detected in any of these cases.

Effects on the Cardiovascular System

Pressor Effects

The pressor response typical of arylcyclohexamines has long been observed, with positive effects on blood pressure as well as potentiation of the response to catecholamines, though tachyphylaxis was noted with repeated dosing [6, 7, 231]. While the mechanisms were not known at the time, this led Ilett et al. to postulate a role of PCP on catecholamine stores – rather than direct receptor agonism – which has since shown to be the case [38]. Interestingly, while the direct effect of arylcyclohexamines in isolation is to increase heart rate and blood pressure, some studies have shown that coadministration of other stimulant drugs, in particular cocaine, may ameliorate the sympathomimetic effects of subsequent arylcyclohexamine use [231], further complicating the clinical presentation of those intoxicated with these drugs. Since then, the effects of PCP and ketamine on the cardiovascular system have been well characterized, displaying modest rises in heart rate and blood pressure at usual recreational doses [232–235]; in one case series, the mean heart rate and blood pressure in PCP intoxicants were 101 bpm and 146/87 mmHg, respectively [235].

It should be noted, however, that arylcyclohexamines may exert cardiodepressant effects in the critically ill. Commonly overshadowed by the stimulant effects exerted by their effect on catecholamine reuptake inhibition, depletion of catecholamine stores in unwell patients can lead to the unmasking of the intrinsic negative inotropic effect of arylcyclohexamines caused by decreased intracellular calcium availability [231, 236–238].

In contrast to PCP and ketamine, MXE is associated with significantly greater cardiovascular pressor effects. In 2012, the first analytically confirmed case series of methoxetamine use provided confirmatory data of these effects [70]. Three individuals from London – two of whom had used only methoxetamine – were found to have tachycardia ranging from 113 to 135 bpm and to be hypertensive (systolic BP 187–201 mmHg, diastolic BP 78–104 mmHg); their serum MXE concentrations were 0.09, 0.12, and 0.20 mg/L. A second case series of three individuals from York also presented with mild cardiovascular stimulation [69]. Interestingly, though the observed effects were milder than those observed in the London group (heart rate 67–107 bpm, systolic BP 148–194 mmHg, and diastolic BP 104–112 mmHg), their blood MXE concentrations were generally higher (0.16, 0.24, and 0.45 mg/L).

Other Cardiovascular Effects

There is little evidence as to the effects of PCP and ketamine on cardiac rhythm, while there are no data for MXE. Further, the paucity of experimental data in this area is somewhat contradictory, with ketamine having been observed to both enhance and diminish adrenaline-induced dysrhythmias in animal models [239–242].

In a series of 233 cases of acute recreational ketamine toxicity, despite palpitations being a common presenting symptom (likely to be related to sinus tachycardia), no cases of significant arrhythmia were recorded [234]; there were also no reports of myocardial infarction. Further supporting this is the considerable experience of ketamine in the setting of cardiac catheterization and as an anesthetic, which have not demonstrated proarrhythmic effects [243, 244].

However, a recent study in rabbits has suggested that chronic use of ketamine may result in remodelling of the myocardium, with ventricular myocardial apoptosis, fibrosis, and increased sympathetic sprouting leading to alteration of underlying cardiac electrophysiological properties and possible arrhythmogenesis [245]. Chronic ketamine use has also been shown in mice studies to cause hypertrophy of cardiac muscle with subsequent lysis and coagulative necrosis, with subsequent ECG tracings showing signs of myocardial ischemia [43].

Effects on the Respiratory System

The introduction of PCP and ketamine as dissociative anesthetic agents was based on the observations that they produced markedly less respiratory depression than traditional anesthetics [3]. In reality, they exhibit mild dose-related respiratory depression [246], though usually only in high or overdose [6, 103, 247]; an animal study using doses of 20 mg/kg of PCP successfully produced respiratory depression. The pattern of respiratory depression differs from that of other anesthetic agents in that they have no effect on the slope of the CO₂-response curve; however, it shares similarity with opiates in shifting the curve to the right; that is, respiratory response to hypercarbia remains intact [248]. This also suggests that at the high doses required to cause respiratory depression, this action is via opioid and sigma receptors [38, 246]; this is supported by the observation that at sub-anesthetic doses the role of opioid receptor-mediated analgesia is minimal [158].

However, at usual recreational doses, the effect on respiratory depression is minimal, with clinical studies demonstrating maintenance of, or even an increase in, minute ventilation, tidal volume, and respiratory rate [8, 249]. In a review of 1,000 cases of acute PCP intoxication, tachypnea was much more commonly observed than was bradypnea [232]. Protection of airway reflexes and maintenance of airway muscle tone has also been demonstrated [250]. Further, ketamine has been noted to have a bronchodilatory effect, postulated to be via two mechanisms: (1) indirect stimulation of β₂ adrenergic receptors via catecholamine reuptake inhibition and (2) direct anticholinergic effect on bronchial smooth muscle [112, 251].

Arylcyclohexamines increase tracheobronchial mucus gland secretion [112], which may be of significance in overdose in conjunction with the potential for respiratory depression at high doses.

Effects on the Urogenital System

The toxic effects of arylcyclohexamines, and ketamine in particular, on the urogenital system have only been described relatively recently [252–254]. Only one study has investigated the effects of ketamine on the kidney after acute administration; after a 50 mg/kg administration of ketamine, rats demonstrated decreased glomerular filtration rate and renal plasma flow lasting 2 days [255]. Most studies and reports describe the effects of chronic ketamine use on the urogenital system; the following relates to chronic use.

Urogenital toxicity from ketamine displays a dose–frequency relationship, with cessation of use correlating with improvement in symptoms; however, severe fibrotic changes in the bladder are irreversible, and cessation of ketamine only prevents worsening of pathology in those with early bladder effects [256]. Common symptoms of urogenital toxicity include dysuria, hematuria, urinary urgency, urge incontinence, frequency, and nocturia [254]. In self-reported ketamine users, the frequency of lower urinary tract symptoms and cystitis has been documented to be as high as 27.0 % and 32.0 %, respectively [178, 254]. The mechanism of urogenital toxicity is unknown. However, it is clear that the lower urinary tract is more commonly affected, with the upper urinary tract usually affected after only more prolonged ketamine use.

It is thought that ketamine in particular, and its metabolites, may be directly toxic to the mucosa of the lower urinary tract, owing to their accumulation in urine. This may be due to effects on the microvasculature leading to ischemia and fibrosis [252, 253].

Bladder Effects

The first stage of effects on the bladder is stasis, resulting from a functional disturbance caused by infiltration of mononuclear cells into the bladder and the kidney, with a resultant increase in the risk of infection [257]. The second stage shows degeneration of nerves and the neuromuscular junction, with subsequent muscle thinning. In the third stage, bladder muscle degeneration is distinct, with replacement by fibrous tissue. The resulting small and rigid bladder causes frequent voiding [43], with bladder volumes reported to be as low as 30–100 mL in severe cases [252]. Bladder biopsies taken from chronic ketamine users have shown chronic inflammatory changes similar to those seen with other forms of interstitial cystitis [253, 258].

MXE was initially marketed as a “bladder friendly” alternative to ketamine. While there have been no human case reports of bladder or renal toxicity with methoxetamine use, animal models have shown that methoxetamine does cause inflammatory changes and fibrosis in the bladder and both tubular and glomerular changes in the kidney, which are comparable to those in similar animal models of chronic ketamine administration [259].

Renal Effects

There is evidence that ketamine causes interstitial changes in the ureters and kidneys similar to those seen in the bladder, with mononuclear cell infiltration in the kidneys seen after between 1 and 3 months of ketamine treatment in mice [260]. It is unclear at this stage whether these chronic changes lead to ureteric or kidney tumors [261].

Hydroureter and hydronephrosis are reported in up to half of chronic users presenting with lower urinary tract symptoms [252]. In some of these cases, this was found to be due to renal papillary necrosis, with sloughing of the papilla into the ureter causing ureteric obstruction. In the majority of cases, however, proximal dilatation of the urinary tract is likely to be a result of the bladder changes described above.

There has been one case report in the literature of ketamine-induced renal infarction in an otherwise healthy 20-year-old female, who had presented with persisting left flank pain for 3 weeks [262]. It is thought that decreased nitric oxide synthesis might be the mechanism in this case, with vasospasm from decreased nitric oxide leading to hypoperfusion and subsequent infarction.

Effects on the Gastrointestinal System

There is little published on gastrointestinal (GI) effects of the arylcyclohexamines. User reports have described nausea, vomiting, and abdominal pain with ketamine use; however, there is no description of GI symptoms associated with use of PCP, MXE, or 3-/4-MeO-PCP in the major published case series [70, 101, 232, 235]. An early report of prominent GI symptoms with PCP use was thought to be due to the presence of a contaminant [263].

Users’ reports of ketamine have commonly described nausea, vomiting, and abdominal cramps (“K-cramps ”); however, there is limited data on the frequency of these symptoms. In a study of ketamine users from the UK, the incidence of nausea was as high as 42.2 %, while K-cramp s were the most common reason for frequent users to seek medical attention [180]. In the largest case series of ketamine use, the incidence of abdominal pain and nausea was found to be 21 % and 9.9 %, respectively [234]. This study also noted abnormal liver function tests and/or dilatation of the common bile duct in 16.3 % and 5.7 % of users, respectively. In another case series, 12 of 14 users (85.7 %) with GI symptoms who had an endoscopy performed were found to have gastritis [264]. While the pathophysiology of these effects is not characterized, improvement in GI symptoms was noted on follow-up after abstinence from ketamine use [264, 265]. A recent online survey advertised at clubs, festivals, underground raves, and on social media found 25.3 % of 506 Spanish respondents to have experienced “frequent” abdominal pain [178].

Clinical Presentation and Life-Threatening Complications

The toxidrome of acute arylcyclohexamine toxicity has been described as including violent behavior and/or agitation, nystagmus, hypertension, tachycardia, and analgesia [266], and these are certainly their most common acute effects (see Table 7 for most common effects). Their toxidrome is not, however, specific, and therefore discerning these patients from users of other recreational substances can be difficult not only because of the varied effects of the arylcyclohexamines but also due to their rare use in isolation [3, 14, 48, 235]. Further, variation in dose, routes of administration, and even purity of the drug makes true description of a “typical” presentation impossible. PCP and ketamine have also been known to be adulterants to other drugs, such as MDMA, cocaine, and even cannabis [3, 14], while MXE may be confused by both dealers and users for its weaker parent drug ketamine [69].

Table 7 Frequency of signs and symptoms of arylcyclohexamine toxicity (%)

The most serious of effects appear to be neurological, including seizures and coma, while reports of cardiac arrest have also been noted with PCP use. All have been associated with mortality, and this is discussed further below.

Typical doses of arylcyclohexamines for recreational use are shown in Table 8; ketamine, in particular, has a wide therapeutic index. Its doses can be compared to its therapeutic use, where an intravenous dose of 1–2 mg/kg (4–10 mg/kg i.m.) is usually sufficient to induce anesthesia and 0.2–0.75 mg/kg i.v. (2–4 mg/kg i.m.) provides sedation and analgesia [267].

Table 8 Typical recreational doses of arylcyclohexamines, by route of use

Phencyclidine

The first classification of PCP intoxication was proposed by Burns et al. in 1975, who described three progressive stages that were associated with increasing dose (Table 9) [35]. It was here that a “diagnostic” triad of nystagmus with hypertension in an agitated, or comatose, patient was first described. Limitations of their categorization included a small sample size of 55 patients, in whom only 10 had confirmatory analysis of PCP use by detection in urine. The route of ingestion was also different between categories – inhalation or insufflation for low and moderate doses and oral for high doses – although it is also important to note that the high-dose category consisted of only one patient from their cohort combined with data from four additional case reports [35]. In 1979, Rappolt et al. drew from their clinical experience of 250 cases to elaborate upon Burns et al.’s work, proposing treatment for the differing stages of PCP intoxication [10].

Table 9 Clinical categorization of acute PCP toxicity, adapted from Burns et al. and Rappolt et al. [10, 35]

The largest clinical case series of PCP intoxication was published by McCarron et al. in 1981 and described 1,000 patient episodes [232]. Of these 1,000 episodes, 59.7 % were due to PCP ingestion only; the most common co-ingestants were alcohol (55.3 %) and cannabis (37.5 %). Most (72 %) had smoked PCP – either by lacing cannabis or mint leaves with the powder or by soaking cigarettes in liquid – 13 % had either snorted or sniffed PCP, 12 % had ingested it orally, and 1.6 % had injected it intravenously; the remainder had used either multiple routes or had been exposed to accidental inhalation of fumes following a laboratory explosion.

McCarron et al. noted the difficulty in correlating the signs and symptoms of PCP intoxication with the ingested dose; in particular, they acknowledged difficulty both in the accurate determination of the dose taken and the timely receipt of serum concentrations [268]. A later quantitative study by Walberg et al. did not show any significant correlation between serum PCP concentrations and degree of intoxication with PCP [269].

From their case series, McCarron et al. proposed a clinical classification of PCP intoxication based on patients’ sensorium and behavior and described four major and five minor patterns of intoxication (Table 10) [268]. These patterns of intoxication were not mutually exclusive, and the authors found that patients often transitioned through more than one pattern during their clinical course. Patients with one of the major patterns of intoxication were more likely to require hospitalization, while those with minor syndromes could be expected to improve more rapidly.

Table 10 McCarron clinical classification of PCP intoxication [268]

Since then, there has been a paucity of large case series of acute PCP intoxication. However, the most recent case series by Dominici et al. [235] showed similar trends to those observed by McCarron et al. three-and-a-half decades earlier. The most common route of administration was still smoking, in 83.2 % of the 184 patients reviewed from June 2011 to March 2013. Two-thirds of patients were male, with a mean age of 32.5 years, and more than half (53.8 %) had co-ingested at least one additional substance, the most common being cannabis and benzodiazepines (see Table 11).

Table 11 Co-ingestions in patients presenting with acute arylcyclohexamine toxicity (%)

Overall, the most common clinical findings were nystagmus (57.4–64.1 % of patients) and hypertension (47.0–57.0 % in McCarron et al., systolic BP >140 mmHg or diastolic BP >90 mmHg; not defined in Dominici et al.). Tachycardia (heart rate >100 bpm) was seen in 30 % of cases in McCarron et al.’s series; however, it was reported that heart rates greater than 120 bpm were unusual, with the highest heart rate recorded being 140 bpm. Interestingly, the series from Dominici reported only two cases (1.1 %) of tachycardia; however, they did not report their definition of this.

Agitated behavior was reported in just over one-third of PCP-intoxicated patients [205, 235]. Between 45.9 and 78.3 % of PCP-intoxicated patients were alert at the time of their presentation to emergency services, one-quarter of whom were found to have retrograde amnesia – 11 of these 46 had co-ingested benzodiazepines [235]. Coma was seen in 106 (10.6 %) cases involving PCP intoxication, with the typical history featuring sudden onset of violent or bizarre behavior, or what McCarron et al. termed “acute brain syndrome” (see Table 10), followed by, an occasionally abrupt, loss of consciousness [268]. Coma was differentiated by length of unconsciousness into mild (<2 h), moderate (2–24 h), or severe (>24 h); 55 of those with coma had lone PCP ingestion, of whom 26 (47.3 %) were classified as mild, 17 (30.9 %) moderate, and 12 (21.8 %) severe. Other neuromuscular effects seen with PCP intoxication were generalized tonic–clonic seizures (3.1 %), posturing (2.8 %), dystonic reactions (2.4 %), tardive dyskinesia (1.7 %), athetosis (1.3 %), and catalepsy (1.3 %) [232]. Muscle spasms mimicking extrapyramidal features were seen in 5.9 % of cases, none of whom were known to have received neuroleptic medications prior to onset of the dystonic reactions; in 80 % of these cases, PCP was the only drug ingested [232].

In McCarron et al.’s case series, tachypnea (respiratory rate >30 breaths per minute) was recorded in 4 % of cases; conversely, apnea occurred in 28 patients (2.8 %), of whom three displayed Cheyne–Stokes respirations and three suffered respiratory arrest. In 17 of the 28 episodes, apnea was associated with a generalized tonic–clonic seizure. The effects of PCP on respiration were not specifically reported by Dominici et al.

Hyperthermia (temperature >38.9 °C) was seen in 26 patients (2.6 %) with acute PCP intoxication; 20 of these were lone PCP ingestion, while no details of the co-ingestants in the other 6 patients were provided [232]. Hypothermia was also reported in 64 patients (6.4 %); 41 had co-ingested another drug, but these were not specified. Further, their definition of hypothermia was a temperature <36.6 °C, while no further breakdown of the clinical data was provided [234, 270].

The autonomic effects reported were diaphoresis (3.9 %), bronchospasm (2.1 %), urinary retention (2.4 %), miosis or mydriasis (2.1 % and 6.2 %, respectively), hypersalivation (1.7 %), and bronchorrhea (0.6 %).

Ketamine

A case series of 116 presentations associated with ketamine use found similar rates of hypertension (38.8 %; systolic blood pressure ≥140 mmHg) and tachycardia (29.3 %; heart rate ≥100 bpm) to those seen with PCP intoxication [271]; however, only 13 of these were due to lone ketamine use.

Data from Hong Kong, where ketamine remains the most abused psychotropic drug, has provided the largest case series, of 233 patients, with acute ketamine toxicity [234]. In this series from July 2005 to June 2008, twice as many males as females – with a median age of 22 years – presented to an emergency department after nasal insufflation (88 %) or oral (5 %) use of ketamine (unknown route of ingestion in 7 %). Of those in whom toxicological screening was performed (126 of 233 patients), 72 % of samples were positive only for ketamine; the most commonly co-ingested drugs were alcohol (10.3 %), ecstasy (6.4 %), and methamphetamine (6.0 %).

As for PCP, the majority (54.5 %) of patients presenting to emergency services after acute ketamine use were alert. However, the most common symptom reported in this cohort was still impaired consciousness (45.5 %), although only 14.0 % had a recorded GCS of less than 15 on examination in hospital. The most common clinical signs were hypertension (4-%; not defined) and tachycardia (39 %; heart rate >100 bpm). Forty-nine patients (21.0 %) reported abdominal pain, 23 (9.9 %) had nausea or vomiting, and 20 (8.6 %) reported dysuria; on examination, 41 (17.6 %) had abdominal tenderness and 32 (13.7 %) were recorded to be hyperthermic (not defined). As mentioned previously, abnormal LFTs (16.3 %) and dilatation of the CBD (5.7 %) were observed among ketamine users, and while these both demonstrated improvement upon cessation of ketamine use, their significance is uncertain [234, 265]; gastritis appeared to have a similar temporal association with ketamine use in a subgroup of 14 ketamine users who underwent endoscopy [264].

In 96 cases of chronic ketamine use from Hong Kong, the most common features on presentation to outpatient clinics were of cystitis (91.7 %) – such as dysuria, urgency, and frequency – while 63 (65.5 %) reported chronic abdominal pain, and 15 (15.6 %) had psychiatric concerns (not further defined) [272].

Methoxetamine

While no animal or human studies have examined the toxicity of MXE, limited data is available from three small series of acute MXE intoxication, two case series of user reports of MXE use, and data from the EMCDDA–Europol Joint Report on MXE and the NPIS [48, 69, 70, 85, 114, 270, 273]. In one report, males accounted for three times as many users as females, with an overall median age of 24 years [85]. It was most commonly taken by nasal insufflation (37 %) or orally (38 %) – either by “bombing” or dissolved in water – other routes included i.m. or i.v. injection (total 7 %), sublingually, or rectally [270].

A total of 110 nonfatal intoxications and 20 deaths have been associated with MXE use. Collectively, the data suggest that the clinical toxicity of MXE is akin to that of PCP and of ketamine. The first report in the literature of MXE use was published in 2011 [67]. A 32-year-old male was brought to the emergency department after he was found by police and paramedics to be agitated; he had used an unknown amount of MXE by i.m. injection and denied any co-ingestion. On arrival, he had a heart rate of 105 bpm, blood pressure of 140/95 mmHg, respiratory rate of 16, temperature of 37.2 °C, and pupils of 6 mm. He intermittently appeared in a dissociated state and was noted to have rotary nystagmus; the remainder of his neurological examination was normal, and he returned to his baseline mental status after 8 h [67].

Initial reports suggested that the cardiovascular pressor effects of MXE were greater than those of either PCP or ketamine [70, 274]; however, the incidence of hypertension and tachycardia appears to be comparative based on the limited data available [48, 85]. Data from case reports are mixed. One male, presenting within half an hour of i.v. injection of an unknown dose of MXE, was both hypertensive (168/77 mmHg) and tachycardic (134 bpm), and his serum was positive for MXE; however, this was not quantified [68]. Another male who presented agitated and aggressive was tachycardic (120 bpm) but had a normal blood pressure (130/80 mmHg), while a third was hypertensive (167/110 mmHg) with a heart rate in the normal range (not reported) [275, 276]. In the case presenting with tachycardia, serum MXE concentration was 0.27 mg/L [275]; blood samples were not taken in the hypertensive case. GI symptoms of nausea and vomiting have also been described in user reports; however, no data is available on their frequency [114, 277].

In those with objective recording of conscious state, MXE caused almost double the number of those with impaired conscious state (27.3 %) as compared to ketamine (14.0 %) [48, 85]. Two reports have associated seizures with MXE use (see Life-Threatening Complications, below) [85, 278], while other common neurological features of MXE toxicity include agitation in around one-third of users – similar to the rate seen with PCP use – and muscular symptoms in up to 18 % of cases [270]. In contrast to PCP and ketamine, a case series of three MXE users reported significant cerebellar toxicity, with their serum MXE concentrations being 0.16, 0.24, and 0.45 mg/L [69].

Methoxylated Arylcyclohexamines

The only case series of other arylcyclohexamine analogues comes from the Swedish STRIDA project, which reported 59 cases of acute 3- and 4-MeO-PCP toxicity [101]; 86.4 % were male and the median age was 26 years. Fourteen cases described route of administration: oral was the most common (9 cases; 64.3 %), while nasal, i.v., and rectal routes were also reported. As with the other arylcyclohexamines, polydrug use was common. Additional substances were detected in the majority of cases (52 of 59; 88.1 %), the most common being cannabis and ethanol (see Table 11); all 7 of the single-substance cases involved 3-MeO-PCP.

The most frequent clinical signs observed with methoxylated arylcyclohexamine use were akin to those observed with PCP, ketamine, and MXE use (see Table 7); altered conscious state (57.6 %), tachycardia (54.2 %), and hypertension (45.8 %) are the most commonly reported, while nystagmus was seen in 30.5 % of cases. In the 7 cases of lone 3-MeO-PCP use, there was a greater incidence of hypertension (100 %; systolic BP ≥140 mmHg) and tachycardia (71 %; heart rate ≥100 bpm), with similar rates of altered conscious state (57.1 %) and nystagmus (28.6 %) [101]. There were no reports of seizures or of altered thermoregulation associated with lone 3-MeO-PCP use.

Emergence Phenomena and the “K- and M- Holes”

The emergence phenomena observed during the recovery phase of anesthesia with arylcyclohexamines has limited their clinical use and was one of the main reasons that PCP was withdrawn less than 10 years after being introduced [9]. Characterized by a confusional state, vivid dreams, and hallucinations, these reactions occurred most frequently in middle-aged males, with an incidence of 17–30 % with use of PCP in clinical anesthesia [38]. Following ketamine anesthesia, studies have shown varying incidences of emergence phenomena, from less than 5 % to more than 30 % of subjects [279–284]. Collectively, the experiences of recreational users during their dissociated state have been commonly referred to as the “K-Hole ” and “M-Hole ” when associated with ketamine and methoxetamine use, respectively [114, 285]. Other descriptions have included a sensation of feeling light, body distortion, experiences of cosmic oneness, and out-of-body and near-death experiences [103, 286].

The underlying mechanism of the emergence phenomena caused by arylcyclohexamines appears to be due to their depressive effects on visual and auditory relay nuclei in the medial geniculate and inferior colliculus, respectively, leading to misperception and/or misinterpretation of visual and auditory stimuli [287, 288]. Further, the loss of proprioceptive sensation results in decreased sensitivity to gravity, providing an explanation for the out-of-body and floating experiences described by recreational users and by those recovering from general anesthesia [76, 114, 286, 289].

In McCarron’s study of 1,000 cases of PCP intoxication, the incidence of emergence patterns was described for each category of coma [268]. A total of 106 cases of coma were described, of whom 55 (51.8 %) had only ingested PCP; of these, 26 (47.3 %) were classified as mild, 17 (30.9 %) as moderate, and 12 (21.8 %) as severe coma. In the 12 patients with severe coma (>24 h) due to lone PCP intoxication, the emergence patterns included five with acute brain syndrome (defined by McCarron et al. as “disorientation with any combination of confusion, lack of judgment, inappropriate affect, or loss of recent memory (with or without bizarre behavior, violence, or agitation)”) (see Table 10) lasting up to 7 days, one with stupor for 10 days, one with catatonia lasting 3 days, one with lethargy for 24 h, three with agitation, and one with psychosis requiring transfer to a psychiatric hospital. In those emerging from moderate coma, eight developed acute brain syndrome for up to 8 days, three were agitated for up to 14 days, one was psychotic for 7 days, and one displayed bizarre behavior for 4 days; only 4 of the 17 (23.5 %) were mentally clear on awakening from coma. For those with mild coma, a majority (14, 53.8 %) displayed acute brain syndrome, which lasted less than 24 h in all cases; four were stuporous; two were agitated; two were psychotic for 34 h and 3 days, respectively; one was catatonic; and one displayed violent behavior. Two (7.7 %) displayed mental clarity on emergence from coma [268].

Aside from this case series by McCarron et al., there is no other data on the incidence or severity of emergence reactions following acute toxicity and coma associated with PCP or other arylcyclohexamine analogues. The availability of pharmaceutical non-racemic ketamine and reports of its recreational use [290], in particular the S(+)-enantiomer, may impact the prevalence of emergence reactions given the evidence supporting a decrease in the incidence of these reactions with its use, as compared with racemic or R(−)-ketamine [106–108].

Life-Threatening Complications

Massive overdoses, of up to 1 g orally of PCP, have resulted in coma for up to 5 days’ duration [115]. McCarron et al. found that, of those with coma, almost one-quarter had prolonged, or “severe,” coma lasting more than 24 h, with the possibility of delayed and prolonged hypoventilation and apnea as a result [268]. Burns et al. also noted that such cases were marked by sustained hypertension, tachycardia, and more prominent neurological features, including seizures [291, 292]. These findings were consistent with McCarron et al.’s results demonstrating increased frequency of almost all clinical features of PCP intoxication with increased duration of coma – with the exception of nystagmus and hypertension, which were uniformly present in around 60 % and 50 % of all PCP intoxicants, respectively, regardless of duration of coma [268].

There is conflicting data regarding the association between serum PCP concentration and clinical findings. Most demonstrate no correlation between the two, with the exception of the finding of a moderate correlation (r = 0.60, p < 0.05) between plasma PCP concentration and hypertension in a series of 5 patients with acute PCP toxicity [293]. It was also noted in the series by Walberg et al. that, of the 183 patients whose blood tested positive for PCP, all 14 of those with coma had uniformly higher serum PCP concentrations (>0.1 mg/L) [269]. One explanation for the lack of association is the high protein- and tissue-bound fraction of PCP, meaning blood PCP concentrations do not reflect total body load [293, 294]. There is also some evidence to suggest that chronic users develop tolerance to PCP, and, as a result, their toxic blood concentrations are significantly higher than for casual users [295, 296].

In the largest case series of 1,000 acute PCP intoxications, there were three cases of cardiac arrest, all of whom were successfully resuscitated; however, no further details were provided [232]. One additional patient died post-cardiac arrest, found on autopsy to be due to pulmonary embolism. Three patients suffered respiratory arrest, from a total of 28 in whom apneas were noted; 16 (57.1 %) had co-ingested a sedative–hypnotic or narcotic drug in addition to PCP. Of the 28 in whom apneic episodes were seen, 17 were associated with seizures, with some requiring intubation and ventilation for prolonged apnea; it was not reported how many of those who had seizures had also co-ingested other substances. A total of 26 patients suffered from one or more generalized tonic–clonic seizures, while a further 5 patients had status epilepticus; in all cases, status epilepticus was terminated by administration of intravenous diazepam [232]. Two cases of seizure have been reported with MXE use: one had multiple seizures terminated by administration of i.v. clonazepam, after inhaling the vapor of 50 mg of MXE powder over a couple of minutes – his serum MXE concentration was 0.03 mg/L [278]; a further case is mentioned in a series of MXE intoxications, but no further details were reported [85].

Rhabdomyolysis (creatine phosphokinase [CK] >16,000 IU/L) was seen in 2.5 % of 1,000 patients with PCP intoxication; of these, 40 % had acute kidney injury (defined as one of (a) rising serum creatinine or urea despite i.v. rehydration, (b) serum urea/creatinine ratio between 9 and 11, or (c) oliguria [urine output <500 mL day] persisting for several days) [297].

The vast majority of patients with acute arylcyclohexamine intoxication will survive with supportive care. However, it is the group that is particularly behaviorally disturbed in whom additional significant harm can occur and who can pose a significant management challenge requiring critical care admission. Owing to the dissociative and analgesic effects in combination with acute behavioral disturbance and agitation, patients may self-inflict significant damage while under the influence of arylcyclohexamines; cases as marked as nuclear evisceration and genital mutilation have been reported [298–300].

Other Considerations

A concern with regard to the potential for serious toxicity and/or mortality associated with recreational arylcyclohexamine use is the prevalence of their misrepresentation, either intentionally or accidentally, as a contaminant. Historically, PCP has a long history of this. Early in its use, it was often found to be adulterated by, or an adulterant of, other arylcyclohexamines. While some were thought to be less potent than PCP, others with increased potency, such as TCP, PCE, and PCPy, had the potential to cause significant inadvertent toxicity. The precursor to, and sometime contaminant of, PCP – PCC – was reported to cause serious toxicity, including coma and death; however, there is little further detail about its clinical effects [3, 62, 64].

More recently, there have been reports of ketamine and MXE being interchanged for each other. Given the binding affinity of MXE for the NMDA receptor is significantly higher than that of ketamine [127], there is the potential for inadvertent overdose. With the marketing of MXE as an “alternate” form of ketamine, the risk of significant toxicity with the use of MXE by those inexperienced in its dosing is compounded by its longer time to onset than ketamine, of up to 90 min, which has caused some users to prematurely re-dose [114].

Further, there are reports of ketamine being sold on the street in its non-racemic form. Given the S(+)-enantiomer is four times more potent in its dissociative effects than its R(−)-enantiomer, this presents further risk to those not aware of the difference in effects between racemic and non-racemic mixtures [40].

Deaths

Throughout its decades of use, PCP has been associated with hundreds of deaths. This is, however, more commonly due to the acute behavioral disturbances associated with mild intoxication [35, 115, 301]. After significant PCP ingestion, coma is protective from potential self-harm in these patients, while risk of death from accidental overdose is minimized by provision of adequate supportive care.

Data from a recent 5-year retrospective review from the New York City Medical Examiner’s Office documented a total of 138 autopsy cases, from 2003 to 2007, in which PCP was detected in postmortem blood samples [296]. Of these, the majority (80; 58.0 %) were involved in violent deaths in which PCP was detected but not thought to be directly related to the cause of death; 52 (37.7 %) were deemed to be as a result of mixed drug intoxication, 5 (3.6 %) were nonviolent deaths in which only PCP was detected, and in one (0.7 %), the cause of death was indeterminate.

Previous case reports have reported lethal serum PCP concentrations of greater than 2.0–2.5 mg/L [302]. However since then, deaths have been reported where the serum PCP concentration was much lower. In one case, a 30-year-old man without any medical comorbidities died from a nonviolent cause [296]. Urine toxicology screen on presentation to the emergency room was positive only for cannabinoids, and serum was negative for ethanol, paracetamol, and salicylates; serum PCP concentration at autopsy was 0.07 mg/L; however, this was 5 days after his initial presentation. A quantitative study by Walberg et al. did not show any significant correlation between serum PCP levels and degree of intoxication with PCP [269]. However, of the 183 patients whose blood was positive for PCP, all of those with coma had a serum PCP level greater than 0.1 mg/L. Indeed, significantly raised serum PCP concentrations are usually seen after massive overdose resulting in death attributable to PCP toxicity, rather than consequences of behavioral disturbance. A serum PCP concentration of 7.0 mg/L was reported in a case of status epilepticus with resultant cardiopulmonary arrest and death [115, 303].

Ketamine

Despite its widespread use, there is little recent data on the number of deaths relating to ketamine use. The first documented blood ketamine concentration in a fatality was reported in 1988, in a 31-year-old female who died after an accidental injection (not reported whether i.v. or i.m.) of up to 900 mg of ketamine; she had a blood ketamine concentration of 7.0 mg/L [304]. A comparison case also detailed in this report, of a 46-year-old man given 100 mg of ketamine (route of administration not documented) during anesthesia for surgery post-gunshot wound, showed a postmortem blood ketamine level of 3.0 mg/L. He suffered cardiac arrest and was pronounced dead 40 min after receiving ketamine; the final cause of death was not reported. A review by the authors of 27 other cases in which ketamine was detected postmortem over a period of 2 years showed blood concentrations of between 0.1 and 9.5 mg/L, in which 15 were less than 1.0 mg/L [304]. No comment was made about the cause of death in these 27 cases or whether other drugs or alcohol were detected; however, their report highlights the issue of timing of postmortem blood and tissue samples and the importance of considering the time of death in interpreting these results.

A review of ketamine-associated deaths examined at the New York City Office of the Chief Medical Examiner between 1997 and 1999 identified 12 nonhospital deaths due to acute intoxication, none of which were due solely to ketamine [305]; of the 87 total ketamine-associated deaths, the vast majority, 72, were inhospital deaths in the setting of treatment with ketamine for burns or surgery.

In 2008, Schifano et al. presented data on 23 ketamine-associated deaths in the UK from 1993 to 2006; of these, only four cases were identified in which ketamine was the only drug found, three of which were deemed to be accidental overdoses, with the other being a case of suicide [306, 307]. Ketamine concentrations were not reported in these cases. Since then, data from the UK’s National Programme on Substance Abuse Deaths (NPSAD) showed a peak of 23 deaths related to ketamine use in 2009, with a subsequent decline to 12 and 8 deaths in 2011 and 2012, respectively [308]. A total of 93 deaths in which ketamine was mentioned on the death certificate were identified from the NPSAD database; in most (70; 75.3 %), other drugs and/or alcohol were also detected [40]. In the remaining 23 cases, where ketamine was the only substance detected, it was not possible to determine whether ketamine was the actual cause of death [40].

Reports of lone ketamine toxicity leading to fatality have reported blood ketamine concentrations varying from 1.8 to 27.4 mg/L [304, 309, 310], while previous studies in squirrel monkeys investigating anesthetic and lethal ketamine doses have led to the estimate of a lethal i.v. dose in humans of 11.3 mg/kg [307, 311]. In comparison, anesthetic studies have demonstrated that i.v. loading doses of 2 mg/kg of ketamine, followed by continuous ketamine infusion, were able to achieve a targeted blood ketamine concentration of between 2 and 3 mg/L, which maintained stable anesthesia [312, 313]. Subjects were found to be rousable to deep stimulus at blood ketamine concentrations of between 1.0 and 1.5 mg/L, while recovery from anesthesia was seen at blood ketamine concentrations between 0.50 and 0.64 mg/L [312–315]. Apart from the limited case reports of fatalities documenting blood ketamine concentrations, there are no other published data for ketamine concentrations in recreational use or overdose.

Methoxetamine

In 2012, no confirmed deaths were attributable solely to the use of MXE [75]. By 2014, 20 deaths associated with MXE use had been reported to the EMCDDA [48, 116]; in 8 cases, MXE was the only psychoactive substance reported. Of those reporting a cause of death, four cases listed MXE in the cause of death, 2 cases were due to mixed overdose, while three cases listed MXE as contributing to death [48]. Interestingly, 4 of the 20 deaths were from drowning, perhaps highlighting the theoretically increased potential for accidental self-harm with MXE given its prominent cerebellar toxicity alongside the dissociation associated with other arylcyclohexamines [69]. Postmortem MXE concentrations in blood were highly variable, ranging from 0.03 mg/L (cause of death: mixed overdose) to 5.2 mg/L (cause of death: drowning) [48, 72, 316]; there was overlap in MXE concentrations with acute toxicity, which in one case series ranged from 0.16 to 0.45 mg/L [69].

The first case report of death involving MXE, published in 2013, was that of a 26-year-old male found deceased in his home, whose postmortem blood MXE concentration was 8.6 mg/kg; three synthetic cannabinoids were also detected (concentrations not reported), as was tetrahydrocannabinol [71]. A year later, a case of death where clinical features were recorded was reported from Poland [73]. A 31-year-old man presented to the emergency department in deep coma; he was tachycardic (heart rate 120–140 bpm) and hyperthermic (temperature 39 °C) and had rhabdomyolysis with a peak CK of 220,531 IU/L. He was also reported to have had seizures, required mechanical ventilation for respiratory failure, and developed acute liver and renal failure; he was admitted to intensive care but died 28 days after his presentation. Serum MXE concentration taken 8.5 h after presentation was 0.32 mg/L; amphetamine was also detected (0.06 mg/L). The first report of fatality from lone MXE use was also reported from Poland, in 2015, of a 29-year-old man who was found deceased in his home [73]; his estimated serum MXE concentration was 5.8 mg/L.

Diagnosis

Clinical

As with all patients who present with recreational drug overdose, collateral history where available provides important information about the dose and route of exposure, co-ingestant(s), and previous drug use. The most common clinical signs in arylcyclohexamine intoxication are nystagmus and hypertension; in the presence of an altered sensorium – either agitation or coma – these patients have been described to have a classic triad of PCP intoxication.

Serum and Urine Arylcyclohexamine Testing

Testing for arylcyclohexamines is not part of routine toxicological screening, and samples are often required to be sent to specialist laboratories for their identification.

Phencyclidine can be qualitatively detected using enzyme-linked immunoassay techniques with a sensitivity of 0.01 mg/L; however, false-positive results are possible due to cross-reactivity with molecularly similar substances such as dextromethorphan [38]. No commercially available immunoassay exists for ketamine; its detection is by gas chromatography and mass spectroscopy. Rapid-detection urine assays have recently been developed and are reported to be sensitive and specific, with a limit of detection of 0.005 mg/L [317, 318]. However, there have been case reports of false-positive results for PCP in those patients also on venlafaxine, with its active metabolite O-desmethylvenlafaxine cross-reacting with the PCP assay reagent [319–321]. Other drugs that have been shown to cause false-positive results in urine PCP assays include dextromethorphan and diphenhydramine [322, 323].

Methoxetamine is tested for using either calorimetric or immunoassay kits and analyzed by gas chromatography–mass spectrometry (GC–MS) techniques. Newer techniques for methoxetamine and other arylcyclohexamine analogues have recently been developed, with the use of high-pressure liquid chromatography (HPLC) able to detect these compounds in both blood and urine [324]. With the rapid appearance of novel arylcyclohexamine analogues, interpretation of immunoassay results must be made with caution. Cross-reactivity between the methoxylated arylcyclohexamines and PCP has been described after confirmatory MS analysis [101].

While toxicological screening for these compounds is not required routinely, it can be useful academically, particularly to help confirm the pattern of toxicity associated with novel arylcyclohexamines. This has been shown in the recent case series of 3- and 4-MeO-PCP as well as the earlier case reports of MXE use [68, 69, 71–73, 101, 275].

Other Investigations

Patients with significant clinical features should have routine hematology and biochemistry testing. Nonspecific laboratory findings include leukocytosis, hypoglycemia, and elevated CK, urea, and creatinine [232, 234]. In the case series of 1,000 patient episodes by McCarron et al., 22 % of those tested had hypoglycemia of less than 70 mg/dL (3.9 mmol/L). Creatinine kinase was elevated above 300 IU/L (reference range 5–200 IU/L) in 70 % of cases, of whom roughly one-third each were violent, agitated, and calm; the highest recorded CK was 423,045 IU/L [232]. Of those who had no other reason to have an elevated white blood cell count (e.g., trauma, rhabdomyolysis, or infection), 39 % were shown to have white blood cell counts of greater than 12,000/ml³. The number of those with leukocytosis (36 %) was similar in a case series of acute ketamine toxicity, while raised creatinine kinase was seen in 32 % of cases [234]

An ECG is an essential investigation in patients with acute recreational drug toxicity in the critical care setting; however, there are no findings specific to arylcyclohexamine toxicity. In one study of anesthesia during electroconvulsive therapy, ketamine was found to cause QTc prolongation from a mean of 410.9 ms at baseline to a mean of 431.6 ms after induction with a 1 mg/kg bolus [325]; there were no episodes of torsade de pointes, and this is therefore of doubtful clinical significance. Isolated reports of ST-segment and T-wave changes with ketamine use also exist. One case of broad-complex arrhythmia with peaked T-waves in the setting of PCP intoxication was likely caused by hyperkalemia of 9.35 mmol/L rather than arylcyclohexamine-specific effects [326]. There has also been a single case report of transient Brugada syndrome in a 31-year-old Caucasian male who had nasally insufflated ketamine [327]. Resolution of the Brugada pattern and subsequent testing demonstrated no other reason for the ECG findings, and it was determined that ketamine’s inhibition of I_Na and I_Ca channels was the cause of the ST elevation [327, 328]. Reports of propofol also causing a transient Brugada syndrome with subsequent malignant arrhythmias is of importance due to its use as a sedative agent in the critical care setting, especially when used to sedate patients with recreational drug toxicity [329].

An electroencephalogram (EEG) may be useful in the undifferentiated comatose patient to exclude ongoing seizures or status epilepticus as a cause for unconsciousness or neuromuscular excitation. Findings seen in arylcyclohexamine toxicity include diffuse slowing with θ and δ waves, which may return to normal prior to clinical improvement [38, 263, 330, 331].

Differential Diagnoses

The varied signs and symptoms seen with arylcyclohexamine intoxication, and indeed the frequency of co-ingested drugs, make for a wide list of differential diagnoses (Table 12). Arylcyclohexamine intoxication therefore requires a high index of suspicion and gathering of as complete a collateral history as possible.

Table 12 Differential diagnoses for arylcyclohexamine intoxication

Management

Conservative and supportive management is the mainstay of treatment in arylcyclohexamine toxicity, with the focus on maintaining a clear airway, adequate respiration, and oxygenation; providing circulatory support where necessary; and being mindful of thermoregulation due to the potentially significant psychomotor agitation and bioamine reuptake inhibition. Restraint and sedation may also be necessary in the most acutely disturbed patients to avoid harm to both the patient and clinical staff tasked with their care; this is described in more detail in the next section. There is no specific antidote for arylcyclohexamine toxicity, and most patients will survive with purely supportive measures.

Intravenous access is required for all patients who present with significant acute arylcyclohexamine toxicity, with blood taken for electrolytes, glucose, lactate, urea, and creatinine. In the significantly agitated patient, rhabdomyolysis may be a feature, with myoglobin-associated acute kidney injury as a cause of high morbidity and mortality associated with PCP intoxication in particular [332, 333]. When present, it should be treated aggressively with intravenous fluids (see chapter on Rhabdomyolysis in the Critically Poisoned Patient).

Seizures can be treated initially with benzodiazepines. In the setting of status epilepticus, or in a patient with an otherwise compromised airway, sedation with a barbiturate or propofol, paralysis, and endotracheal intubation is indicated. In patients requiring critical care intervention and management, significant hyperthermia is associated with poor prognosis [334–336]. Therefore, particularly in the setting of hyperthermia, prompt anticonvulsant therapy should also be instigated to minimize additional thermogenesis from seizure activity. In those patients with significant hyperpyrexia (>39–40 °C), aggressive cooling methods should be employed, including the use of cold intravenous fluids, packing of the axillae/groin with ice, and, if necessary, immersion in an ice water bath or the use of an intravascular cooling device (Grade III recommendation).

Patients should have continuous cardiac monitoring, due to the prevalence of tachycardia and reports of arrhythmias. While hypertension is common, it is usually short-lived, with most cases resolving without the need for specific antihypertensive treatment. Those that require medicating for behavioral disturbance, agitation, or seizures – such as benzodiazepines – are likely to see a beneficial effect of this on blood pressure. In 563 patients with PCP-induced hypertension, only one required specific antihypertensive medication; seven additional patients had known essential hypertension, of whom five received antihypertensives [232]. If hypertension is persistent and severe (systolic BP >200 mmHg or diastolic BP >120 mmHg), we recommend treatment with intravenous glyceryl trinitrate as its short half-life enables it to be readily titrated (Grade III recommendation). While this is also true of sodium nitroprusside, it has the potential to increase catecholamine secretion and worsen tachycardia [337, 338]. The longer half-lives of calcium channel blockers make them impractical for rapid titration in the critical care setting, while beta-blockers are generally not recommended in the setting of acute stimulant toxicity as they can result in unopposed alpha activity and worsen hypertensive crises.

Agitation and Behavioral Management

As the psychotomimetic features of acute intoxication are similar to those seen in emergence reactions, it is useful to be mindful of the importance of non-pharmacological modalities of treatment in caring for the patient intoxicated with arylcyclohexamines. However, the spectrum of behavioral disturbance is wide. Indeed, the major toxicity of arylcyclohexamines is related to these effects, with unawareness of surroundings and oblivion to pain due to their dissociative and analgesic effects leading to the potential for self-inflicted injuries and extreme overexertion. The potential for accidental self-harm is theoretically exacerbated with MXE given its reported cerebellar toxicity [69].

In mild intoxication, management may involve maintaining an environment of relative sensory deprivation [339]; however, this has not been formally studied (Grade III recommendation). Patients with more severe toxicity and acute behavioral disturbance may initially require physical restraint in order to establish intravenous access to facilitate chemical sedation. Overall, chemical restraint is preferred, as physical restraint may exacerbate underlying rhabdomyolysis. Benzodiazepines are the preferred sedating agent, as antipsychotic medications may exacerbate hyperthermia, lower the seizure threshold, or worsen dystonic and anticholinergic reactions [266].

If an antipsychotic medication is required, atypical agents offer a better choice than traditional antipsychotics due to their more favorable side effect profile, in particular their less pronounced extrapyramidal side effects [340]. In one series of PCP intoxication, 168 patients received haloperidol and 66 received chlorpromazine for agitated behavior, in whom 10 (6.0 %) and one (1.5 %), respectively, developed dystonic reactions [268]. There is limited published experience with newer sedatives, such as dexmedetomidine, or newer atypical antipsychotics. In our opinion, therefore, benzodiazepines should remain first line in these patients (Grade III recommendation). Initial intramuscular dosing, using an agent with good systemic absorption by this route such as lorazepam, may be required in those with very severe behavioral disturbance; however, i.v. administration is the route of choice as it allows for more titrated dosing.

Decontamination and Elimination

Activated Charcoal

There are no human studies investigating the role of activated charcoal (AC) administration on arylcyclohexamine toxicity; data on the use of AC is limited to a single study investigating PCP in dogs [341]. Patients with significant arylcyclohexamine toxicity generally have significant behavioral disturbance and/or drowsiness which would limit AC administration and routine use of AC is not recommended.

Urinary Acidification

As the arylcyclohexamines are weak bases, acidification of urine theoretically enhances their elimination; indeed, the use of ammonium chloride was previously advocated for this purpose [342]. However, the risk of inducing systemic acidemia outweighs the potential benefits (Grade III recommendation). Urinary acidification also has the potential to increase the risk of acute tubular necrosis in patients with rhabdomyolysis [333, 343]. Furthermore, while renal clearance of PCP has been shown to be increased by up to 23 % following urinary acidification, this accounts for only 1 % of total PCP clearance due to the majority of its clearance being via hepatic metabolism [109, 344]. Given these concerns, coupled with the fact that the majority of patients with arylcyclohexamine intoxication will survive with supportive care, urinary acidification is not recommended.

Hemoperfusion and Hemodialysis

Due to the large volume of distribution of PCP and the other arylcyclohexamines, along with their high lipid solubility, hemodialysis and hemoperfusion are not effective in managing arylcyclohexamine toxicity [345].

Emergence Reactions

As described above, McCarron et al. reported a high incidence of emergence reactions following acute PCP toxicity, with more than 75 % of those with moderate or severe coma displaying confused, agitated, or psychotic behavior [268]. In those with mild or moderate coma, only 13.9 % emerged from their altered conscious state with clear sensorium, while none of those with severe coma did so.

There are no data for the treatment of emergence reactions following acute arylcyclohexamine toxicity causing coma. Indeed, apart from the study by McCarron et al., there are no data on the frequency of emergence reactions following arylcyclohexamine toxicity in the recreational setting. Anesthetic studies demonstrate that the incidence of emergence phenomena after general anesthesia with PCP and ketamine ranges anywhere from less than 5 % to more than 30 % [38, 103]. While the incidence of emergence reactions following anesthesia with PCP has been reported to be higher in males [38], it appears to be higher in women after ketamine anesthesia [104, 267, 346].

The role of preanesthetic medication in reducing the incidence and/or severity of emergence reactions following ketamine anesthesia was acknowledged even after its first report; however, it was not investigated at that time [11]. Initial studies investigating various premedications offered conflicting results about the beneficial effect of agents such as droperidol, chlorpromazine, and opiate-containing regimens [346–351], and these are not recommended [103, 351].

Benzodiazepines have been shown to reduce the incidence of emergence reactions following ketamine anesthesia, with initial studies suggesting lorazepam to be superior to other benzodiazepines, including diazepam (Grade II-1 recommendation) [351–353]. However, the introduction of midazolam has since superseded their use due to its shorter half-life and water solubility and has been shown to be superior to diazepam in reducing the incidence of emergence reactions (Grade I recommendation) [354–356]. In one randomized, double-blinded study of those induced to anesthesia using 250 mg i.v. ketamine, only 4 % of those who also received 12.5 mg i.v. midazolam had emergence reactions compared to 36 % of those who received 20 mg i.v. diazepam in addition to ketamine [355].

While these data provide good evidence of the efficacy of benzodiazepines, and in particular midazolam, for reducing the incidence of emergence phenomena, they must be interpreted with some caution in managing the patient with acute arylcyclohexamine toxicity. Not only are they based on controlled anesthetic studies, most studies administered benzodiazepines as a premedication or at the time of induction of anesthesia, though other studies have suggested that they are equally effective administered at the completion of a procedure [280, 357]. A study by Levänen et al. comparing dexmedetomidine (2.5 μg/kg) with midazolam (0.07 mg/kg) noted a significant reduction in CNS effects in the dexmedetomidine group; however, these were mostly unpleasant dreams, with no reports of agitated or psychotic behavior [358]. A more recent study comparing dexmedetomidine (1 μg/kg) and midazolam (0.02 mg/kg) also demonstrated less subjective CNS effects in the dexmedetomidine group; however, their findings were not significant [359]; similarly, most of the described effects were mild or moderate dream states, without any significant behavioral changes reported.

As an aside, there is also evidence that the S(+)-enantiomer of ketamine is responsible for less emergence phenomena than either the racemic mixture or, in particular, the R(−)-enantiomer [106–108]. As the S(+)-enantiomer is now available in a pharmaceutical preparation [267], its use would be recommended above the racemic mixture for anesthesia and for sedation in the critical care setting.

Disposition

Most patients with mild intoxication with arylcyclohexamines will be able to be discharged from the emergency department after a period of monitoring that has demonstrated normal cognition, normal and stable vital signs, normal ECG, and normal biochemical markers. Those with persisting signs of mild arylcyclohexamine toxicity or requiring repeat blood tests, and who do not have significant behavioral disturbance, may be managed in an observation unit or with standard ward-based care.

Patients with severe toxicity, as highlighted in Box 1, will require critical care admission for further targeted management as discussed in the sections above.

Box 1

Criteria for critical care management of acute arylcyclohexamine toxicity

Criteria for admission to intensive care unit	Criteria for discharge from intensive care unit
Agitation or violent behavior requiring sedation and/or intensive monitoring	Settled or settling behavior
Rhabdomyolysis, requiring hemodialysis	Maintenance of adequate urine output and improving renal function
Severe, refractory hypertension
Severe hyperthermia (temperature >39–40 °C)	Thermoregulation not requiring exogenous cooling
Prolonged coma	Improvement in mental status
Status epilepticus or seizures requiring therapy	Resolution of seizures
Other reason(s) for endotracheal intubation	Successful extubation and maintenance of patent airway
Injuries/trauma secondary to behavioral disturbance requiring ICU level care

Special Populations

Dependants

The impact of the behavioral effects of the arylcyclohexamines, including agitation, violence, and altered sensorium, necessitates consideration of the resulting social issues [360–362]. While not specific to arylcyclohexamines, these issues require the assessment and safeguarding of vulnerable adults – and, indeed, children – due to the potential for significant intimate partner domestic abuse, crime, and incarceration [298, 363].

Children

In addition to the potential for ingestion of arylcyclohexamines in either liquid, tablet, or powder forms, significant accidental poisoning, in particular with PCP, can occur from exposure to passive smoke [364]. Signs may be as subtle as listlessness and decreased responsiveness to tactile stimuli, irritability, and poor feeding in the very young [365]. Other neurological signs observed in adult presentations may also manifest in children, with nystagmus, ataxia, increased muscle tone, and opisthotonus all reported [364–366]. Hypertension was also common, found in 30 % of children in one study, while seizures, respiratory depression, and apnea were reported more frequently than in adults; agitation and violent behavior were less commonly seen [365, 366].

All pediatric patients with acute arylcyclohexamine toxicity require continuous monitoring to observe for any fluctuation in consciousness or cardiorespiratory status [266]. Infants may require longer observation and admission due to case reports of their delayed recovery [364]. Any child who presents after exposure to arylcyclohexamines should have safeguarding procedures enacted.

Pregnant and Breast-Feeding Patients

Phencyclidine has been shown to cross the placenta and is associated with poor feeding, irritability, and hypertonicity in neonates demonstrating possible withdrawal from PCP [367, 368]. It is also excreted into breast milk, where it may be present for up to 40 days after last maternal use and in concentrations of up to ten times plasma levels [369, 370].

Studies investigating the impact of intrauterine exposure to PCP have found it to cause symmetrical intrauterine growth retardation [371, 372]. In one study, 42.2 % of infants had a birthweight for gestational age below the 25th percentile, and 45.7 % had a head circumference below the 25th percentile; while clearly significant, these were less than a comparison group exposed to cocaine, in whom 55.1 % and 67.9 % of newborns were below the 25th percentile for birthweight and head circumference, respectively [372].

As ketamine is also known to cross the placenta, appropriate post-delivery care should be taken in those newborn infants who are born via Caesarean section under ketamine anesthesia [112]. A recent systematic review and meta-analysis of i.v. ketamine during spinal and general anesthesia for Caesarean section found no significant effects on neonates, though their findings were limited by a paucity of data for both maternal effects and neonatal well-being [373]. There are no human studies investigating the teratogenicity of ketamine or other arylcyclohexamines. Injections of 25 mg/kg ketamine of female dogs in their first trimester did not demonstrate any adverse effects to either the mother or her pups [374]. Another study found that ketamine potentiated the teratogenic effects of cocaine in mice but that it was not teratogenic on its own [307, 375].

There are no data for MXE to be able to determine whether it crosses the placenta and/or whether it is excreted into breast milk.

Grading System for Levels of Evidence Supporting Recommendations in Critical Care Toxicology, 2nd Edition

• I Evidence obtained from at least one properly randomized controlled trial.

• II-1 Evidence obtained from well-designed controlled trials without randomization.

• II-2 Evidence obtained from well-designed cohort or case–control analytic studies, preferably from more than one center or research group.

• II-3 Evidence obtained from multiple time series with or without the intervention. Dramatic results in uncontrolled experiments (such as the results of the introduction of penicillin treatment in the 1940s) could also be regarded as this type of evidence.

• III Opinions of respected authorities, based on clinical experience, descriptive studies and case reports, or reports of expert committees.