Drug Abuse

CHAPTER 2 GENERAL INTRODUCTION

This chapter provides a general overview of cannabis, starting with a historical perspective which is followed by a description of its current usage prevalence. Then a short introduction to the plant and its derivatives is given, followed by an extensive review of the drug's chemistry and pharmacology. The concluding part is dedicated to an overview of the acute and chronic effects of cannabis use. Finally, the drug's therapeutical applications are shortly summarized.

2.1 History of Cannabis Use

Throughout recorded history man has shown a great interest in Cannabis sativa, also known as Indian hemp, which is an herbaceous annual. That interest extends backward for about twelve thousand years (Abel, 1979). The ancient Chinese used cannabis as an ingredient in clothes and ropes. The Greeks became familiar with this application of cannabis during the sixth century BC. At first they obtained cannabis from Milesion colonies, but by the third century BC, primarily from the Rhone valley in France. The Romans were aware of the tensile strength of cannabis rope and used it extensively for naval construction. Early in American history the plant was widely cultivated for its commercially useful fiber, beginning in Jamestown, Virginia, in 1611 (Grinspoon and Bakalar, 1992).

Cannabis has long been used as a medicine in India, China, the Middle East, Southeast Asia, South Africa, and South America (Grinspoon and Bakalar, 1993). The earliest description of the drug, however, dates back to the twenty-eight century BC (Abel, 1979). The Chinese emperor Shen-Nung described hundreds of drugs, among them cannabis, in an herbal (the Pen Ts'ao) that eventually became the standard Chinese compendium of medicines. Cannabis was recommended for relief or recovery from rheumatic pains, constipation, malaria, 'absentmindedness', and female disorders. Chinese physicians discovered the painkilling properties of cannabis and the drug was applied as an analgesic during complicated surgical procedures in the second century AD. In India, the anxiolytic and euphoric properties of cannabis were discovered some time between 2,000 and 1,400 BC when it was used to concoct one of India's favorite beverages. It was further recommended for quickening the mind, lowering fevers, inducing sleep, curing dysentery, stimulating appetite, improving digestion, relieving headaches, and curing venereal disease. The medicinal effect of cannabis has also been recognized in Western societies, some millennia later. During its heyday, from 1839 to the turn of the century, more than one hundred papers appeared in the Western medical literature (Grinspoon and Bakalar, 1992). Cannabis was prescribed for various diseases and discomforts, such as insomnia, asthma, coughing, fatigue, rheumatism, delirium tremens, and pain with menstruation. It was included in the United States Pharmacopeia from 1850 to 1942 (Himmel-stein, 1983). At the turn of the twentieth century, however, the medicinal use of cannabis waned, mainly due to the development and introduction of synthetic medicines. Yet there is at present renewed scientific interest in the medicinal applications of cannabis due to the identification of the active cannabinoids and the development of synthetic analogs. Research efforts in this area will very likely increase dramatically in the very near future because of the recent discovery of cannabis receptors and a potential endogenous ligand in the human brain (Section 2.4).

Among other medicinal properties of cannabis, the Chinese recognized its euphoric and mind-altering potentials. Then as now opinions regarding these effects were divided. Some were convinced that cannabis would destroy society and others believed it would be a boon to mankind. The matter was resolved about 600 BC by the Taoist condemnation of anything that enfeebled the body. Afterwards both medicinal and recreational usages of the drug disappeared from China (Abel, 1979). In Arab lands, hashish became a commonly used drug for its euphoric properties by the thirteenth century AD. Though Cairo authorities disapproved of and tried to suppress cannabis use, it had become too ingrained to be eradicated. One tale of hashish' effects dating from those times played a major role in generating the drug's poor reputation among the general public, namely the story of the hashsba—slirin which means hashish-eaters. The Arabic name, from which the English word assassin originates, was given to a group of moslem terrorists who murdered Christians at the time of the Crusades. Although they were users of hashish, as many others in the Arab world were, Abel states that there is no evidence that the terrorists took the drug to fortify themselves for errands of death. The Arabs themselves have rarely regarded hashish as a drug leading to violence and crime. Indeed, cannabis may suppress violent behavior because of the mild lethargy it induces (Grinspoon and Bakalar, 1992). Nevertheless, this tale was readily used for anti-cannabis campaigns over subsequent centuries.

Although cannabis was primarily used by the Europeans for making rope during the Middle Ages, some were already aware of its inebriant properties. Sorcerers and witches were accused of incorporating cannabis in drugs. Pope Innocent VIII in 1484 included cannabis in the list of satanic compounds. As the euphoric properties of the drug became more widely known a group of French writers formed the 'Club des Hashischins' in Paris in the 1850s. They were fascinated by the stories of the mind-altering effects of hashish that came in the wake of Napoleon's retreat from Egypt and wrote novels under its influence, e.g. Artificial Paradises by Charles Baudelaire (1860). North Americans became aware of the drug through Bayard Taylor, writer and traveler best known for his translation of Goethe's Faust. He tried it during his visit to Egypt in 1854 and wrote two books of which The Land of the Saracens (1855) contains the most comprehensive description of his hashish experiences. Hashish acquired a lurid reputation through these writings and use of the drug waned considerably in the Western world.

A revival of cannabis use was seen in the United States with the passage of the Eighteenth Amendment in 1920 which prohibited the sale of alcohol. Cannabis began to be brought across the border from Mexico and smoked for pleasure by minority groups. The common name for any part of the cannabis plant became marijuana after the Mexican Spanish word maraguan quo meaning an inebriant plant (Maykut, 1985). Still, very little was known about the drug. Little attention was paid to marijuana use until Commissioner Ansinger of the U.S. Federal Bureau of Narcotics and Dangerous Drugs - according to Abel, in an effort to dissuade Congress from reducing the bureau's budget - started a media campaign against marijuana in the 1930s. The Bureau of Narcotics and Dangerous Drugs circulated stories about crimes attributed to marijuana use and later used the published stories as evidence in support of its policies. Cannabis was banned in the USA by the 1937 Marijuana Tax Act which required anyone using it to register and pay a tax of one dollar an ounce for industrial or medical purposes and of hundred dollars an ounce for unregistered transactions having any other purpose (Grinspoon and Bakalar, 1993). Although primarily designed to prevent 'recreational' use, it also caused cannabis to lose its medical status. In the Netherlands, cannabis was proscribed by the Opium Law of 1928 when it was still rarely used.

The second revival of marijuana use occurred during the 1960s, at the time of the 'flower power' movement when a large number of young people began to use marijuana recreationally. Many of the American users during this period were Vietnam War veterans who were introduced to cannabis and other intoxicating drugs while on duty in Southeast Asia. Cannabis' popularity became widespread, both in the United States and Western Europe, not only in lower social classes and minority groups but throughout all levels in society. Cannabis has ever since become the most widely used illegal drug in most of these countries. As a consequence, public and scientific interest in the plant and its usages increased dramatically. About 8,000 references — dealing not only with the psychic and physical effects of the drug but also with other topics such as industrial usage, cultivation, history, and legal status — to papers published prior to 1978 are collected in Abel's (1979) Comprehensive Guide to the Cannabis Literature. The number of articles has since increased by many thousands - in the years 1990-1993, more than 1,200 references to scientific articles can be found - and it is very unlikely that the rate of publications will decrease after recent breakthrough discoveries concerning the drug's pharmacology.

2.2 Prevalence of Cannabis Use

Cannabis is by far the most widely used illicit drug in most countries all over the world, including Nigeria (Morakinyo, 1983), Australia (Rankin, 1985), Israel and France (Kandel, 1984), Greenland (Pedersen, 1992), Canada (Campbell and Svenson, 1992), and Spain (Alvarez et al., 1992). Unlike alcohol and tobacco that are used by people of all ages, a preference for cannabis predominates among the young people who apparently use it less when growing older.

In the United States, marijuana usage prevalence peaked in the late 1970s and has been declining ever since. But even there marijuana is the most commonly used illicit drug (Jessor et al., 1986; Johnston et al., 1992). The most recent data about marijuana usage prevalence are available from the 17' National Survey of American High School Seniors, and the 12th National Survey of American College Students (Johnston et al., 1992). Lifetime prevalence among high school seniors declined from 60.3% in 1980 to 36.7% in 1991; and, among college students, from 65.0% to 46.3%. Thirty-day prevalence declined in both groups from about 34% in 1980 to 14% in 1991. While 9.1% of high school seniors and 7.2% of college students admitted daily use of marijuana in 1980, only about 2% of both groups did in 1991.

In the Netherlands, prevalence data have been obtained by the Youth Health Care from students in grades 7 and 8 of primary education and all levels of secondary education in 1984, 1988, and 1992 (Van der Wal, 1985; Plomp et al., 1990; De Zwart et al., 1993). A three-fold increase in cannabis prevalence was found among the total group during this period. Lifetime prevalence increased from 4.8% in 1984 to 13.6% in 1992; thirty-day prevalence, from 2.3% to 6.5%. Considering only those of 18 years or older, lifetime prevalence of cannabis increased during this period from 15.7% to 46.8% among males, and from 13.5% to 22.0% among females. Thirty-day prevalence increased from 9.2% to 17.0% among males and decreased from 8.1% to 6.1% among females.

In short, relatively more young Americans formerly used and still use cannabis than their Dutch counterparts, but the disparities between lifetime and current use prevalences are narrowing due to opposite trends in the two countries. It is perhaps important to note in this context that possession of cannabis is prohibited by law in The Netherlands, as in the United States. The seriousness of the offense is, however, determined by the amount found in the Dutch user's
possession and prosecution is unlikely to occur when that is less than 30 g (1 oz).

2.3 The Cannabis Plant and its Preparations
There is only one species of the genus Cannabis, Cannabis sativa, with two subspecies of sativa and indica, each with wild and cultivated varieties (Small and Cronquist, 1976). The subspecies sativa has been cultivated for fibre obtained from its stem to produce rope and linen and for its seeds to produce birdseed and oil for quick-drying paint. The principal interest in cannabis, however, has been in the sticky golden resin that covers the flowers and top leaves of the subspecies indica. The resin contains the psychoactive chemical compounds known as cannabinoids. The cannabinoid content of the drug varies widely depending on type of plant, climate, soil, cultivation, and part of plant. The highest concentration of cannabinoids is found in the flowering tops of the plant, followed by its leaves. Very small quantities of cannabinoids are found in the stem and roots of the plant, and none in the seeds. Female plants produce higher concentrations of cannabinoids than their male counterparts.

Roughly speaking, three grades of drug preparations exist, identified by the Indian names bhang, ganja, and charas (Maykut, 1985; Grinspoon and Bakalar, 1992) . Bhang is obtained from dried leaves and flowering tops of uncultivated plants and has a low resin content. Ganja is made from the dried leaves and flowering tops of carefully selected cultivated plants and contains larger quantities of resin with a higher quality. Charas is prepared from the resinous exudate itself, obtained from the flowering tops of mature cultivated plants. The first two grades are known as marijuana, the latter as hashish which is five to ten times stronger than marijuana. Hashish oil, a concentrated extract of hashish sometimes mixed with alcohol, is even stronger. Many street names of the drug exist, such as: grass, hemp, pot, rope, weed, and sinsemilla. The latter, literally meaning 'without seeds', is a potent and popular street preparation which results from careful manicuring of the unfertilized flowering tops of the female plant. The plant is normally cut, dried, chopped, and incorporated into cigarettes with or without tobacco. The drug can also be smoked in a waterpipe, chewed, prepared as tea, or eaten in baked goods.

2.4 Chemistry and Pharmacology

The hemp plant contains more than 400 chemical compounds. More than 60, the cannabinoids, are specific to that plant and belong to the terpenophenolic chemical class (Turner et al., 1980). The term cannabinoid is used for the typical C,1-compounds present in Cannabis sativa and includes their analogs and transformation products. Cannabinoids are very lipid soluble and water insoluble due to nitrogen lack in the molecule. The majority of the cannabinoid products are pharmacologically inactive. The main active ingredients of cannabis are cannabinol (cBN), cannabidiol (cBD) and several isomers of tetrahydrocannabinol (THc). The constituent primarily responsible for the physiological and psychological effects of cannabis is A9-THC (Gaoni and Mechoulam, 1964). Its pharmacological activity is stereoselective, the (—)-trans isomer of Y-THC being 6-100 times more potent than the (+)-trans isomer dependent upon the species as well as the pharmacological test (Dewey et al., 1984). Another physiologically active isomer, ,60-THC, is about equipotent as 6,9-THc but is of trivial practical importance because of the minute amounts available in the plant material. Unless stated otherwise, the abbreviation THC is hereafter used to refer to A9-THC. CBD is devoid of psychoactive properties (Karniol and Carlini, 1973; Belgrave et al., 1979) whereas CBN possesses about 1/10 the potency of THC in man (Hollister, 1973; Perez-Reyes et al., 1973). Although most cannabinoids do not posses psychoactive properties themselves, it is quite possible that they interact with THC in a synergistic, additive, or even antagonistic manner. It has been suggested, for example, that CBI) delays both the onset and offset of THC's effects, and either antagonizes the stimulatory or potentiates the depressant properties of THC (e.g. Karniol and Carlini, 1973).

Since THC is the most prominent psychoactive compound in cannabis its concentration mainly determines the psychotropic activity of the drug. It should be mentioned that the THC in the plant material is predominantly in the form of a TI-IC acid derivative. Only after heating, the acid is instantly and totally decarboxylated to TI-IC itself. Therefore, hemp should first be heated before eaten or drunk. When smoked, the acid derivative is totally converted to THC and only Ti lc itself appears in the smoke. The combined concentration of THC and THC acid in marijuana cigarettes varies in the United States from about 0.5 to 11% (Jaffe, 1990). In The Netherlands, seized hemp material usually contains about 10% THC, though it may range from 5 to 15% and, in exceptional cases, to 25% (Dutch Forensic Laboratory, personal communication).

Cannabis is usually smoked as a 0.5-1 g marijuana cigarette. Hashish, having a far greater potency, is often mixed with tobacco before smoking. The characteristic odor of cannabis is not due to the cannabinoids but to volatile essential oils. Tobacco and marijuana smoke are quite similar in many respects. The most obvious difference is that nicotine is present in tobacco while THC, CBD, and CBN are present in the marijuana smoke condensate (Harvey, 1984). Compared to a nicotine cigarette, smoking a marijuana cigarette delivers greater amounts of the carcinogenic products, tar and benzopyrene, to the smoker's mouth, and leads to greater retention of the inhaled tar in the lung as well as greater boosts in blood carboxyhemoglobin (Salemink, 1984; Wu etal., 1988; Tashkin etal., 1991). Davis et al. (1984) studied the smoking characteristics of marijuana cigarettes under smoking-machine conditions. They found that the humidity of the marijuana cigarette - humidification is often used to moderate 'harshness' upon smoking - and type of cigarette paper used do not alter the amount of Ttic delivered to mainstream smoke. When a complete marijuana cigarette was consumed in a single puff, 69% of the available THC was recovered in the mainstream smoke. Apparently, about 30% of the THC was destroyed by pyrolysis. When they simulated the puff duration and puff volume that many marijuana smokers practice, only 16 to 19% of the THC was found in the mainstream smoke condensate. Others have found numbers in the range of 10 to 25% (summarized by Agurell et al., 1986), the former being more characteristic of light users, the latter of heavy users. When administered orally, TI-IC's bioavailability is only about 6% due to its sensitivity to acidic gastric juice and first pass metabolism in gut and liver.

The THC dose required to produce different pharmacological effects in humans ranges from 2 to 22 mg for smoking and 20 to 90 mg when THC is administered orally (Martin, 1986). A profound psychological 'high' can be experienced after smoking a cigarette containing 10 mg THC. If only 10-20% of the available TFIC enters the circulation when the drug is smoked, the TI IC dose range would he reduced to 0.2-4.4 mg. Animal studies showed that the TI-IC level in the brain is surprisingly small, at most 1% of the administered dose at peak concentration (Agurell et al., 1986). Assuming a similar distribution in humans, only 2-44 tig THC would be expected to penetrate the brain with less than 20 Ag being necessary to produce a profound 'high'.

Although the required dose to produce pharmacological effects is quite small, huge doses of cannabinoids can be administered without causing death. The lethality of drugs is normally expressed in the 1,D 3, the dose that will cause death in 50% of the humans or animals taking it. The 1_11_ in humans is, however, not known because lethal effects of overdose by humans are nonexistent or rare. The toxicity of drugs is expressed as a therapeutic ratio of safety factor which is the ratio of lethal to effective dose. On the basis of animal studies estimates of the safety factor of THC in humans vary from 4,000 (Abel, 1979) to 40,000 (Grin-spoon and Bakalar, 1993). The safety factor for many cancer chemotherapy drugs is only 1.5 and, for alcohol, 4 to 10. In this respect, cannabis is a remarkably safe drug.
The metabolism of THC is exceedingly complex and more than 80 metabolites are known to be formed in man. After marijuana smoking or THC injection, the first metabolite, 11-hydroxy-A9-Thic (11-01-1-Thic) is formed in the lungs and liver. Its peak concentration in relation to the parent compound's is about 1:10-20 (Wall et al., 1983; Huestis et al., 1992). After oral THC the ratio is about 1:1-2. Because this metabolite's psychotropic activity is equipotent to the parent's, it contributes to the total cannabis effect, particularly when the drug is ingested.

11-0H-THC is converted by the liver into a number of inactive metabolites. The primary pathway leads to the formulation of 11-nor-Y-THc-9-carboxylic acid (Thic-cooH), the most abundant inactive metabolite in plasma, and in urine where it is partially conjugated.
Plasma concentrations of THC peak during the smoking process (Perez-Reyes et al., 1982; Huestis et al., 1992; Mathew et aI., 1993) and decline exponentially in two sequential phases. In the initial distribution (a) phase, the drug passes rapidly out of the plasma and into highly perfused fatty tissues such as liver, lung, kidney, and spleen. Although the brain also receives a high blood flow it is, on the basis of animal studies, generally assumed that only a very small proportion of THC will pass the blood-brain barrier. The first phase is followed by a much more prolonged elimination (0) phase that contributes to the accumulation of THC in poorly perfused tissues and wherein THC is redistributed from tissue into blood, metabolized and excreted in urine and feces. The a-phase half-life (t) is only about 30 minutes, whereas that of the 0-phase (t") was estimated to vary between 18 and 36 hours depending upon the individual (Wall et al., 1983; Chiang and Barnett, 1984), which would be in the same range as many psychoactive drugs such as amitriptyline, haloperidol, and nitrazepam (Agurell et al., 1986). Yet recent studies provided evidence that the earlier reported terminal elimination half-lives were underestimated due to short sampling times (Johansson et al., 1988, 1989a). In the latter study, ten marijuana smokers who habitually smoked at least one cigarette per day were asked to smoke four cigarettes during a two-day period delivering a total 'loading dose' of approximately 56 mg THC, and then to abstain from cannabis use for four weeks. Eight subjects smoked marijuana containing deuterated THC, two smoked unlabeled THC. Plasma samples were analyzed using gas chromatography followed by mass spectrometry (Gc/ms), with an improved detection limit of 20 pg/ml. The decline in THC's plasma concentrations followed the same pattern as previously observed and were less than 1 ng/ml on the second day of abstinence in all subjects except one. THC remained detectable between 10 to 15 days in the subjects who had smoked deuterated THC, and the terminal elimination half-lives ranged from 2.6 to 7.2 days (mean+sp = 4.3 + 1.6). THC remained detectable for 24 and 28 days in the other two subjects, with half-lives of 9.6 and 12.6 days, respectively.

The peak plasma concentration of 11-0H-Thic is achieved within 15-30 minutes and from there declines according to essentially the same pharmacokinetic profile as its parent. The rise in THC-COOH's plasma concentration is relatively slow, reaching an ill defined peak in different individuals within 1-2 hours. Its elimination follows a monoexponential profile with various individuals showing L„3's from less then 24 to more than 72 hours. Given the above evidence for prolonged terminal elimination half-lives for THC, one should not be surprised that future studies may show that these periods were also underestimated.

Despite what might be expected from the elimination half-lives, there is hardly any accumulation of THC in the blood. Nahas et al. (1981) demonstrated that after repeated intramuscular injections of 6,8-THC in rats accumulation does occur in neutral fat and liver but not in blood and brain. In man, THC was found in fat biopsies four weeks after drug administration (Johansson et al., 1989b) and THC metabolites were present in urine for several days to more than a month after the cessation of marihuana use (Hollister and Kanter, 1980; Dackis et al., 1982; Ellis et al., 1985). Therefore, the presence of cannabinoids in urine provides little information concerning the time of last drug use. Urine screening may be sufficient in epidemiological studies investigating the extent of use of cannabis in selected population groups. But for an indication of intoxication in studies to show the role of cannabis as a cause of road traffic accidents, plasma samples need to be taken and assayed for THC. Radioimmunoassay is widely applied in qualitatively analyzing urine samples, GC/MS is the procedure of choice when plasma samples need to be quantitatively analyzed (Cook, 1986).

Though peak concentrations of THC are achieved during smoking, the maximum psychological effect (or 'high') occurs 15-30 minutes after its cessation. This temporal dissociation between plasma concentration and effect suggests that brain concentrations increase as plasma concentrations decrease, possibly due to slow penetration of the blood-brain barrier, slow distribution within the brain, and a lag-time in pharmacological activity (Agurell etal., 1986). Both peak plasma concentrations and maximum psychological 'high' are roughly proportional to the inhaled THC dose, but correlations between these parameters measured simultaneously at times 3-240 min after the cessation of smoking are, albeit significant, not especially strong. For example, Ohlsson et al. (1980) found the overall correlation for repeated measurements obtained from 11 experienced smokers to be r = 0.53. After four hours, the psychological 'high' had vanished and plasma THC levels were very low. Great interindividual variation exists in plasma levels of THC after smoking and this variation cannot be attributed to the regularity of cannabis use (Lindgren etal., 1981). It was even noted when a paced smoking protocol was used (Huestis et al., 1992).

Available evidence leads to the conclusion that it is usually impossible to predict the psychological effects of THC from its determination in a single plasma sample. But this is not the same as saying that no biological index of cannabis intoxication will ever be found. One possible candidate is THC's inactive metabolite THC-COOH. The relationship between this metabolite's plasma concentration and the perceived 'high' after cannabis smoking has never been defined, although both parameters were measured in the study by Perez-Reyes et al. (1982). Peak and time integrated THC-COOH concentrations were proportional to the administered THC doses. Interestingly, the occurrence of the peak THC-COOH concentration coincided in time with the subjects' report of maximum 'high'. The authors failed, however, to measure, or at least report, the correlation between plasma THC-COOH concentration and subjective feelings because of the metabolite's pharmacological inactivity. Yet this coincidence might signify a useful epiphenomenal correlation. This possibility was repeatedly explored in this program.

Until about a decade ago, the mechanisms by which THC exerts its effects on the CNS were poorly understood. One of the problems was that cannabinoids affect almost every system in which they are examined (Dewey, 1986). THC partitions into biological membranes and affects many membrane related functions including neurotransmitter uptake systems, enzymes, and receptors. It has long been hypothesized that cannabinoids alter membrane properties nonspecifically without interacting with any specific membrane receptor (e.g. Roth and Williams, 1979). But during the 1980s evidence for the possible existence of receptors slowly emerged. Both the stereoselectivity of cannabinoids (Dewey et al., 1984) and the observation that modest structural modifications of the THC molecule results in profound changes in behavioral activity (Razdan, 1986) argued for an receptor-mediated mechanism of action. Martin (1986) reviewed the literature on the cellular effects of cannabinoids and was tempted to think of receptor-mediated effects because unique pharmacological effects occur at reasonably low drug concentrations. The diverse effects of cannabinoids on enzymes as well as neurotransmitter and opioid receptors at high concentrations would seem to be produced by a general perturbation of membranes.
Understanding of the mechanism of action of cannabinoids rapidly progressed by using a class of high-potency synthetic compounds originally developed for their analgetic properties such as levonantradol, its active metabolite desacetyllevonantradol, and CP-55,940 (Johnson and Melvin, 1986). Howlett and her colleagues (Howlett, 1985, 1987; Howlett et al., 1988) showed that centrally acting cannabinoids attenuate cyclic AMP accumulation by inhibition of adenylate cyclase, and that the ability of inhibition caused by a series of synthesized cannabinoids occurred in a reversible, cell type-specific, potent, and stereoselective manner. The same group of researchers identified and characterized a high-affinity, stereoselective, pharmacologically distinct receptor in membranes from rat brain by using tritium-labeled CP-55,940 (Devane et al., 1988). These receptors are coupled to adenylate cyclase via the inhibitory guanine-nucleotide protein, G„ and thereby inhibit cyclic AMP production. Furthermore, they are involved in the regulation of K+ and Ca' currents in neuronal cells (Abood and Martin, 1992).

Herkenham et al. (1990) localized cannabinoid receptors in brain sections of rat, guinea pig, dog, and rhesus monkey, as well as in human brains obtained from people who died from nonneurological disorders. Autoradiography using [3MCP-55,940 revealed similar distributions of binding in all species with the greatest abundance of binding sites in the globus pallidus, substantia nigra pars reticulata, and the molecular layers of the cerebellum and the hippocampal dentate gyrus. Receptors were also dense in the cerebral cortex, striatum, and the remainder of the hippocampal formation. Comparatively sparse binding was found in lower brain stem areas and spinal cord. Identical receptors were also found in human testis (Gerard et aL, 1991), and human spleen, tonsils and peripheral blood leukocytes (Bouaboula et al., 1993), albeit in smaller concentrations than in the brain. Cannabinoid receptor genes have now been cloned in both rat (Matsuda et al., 1990) and human (Gerard et al., 1991), and the deduced amino acid sequences exhibit more than 97% identity. The latest breakthrough is the isolation of a potential endogenous ligand for the cannabinoid receptor (Devane etal., 1992). The compound was named 'anandamide' from the Sanskrit word ananda, meaning bliss, and from its chemical name, arachidonylethanolamide.

The single essential commonality of recreational and abused drugs is that they all act on the brain reward system. It is hypothesized that this action results in the 'high', 'rush' or 'hit' sought by its users (Gardner and Lowinson, 1991). Herkenham et al. (1991) failed to discover cannabinoid receptors on the dopaminergic neurons comprising the brain 'reward' system, suggesting that cannabinoids do not directly act there. Animal studies showed, however, that THC lowers brain reward thresholds in the medial forebrain bundle, thereby enhancing dopamine release in the reward system (Gardner et al., 1988; Gardner and Lowinson, 1991). Both effects were reversed by naloxone, an opiate receptor antagonist, suggesting that THC acting at its own receptor causes the release of an endogenous opioid which in turn acts upon j or other receptors causing the euphoric effect. Since THC is not itself an opioid receptor agonist one should not expect it to produce the same dependency and withdrawal syndrome upon abstinence as if it were. Nonetheless the modestly euphoric and anxiolytic effects of releasing an endogenous opioid might be something heavy cannabis users wish to perpetuate. They could become psychologically dependent upon this process as the means for coping with stress or simply enriching an objectively dull life.
The recent major steps forward in understanding the pharmacology of cannabinoids tempted many investigators (Herkenham etal., 1990; Gerard et al., 1991; Abood and Martin, 1992) to speculate on the causal role cannabinoid receptors play in the multiplicity of effects observed after ingestion or inhalation of cannabinoids. The high density of receptors in the cortex may explain the diverse cognitive effects found after cannabis smoking (below). The hippocampus plays a crucial role in memory consolidation so it should come as no surprise that THC affects short-term memory. Receptors in the basal ganglia and cerebellum may explain the effects of cannabinoids on movement control; and, those in human testis, the depression of reproductive functions observed after chronic cannabis use. The low toxicity of cannabinoids probably reflects the paucity of receptors in medullary nuclei that mediate respiratory and cardiovascular functions. Similarly, the deficit of receptors in these areas and also the mesen- cephalic reticular formation and posterior diencephalon would not lead one to expect a primary influence of cannabinoids upon the major CNS arousal systems that arise there.

2.5 Effects of Cannabis Use

The cannabinoids comprise a unique pharmacological class of compounds producing a multiplicity of effects. They produce mixed stimulation and depression of CNS activity in different areas and therefore partially mimic the activities of other centrally acting drugs, including stimulants, sedatives, tranquilizers, and hallucinogens. However, the degree to which cannabinoids' effects resemble those of any other single class of psychoactive drugs is too low to allow their joint classification (Consroe et al., 1976; Razdan, 1986). In addition, THC has also been wrongly classified as a narcotic and should be viewed as distinct from the opiates in this respect (National Commission on Marihuana and Drug Abuse, 1972).

2.5.1    Acute Effects

THC is rapidly absorbed during cannabis smoking, and its acute effects appear shortly thereafter; i.e. within 15 minutes with peak effects occurring between 30 and 60 minutes. Acute subjective effects are dose-dependent and generally last for about two to four hours. Nonetheless, performance decrements may persist for several hours after the feeling of intoxication has passed (Barnett et al., 1985). Intravenous (i.v.) injection of THC produces a similar profile of effects as after smoking (Ohlsson et al., 1980; Lindgren et al., 1981). When cannabis is ingested, THC's onset of action is some hours delayed and subjective effects last for 5-12 hours without a clear peak, which is consistent with the drug's pharmacokinetics. It is still uncertain whether THC adversely affects performance during the day after cannabis smoking. Some studies provided evidence for a 'hangover' syndrome (Yesavage et aL, 1985; Heishman et al., 1990; Leirer et al., 1991), but others demonstrating acute performance impairment failed to reveal the existence of clinically significant residual effects the following day (Rafaelsen et al., 1973a, 1973b; Barnett et al., 1985; Leirer et al., 1989; Chait et al., 1985; Chait, 1990).

An early clinical account of cannabis intoxication was given by Bromberg (1934), a psychiatrist, on the basis of his own experience and observations of, and discussions with, other users.
"The intoxication is initiated by a period of anxiety within 10 to 30 minutes after smoking, in which the user sometimes .. . develops fears of death and anxieties of vague nature associated with restlessness and hyperactivity. Within a few minutes he begins to feel more calm and soon develops definite euphoria; he becomes talkative . . . is elated, exhilarated . . begins to have an astounding feeling of lightness of the limbs and body . . . laughs uncontrollably and explosively . . . without at times the slightest provocation has the impression that his conversation is witty, brilliant . . . The rapid flow of ideas gives the impression of brilliance of thought and observation  [but] confusion appears on trying to remember what was thought. . . he may begin to see visual hallucinations . . . flashes of light or amorphous forms of vivid colors which evolve and develop into geometric figures, shapes, human faces, and pictures of great complexity . . . After a longer or shorter time, lasting up to two hours, the smoker becomes drowsy, falls into a dreamless sleep and awakens with no physiologic after-effects and with a clear memory of what happened during the intoxication" (Bromberg, quoted by Grinspoon and Bakalar, 1992).

Grinspoon and Bakalar considered Bromberg's account as a composite, exaggerated and overinclusive description of cannabis 'highs'. Nevertheless, it clearly demonstrates how diverse the subjective effects of cannabis can be. After low doses of THC, the effects are subtle and of short duration, and intoxication is generally undetectable to the observer. After high doses, intoxication may still be hardly noticeable but the most reliably produced physiological signs include reddening of conjunctivae due to dilatation of blood vessels and increased heart rate with a concomitant peripheral vasodilation (e.g. Weil et al., 1968; Benowitz and Jones, 1981; Hollister et al., 1981; Maykut, 1985). Blood pressure slightly increases in the supine position but decreases upon standing, which may result in orthostatic hypotension and syncope in some individuals (Maykut, 1985). THC exaggerates task-elicited tachycardia as well as mean arterial blood pressure, suggesting that it may increase cardiovascular responsivity (Capriotti etal., 1988). Respiratory depression, bronchodilation, and decreases in intraocular pressure, salivary flow, skin but not oral temperature, grip strength, and REM sleep have also been observed (Hollister, 1971a, 1986; Jones et al., 1981). Despite anecdotal reports, pupil size is not affected by the drug (Weil et al., 1968), Neither are the electroencephalogram and deep tendon reflexes (Dewey, 1986; Hollister, 1986). Intravenous administration of THC produces variable changes in total cerebral glucose metabolism but consistently increases cerebellar metabolism (Volkow et al., 1991). Mathew and associates showed that cerebral blood flow decreases in all brain regions in inexperienced smokers but increases in some brain areas, particularly the frontal lobe, in experienced smokers (Mathew etal., 1989, 1993).

The principal psychological effect of cannabis is the 'high' which is sometimes a dream-like, euphoric state. The user's mood may vary from exhilaration to quiet introspection. The euphoric state is sometimes accompanied by mild state anxiety, tension, anger, and confusion (Mathew et al., 1993). Time perception is consistently changed. Time seems to slow down so that minutes may seem like hours (Hollister and Gillespie, 1970; Bech et al., 1973; Borg et al., 1975; Hicks et al., 1984). Temporal disintegration is associated to the altered time sense as another sign of cannabis intoxication (Melges et al., 1970a; Mathew et aL, 1993). The latter was defined by Melges et al. as the difficulty to retain, coordinate, and serially index those memories, perceptions, and expectations that are relevant to the goal one is pursuing. The user has difficulty concentrating and thinking is troubled by the intrusion of thoughts with more unusual associations, which may result in disorganized speech. Both studies showed further depersonalization, i.e. an alteration in the perception or experience of the self whereby the usual sense of one's own reality is temporarily lost or changed (Dsm-III-R). Depersonalization can be prominent, particularly after consuming high doses of TI-IC. Users can become detached observers of their own intoxication, possibly explaining why the more experienced may appear normal in public. Depersonalization caused by other factors is often said to be a cognitive defense against dysphoric mood states such as anxiety and depression. However, many users react to the combined experience of temporal disintegration and depersonalization — symptoms that are also produced by psychotomimetics — with euphoria. Mathew et al. (1993) therefore concluded that it is unlikely cannabis-induced depersonalization serves this function. Furthermore, they and many others have demonstrated that all of the above-mentioned aberrations are transient: normal self-perception returns within a few hours after smoking.

As cannabis generally intensifies a user's prevailing mood, dysphoric reactions may also occur. Some individuals, particularly naive users or more experienced users consuming an unexpectedly high dose, may exhibit anxiety which is sometimes accompanied by paranoid thoughts. Anxiety may become so severe as to induce panic reactions, probably due to the user's fear to lose control of thinking and actions, and that these effects might never wear off. Due to a distorted perception of the body, the anxious user may think he is becoming insane or undergoing a life-threatening physical catastrophe. Although relatively uncommon, such panic reactions are probably the most frequent adverse reaction to moderate cannabis use (Grinspoon and Bakalar). If the user has some degree of underlying depression, cannabis may produce an acute depressive episode. This reaction is, however, rarely seen in experienced users. Another infrequently reported adverse reaction is the occurrence of a psychotic syndrome, described by Chopra and Smith (1974) as lying on a continuum from "an acute confusional state" to "a full-blown toxic psychosis." This acute organic psychosis is more likely to occur after consumption of an extremely high cannabis dose, particularly in individuals with a previous history of psychosis (Chaudry et al., 1991; Mathers and Ghodse, 1992). The syndrome resembles the delirium of high fever and includes disorientation, confusion, and both auditory and visual hallucinations (Weil, 1970; Maykut, 1985; Wert and Raulin, 1986). It can be treated effectively with cannabis withdrawal • or antipsychotic drug administration, without residual effects (Chaudhury et al., 1989; Chaudry et al., 1991; Van Brussel, 1993). Cannabis may precipitate schizophreniform episodes (Thornicroft, 1990) but, as the drug may exacerbate pre-existing mental illness, these are usually relapses of known cases (Tunving, 1985; Hollister, 1988).

Disruption of memory processes is the single most consistently reported cognitive deficit following cannabis use. Marijuana smoking impaired immediate free recall of digits (Tinklenberg et al., 1970; Heishman et al., 1989, 1990), words (Abel, 1971; Chait et al., 1985), prose (Miller et al., 1977a), and picture/word combinations (Miller et al., 1977b). Hooker and Jones (1987) found an increased number of short story omissions following marijuana smoking, and intrusions occurred in delayed free recall. However, neither immediate and sustained attention nor controlled retrieval from semantic memory were affected. Darley et al. (1974) found that subjects treated with drug and placebo did not differ with respect to either their delayed recall or delayed recognition of word lists that were learned four days earlier. This means that THC impairs acquisition or consolidation but not retrieval processes.

The recent breakthrough in localizing the sites of TI-IC activity within the brain, coupled with earlier psychological concepts of the relationship between working memory and the temporal organization of behavior, may offer an explanation for the most prominent cognitive deficit in THC intoxication. Working memory is conceived as the brain's limited-capacity 'work space' for temporary storage and processing of sensory information in relation to that retrieved from long-term memory (Baddely and Hitch, 1974). Melges et al. (19706) demonstrated that working memory impairment is at least partially responsible for temporal disorganization of behavior: errors in their Goal-Directed Serial Alternation Task were mainly due to serially organizing and retaining information arriving in working memory. This finding suggests a crucial locus within the CNS which accepts incoming information, serially encodes it with respect to time of arrival and holds it long enough to form associations with information elicited from long-term memory. Abundant evidence from neuropsychological research indicates the hippocampus is that crucial coordinating structure (Gray, 1982). The localization of a high density of cannabinoid receptors within the hippocampus encourages one to speculate concerning the drug's probably related amnestic and temporally disorientating properties. As mentioned above, the disruption of organized consciousness by the intrusion of irrelevant thoughts is quite common following cannabis use. Since the hippocampus functions to inhibit the inappropriate recall of long-term memory associations as well as facilitating the recall of appropriate ones (Valzelli, 1980), it is tempting to speculate that cannabinoid receptor agonists, including THC, reduce the internal inhibition of associations. The intrusion of inappropriate associations in working memory would result in impaired learning (Hooker and Jones, 1987).

THC can have other disinhibiting effects on information processing. The Stroop Color-Word Test was specifically designed to measure a subject's ability to inhibit a stereotyped, but under the circumstances, inappropriate response. In this test, the names of colors are rapidly presented in compatible or incompatible colors. An example of the latter would be the presentation of the word 'red' in a blue color. The subject has to inhibit the stereotyped response of reading ('red'), and instead, specify the color of the characters (blue). Disinhibitory drugs cause the tendency to revert more often to the stereotyped response. THC is disinhibitory: Hooker and Jones found that THC produced significantly more erroneous stereotyped responses, while independent measures of word reading and color naming speed were not affected.

Besides affecting memory processes and time estimation, several reviews demonstrated that THC produces performance decrements in many different tests of perceptual, cognitive, and motor skills, such as signal detection, attention, motor coordination, and reaction speed (Moskowitz, 1985; Murray, 1985; Chesher, 1986). The effects of THC on skills important for driving, as measured in the laboratory and in driving simulators, are more extensively discussed in the next chapter.

Many users of cannabis claim that the drug heightens their sensitivity to external stimuli, sharpens their vision (though with many visual distortions), makes colors appear brighter, enhances the appreciation of music, and reveals details that are normally neglected. In the words of Grinspoon and Bakalar "It is as though the cannabis-intoxicated adult perceives the world with some of the newness, wonder, curiosity, and excitement of a child (p. 237)." Possibly these are the positive aspects of perceptual disinhibition, allowing the formation of associations that have been long since relegated to the 'irrelevant' by the process of habituation. As a result, some people believe that they become more creative and perform better while under the influence of the drug. Notwithstanding the sincerity of their claims, enhanced creativity under the influence of THC has never been demonstrated in a manner that would withstand scientific scrutiny.

Another argument advanced by THC advocates is that the drug increases appetite (Tart, 1970). This assertion was repeatedly investigated and confirmed in laboratory settings, i.e. acute administration of marijuana selectively increased food consumption (Weil et al., 1968; Hollister, 1971b; Foltin et al., 1986, 1988; Kelly et al., 1990). Weil and his colleagues measured but failed to find concomitant changes in blood glucose levels and therefore postulated a central rather than peripheral physiological trigger for increased appetite. Foltin and his colleagues found that the total increase in caloric intake on days that subjects smoked marijuana was not attributable to eating during regularly scheduled meals but rather to ad lib eating between meals. Apparently, THC has less effect upon mechanisms of hunger and satiety than it does upon those that determine the positive hedonic value of certain foods. As a result of increased food intake, body weight increases also but to a greater extent than predictable on the basis of caloric intake alone (Foltin et al., 1988).

Cannabis is sometimes referred to as an hallucinogen. Many of its perceptual and cognitive effects are also produced by lysergic acid diethylamide (BD), including depersonalization, temporal disintegration, and distortions in color vision. Furthermore, both drugs may incidentally produce anxiety and paranoid reactions. But, according to Grinspoon and Bakalar, there are also important differences between subjective effects of the two drugs. LSD's are far less controllable by the individual. The so-called 'bad trip' — an experience of agonizingly nightmarish reactions — is far more common among LSD users, even experienced ones, than among cannabis users. But there is a more important difference between cannabis and hallucinogens. Except after very high doses, whatever the cannabis user subjectively experiences as a unique drug effect is superimposed on the normal stream of consciousness. What the LSD user experiences sometimes replaces it. Grinspoon and Bakalar therefore concluded that it is unlikely that cannabis, when consumed in normal doses, will produce true hallucinations. After very high doses of marijuana, however, hallucinogenic trips that approach those of LSD may occur (Tart, 1970; Fabian and Fishkin, 1981).

Cannabis has been used as a stimulant for work and a depressant for relaxation (National Commission on Marihuana and Drug Abuse, 1972; Murray, 1985; Van Ree and Essenveld, 1985). It is sometimes said that the drug-induced euphoric state is followed by drowsiness (Maykut, 1985; Hollister, 1986). But this opinion seems questionable from the failure to locate dense concentrations of THC receptors within the brain's major arousal generators, mentioned above, and the lack of consistent EEG signs of diminished electrocortical arousal following acute drug administration (Dewey, 1986). Chait et al. (1985) found no TI IC effect on reported quality of sleep, sleep onset latency, or number of awakenings during the night after marijuana smoking. In a similar study, but now using multiple doses over several consecutive days, Chait (1990) found a significant effect on only one of the four factors measured by the Leeds Sleep Evaluation Questionnaire: "getting to sleep" was easier after continual marijuana than placebo smoking. This effect seems more attributable to increased relaxation than sedation. Jones (1971) showed that the occurrence of drowsiness may be contingent upon the social setting: when tested individually, subjects demonstrated slight drowsiness but, when tested in a group situation, there was a marked lack of sedation. Furthermore, Jones noted that symptoms indicative of drowsiness, such as feeling sluggish or sleepy, were more often reported in the placebo than active drug condition. Thus, there is no convincing evidence that THC itself produces pronounced sedation.

2.5.2    Factors Influencing the Drug's Acute Effects

The wide range of different individuals' reactions after cannabis smoking is well known. But the same individual can also react inconsistently after smoking on different occasions. Several factors responsible for inter- and intrasubject variability have been identified. They include the route of administration, the THC dose, the smoking technique, the drug's potency, the individual's previous drug experience, and the psychological set and physical or social setting.

It is obvious that the THC dose and route of administration are important factors in determining the timing and magnitude of acute effects. As mentioned above, increasing THC doses not only increase all acute effects but may eventually lead to qualitatively different reactions. Despite every effort to control the dose via a chosen route of administration, it is usually quite difficult to ensure that the quantity of TI-IC entering the body, much less the brain, will be the same in different individuals or the same tested repeatedly. Nearly perfect control over the administered amount of THC is achieved when it is injected intravenously. This route of administration has, however, some important disadvantages (Dewey, 1986). First, the cannabinoids are very lipid soluble and posses hydrophobic properties. They are difficult to handle because of their high viscosity, illustrated by Dewey's term "rubber cement", making it difficult to accurately prepare small doses. Secondly, different vehicles have been used to put THC into solution, including ethanol and many others. Since the vehicle might interact with TI-IC in different ways, discrepancies in effects can easily appear when different vehicles are used. Finally, and due to these difficulties, cannabis users do not inject their drug. Using this route of administration in experimental studies with humans may therefore lead to results that are not representative of the naturally occurring events.

Administering THC by the oral route provides good control of the administered dose but much less over the amount of THc entering the systemic circulation, due to the great interindividual variability in rate of absorption from the digestive tract and first-pass metabolism in gut and liver. Another disadvantage of oral dosing is the difficulty it imposes on titrating the plasma/brain concentration to achieve the desired effect: underdosing with no effect and overdosing with adverse reactions would be the norm rather than the exceptions. But the most important disadvantage is again that it is not the route of drug administration preferred by the majority of cannabis users. Results obtained using this approach are unlikely to be representative of THC's effects in real life.

Cannabis smoking is generally associated with a considerably different smoking topography than that observed during tobacco smoking, including larger puff volumes and somewhat larger inhalation volumes (Wu et al., 1988). Smokers of cannabis also tend to hold the smoke in their lungs for longer periods of time, typically 10 to 15 seconds, owing to their belief that prolonged breathholding increases THC absorption and, consequentially, the subjective effects (Perez-Reyes et al., 1981, 1982; Wu et al., 1988). Zacny and Chait (1989) measured the psychological and physiological effects of THC under three different breathholding conditions, 0, 10 and 20 seconds, while maintaining the inhaled volumes of smoke and ambient air constant. Though the 0-second condition implied that smoke should be immediately exhaled after inhalation, the actual duration that smoke was in the lungs was estimated by the investigators as probably longer than five seconds. Compared to baseline, THC produced significant elevations on the primary measures - perceived 'high', heart rate and expired air carbon monoxide (co) - but the effects were not significantly related to breathhold duration. The study was replicated by the same investigators but now with the addition of a placebo condition and the omission of the 10-second breathholding condition (Zacny and Chait, 1991). Compared to placebo, THC increased heart rate and subjective effects. Prolonged breathholding magnified the heart rate elevations moderately (about 20%), albeit with borderline statistical significance, but had no effect on subjective measures. In this study, expired air co did increase with prolonged breathholding but to a similar extent in both the placebo and active drug conditions, indicating that THC does not affect alveolar absorption of co. Tashkin et al. (1991) executed a similar study, but they also measured drug plasma concentrations. They examined the effects of breathhold duration (4 or 14 seconds), puff volume (70 or 45 ml), and number of puffs (6 or 10) on the amount of inhaled tar, tar retention in the lung, blood carboxyhemoglobin, plasma THC concentration, heart rate, and subjective 'high'. Changes in puff volume and number of puffs had no effect on any of these measures if cumulative puff volume was held constant. Prolonged breathholding had no effect on the amount of inhaled tar, but did produce greater retention of tar in the lungs and greater blood carboxyhemoglobin concentrations. Furthermore, it produced significantly higher plasma concentrations of THC two minutes after completion of smoking, accompanied by somewhat greater elevations in heart rate and slight increases in subjective 'high'. Block et al. (1992) measured the acute effects of smoking marijuana cigarettes containing 19 mg , and placebo, on a wide variety of cognitive and psychomotor tests. They also examined the effects of breathhold duration, i.e. 7 and 15 seconds. As the latter was only a secondary goal of their investigation, they failed to assess other parameters of the subjects' smoking behavior. Nor did they measure heart rate, plasma concentrations of THC, or co absorption. The results showed significant performance decrements in most tests following THC. Subjects' performance was worse after long than short breathhold duration, but this was independent of the material smoked. In other words, prolonged breathhold duration did not potentiate THC's adverse effects. Together, these studies show that prolonged breathholding produces increased absorption of THC, somewhat greater tachycardia, but only slightly greater subjective effects and hardly any larger performance deficits.
Thus, it seems unnecessary to control breathholding time in experimental studies.

As mentioned earlier, marijuana's THC concentration, or potency, is increasing rapidly. It is therefore important to examine whether marijuana smokers modify their smoking behavior depending on the potency of the drug, like tobacco smokers do depending on a cigarette's nicotine yield (Herning et al., 1981; Gust and Pickens, 1982). If individual cannabis smokers always titrate plasma/brain concentrations to the same levels, the magnitude of effects produced by different potencies should be the same in any given one. Yet there are reasons to believe that this is not the case. Heavy users showed pronounced subjective effects to placebo cigarettes in Jones' study (below) and several others failed to reveal significant changes in smoking patterns as a function of THC concentration. Cappell et al. (1973) allowed subjects to smoke as many marijuana cigarettes, containing either 0.2, 0.4, or 0.8% THC, as they needed to achieve a 'nice high'. The estimated weight of marijuana consumed was inversely related to drug potency but the 4-fold increase from lowest to highest potency yielded only a 27% decrease in marijuana consumption. Puff duration and interpuff interval were nearly identical in the three conditions but cumulative breathhold duration and number of puffs were inversely related to potency. Though consistent with a titration process, the latter results were not significant. Similar results were achieved when the study was replicated with potencies of 0.36, 0.73, and 1.45% (Cappell and Pliner, 1974). Perez-Reyes et al. (1982) observed smoking patterns when experienced smokers consumed marijuana cigarettes containing 1.32, 1.97, and 2.54% TI-IC but also failed to show significant changes as a function of potency. These subjects inhaled more THC with higher potencies resulting in higher plasma levels of cannabinoids as well as greater psychological and physiological reactions.

Jr. these studies, the investigators measured puff duration and number of puffs but failed to obtain volumetric measures. Herning et al. (1986) did while heavy smokers consumed marijuana cigarettes containing 1.2 and 3.9% THC. Their interpuff interval and number of puffs increased while smoking the high potency cigarette, resulting in a prolonged smoking period. They did not alter puff duration but inhaled a larger volume of air, thereby diluting the marijuana smoke. Although these smokers adjusted their smoking pattern they still did not titrate THC because the cumulative puff volume of the high potency cigarette was twice that of the low potency. Chait (1989) supposed that the failure of previous studies to demonstrate true titration might be attributable to the subjects' unfamiliarity with the different potencies presented in the cigarettes given by the investigators. In his study, subjects were given the opportunity to experience the respective potencies (0.9, 1.7, and 2.7%) of each color coded marijuana type as they smoked them on five separate occasions. In all cases these subjects were allowed to self-administer as much marijuana as they wanted over a 30—min period. Yet the subjects' THC consumption, as measured by several reactions, increased as a function of cigarette potency. They did not regulate the dose in spite of their knowledge of the different potencies.

In contrast to previous results, subjects in a study by Heishman et al. (1989) appeared to adjust their smoking pattern depending on the marijuana potency. The high potency cigarette (2.7% THC) was smoked with a shorter puff duration and a lower puff and inhalation volume than the low potency cigarette (1.3% THc). Interestingly, smoking patterns began to diverge after the second or third puff. Whether the subjects' ability to adjust their smoking pattern with higher potency cigarettes was due to pharmacological factors is, however, doubtful. They showed it more unpleasant to smoke the high potency cigarettes owing to 'harshness', which may have strongly contributed to the observed differences. In conclusion, there is only weak evidence that marijuana users adjust their doses of inhaled THC so as to titrate plasma/brain concentrations to a desired level. But before closing the issue, it should be recognized that the range of potencies used in experimental studies was always narrow. The smokers may have been unable to discriminate among them and titrate successfully. Users in real life are not only confronted with a wider range of potencies. They are usually more aware of the potency of the material they smoke, from prior experience, upon the advice of other users or even from its 'street' price. It might well be that ordinary users regulate the consumed dose more accurately than shown by these studies.

In many studies in which THC was administered by smoking, subjects were allowed to smoke the cigarettes ad lib, i.e. in their customary fashion and as much of the administered dose as they wished. This was done to keep the experimental setting as natural as possible. Those investigators who applied strict smoking procedures, with or without the use of special devices, did so in order to exert greater control on the amount of THC inhaled by the subjects. Besides being less natural and probably also annoying, the latter procedure may result in large numbers of subjects reporting discomfort and dysphoric reactions (Mathew et al., 1989). The major drawback of the ad lib procedure is, of course, that one is less certain of how much THC was actually delivered. Cumulative puff volume may be used to estimate the inhaled dose, but measuring it is probably as troublesome to the subjects as the rigidly structured smoking procedures. Direct comparisons between the observed effects of THC in different studies can only be reasonably made if investigators measure and report at least the administered dose, but preferably also drug plasma concentrations. The total amount of smoked THC can be estimated gravimetrically by multiplying the potency of the drug by the weight of the smoked portion. This is a legitimate procedure: it has been shown that the THC concentrations in the unsmoked portions of marijuana cigarettes are very similar to those found in the unlit cigarette (Perez-Reyes et al., 1982). One is tempted to conclude from the results reported in the literature that the different smoking procedures do not produce disparate results. The main reason is that THC concentrations in plasma show great interindividual variability whether an ad lib or paced smoking protocol was used. Apparently, individual differences in absorption, distribution, and metabolism of THC have a greater impact on the observed plasma concentrations than the smoking procedure. Yet the two procedures have never been directly compared in a single study. A conclusive answer concerning the differential effects of smoking procedures on the delivery of THC or its effects can not be provided for that reason.

A factor that has probably a much greater impact on the severity of intoxication and the extent of performance deficits than the smoking technique is the individual's previous cannabis experience. While light and heavy users — light meaning usually not more than once a month, and, heavy, at least once daily — reported similar subjective 'highs' following THC administration by either smoking or i.v. injection (Jones, 1971; Perez-Reyes et al., 1974; Lindgren et al., 1981), nonusers experienced significantly less effects than users (Weil et al., 1968; Casswell and Marks, 1973; Milstein et al., 1975). These observations agree with anecdotal reports of chronic users that a novice user has to 'learn to get high'. Weil and his colleagues advanced two alternative interpretations of this phenomenon. The first is that 'reverse tolerance' or some sort of pharmacological sensitization occurs after repeated exposure to cannabis, but no study has provided evidence for this hypothesis (Hollister, 1986). Another possibility hypothesized by Weil et al. is that novice users are psychologically inhibited from experiencing any effect, and that the inhibition is reduced after repeated exposure to cannabis. A more likely explanation, though related to the latter, is that users and nonusers have different expectancies that modify their reactions. This possibility is discussed below.

Whereas subjective measures of intoxication showed consistently greater effects in users than nonusers, objective tests of cognitive and psychomotor performance have provided inconsistent results. Experienced users exhibited similar impairments in cognitive functioning (Casswell and Marks, 1973), greater impairments in perceptual-motor tasks (Milstein et al., 1975), and less impairment in a divided attention task (Marks and MacAvoy, 1989). Compared to light or casual users, however, heavy users showed consistently less impairment in digit symbol substitution, complex reaction time, and continuous performance tests following active drug administration (Jones, 1971; Meyer et al., 1971). These observations seem to confirm the conviction of many experienced marijuana users that they can volitionally control THC's effects (Tart, 1970). The notion is supported by the common observation that the total effects of marijuana smoking contain a large placebo component (below). It was experimentally tested by Cappell and Pliner (1973). Twenty subjects smoked a marijuana cigarette containing 12 mg THC or placebo on two separate occasions. Before smoking, baseline measurements were taken of heart rate and performance in four tasks, i.e. time estimation, immediate recall of words, solving arithmetic problems, and backward digit span. After smoking, the subjects rated their level of intoxication. At this point, the major experimental manipulation was introduced. Unbeknown to them, the subjects were equally divided between two groups matched on the basis of self-reported current marijuana use. The low-motivation group was simply required to repeat the same tasks. The high-motivation group received the instruction to repeat the tasks while trying as hard as possible to overcome the drug's interfering effects. After they were over, both groups again rated their intoxication. Instructions had no effect on heart rate or feelings of intoxication. Arithmetic and backward digit span were unaffected by THC and so showed no effects of instructions. Immediate recall and time estimation were impaired in the low-motivation group and memory was likewise impaired in the others. However, the high-motivation group showed significantly less drug effect on time estimation. This result shows that some volitional control of THC's effects is possible. It also indicates that the reasons large THC effects on performance have been found in some experiments but not others is not only related to differences in the investigators' testing procedures but also to their respective subjects' motivation for overcoming those effects by compensatory effort.

As already mentioned, the high variability of cannabis' effects between individuals may be partially attributable to differences in the expectations of the user. Jones (1971), for example, found that ratings of psychological 'high' were greater in heavy than light users following placebo but similar following the active drug. This suggests that heavy users tended more to respond to cues as smell and taste which were similar for both placebo and active marijuana. This interpretation was further strengthened by the lack of a subjective effect in both heavy and light users following the oral ingestion of placebo marijuana that had a disagreeable taste. Cami et al. (1991) also showed that expectancy may influence smoking behavior. Subjects who expected and received active drug (hashish incorporated into a tobacco cigarette) showed higher plasma levels of both THC and THC-COOH as well as faster metabolism than subjects who had not expected but received active drug. Furthermore, the elevation in heart rate was significantly greater in the former group who also experienced more marked subjective effects. Interestingly, another group of subjects who expected active drug but received placebo, showed pronounced subjective effects. A similarly strong placebo effect was found by Chait and Perry (1992) . Their regular marijuana smokers were allowed to smoke placebo cigarettes freely for 60 minutes during four identical weekly sessions. Some were told that the cigarettes contained THC (deceptive administration), and others, that they might or might not (double-blind administration). The deceived subjects smoked more placebo marijuana and reported a greater subjective response than the others, though the difference was confined to the first session. Together, these studies show that subjects' expectancies can have a significant impact on perceived THC effects. Moreover, they
Clearly demonstrate the need for inclusion of placebo marijuana (or hashish) in experimental research.

Another factor deemed important in determining the effects of cannabis is the setting, i.e. the prevailing environmental and social conditions during drug use. Hollister et al. (1975) investigated the effects of two settings, one 'favorable' and the other 'neutral', on the psychological and physiological effects during marijuana smoking. Twelve subjects were tested in both settings, once after smoking placebo and once after active marijuana. There was a clear effect of active drug along with substantial variability between subjects' reactions, but the actual setting in which the drug was administered did not significantly alter those effects. One's company during marijuana smoking seems a more powerful mitigating factor than the physical environment. As mentioned above, Jones found substantial differences in reactions to marijuana smoking between subjects tested individually and those tested in a group. The former exhibited the relaxed, slightly drowsy, and undramatic state typically seen in laboratory settings, whereas the latter showed elation, euphoria, uncontrolled laughter, and a marked lack of sedation. Further studies demonstrated that an individual's mood after marijuana smoking is significantly related to the prevailing moods of the group (Rossi et al., 1978); and, that THC produces greater perceived 'highs' when smoked in the presence of friends than strangers (Marks and Pow, 1989).

In conclusion, many factors may influence THC's effects but only a few seem to really matter. The physical setting and smoking technique appeared to only slightly influence THC's effects whereas the social setting and the user's expectations based on previous cannabis experience influenced them substantially. Potent psychosocial factors, more than differences in the relatively low to moderate THC doses that are typically consumed for recreational purposes seem to determine how THC affects behavior in real life situations. Or, as Jones concluded more than two decades ago, ". . many people have uncritically accepted the belief that the drug has specific effects on behavior and experience and that these can be readily identified . . . Although at high doses such a model may be valid, at the doses most youthful drug users are discussing there is ample evidence that the effects of psychoactive drugs on behavior and experience are often to a great extent independent of the drugs' pharmacological effects (p. 368)." Regarding the design of experimental research on cannabis, these observations strongly imply the need for placebo-controlled, double-blind, repeated-measures designs, in which each subject acts as his or her own control; and, to maintain the constancy of psychosocial factors, when the design is to measure THC's pharmacological effects, or systematically vary them for measuring the mitigating effects.

2.5.3    Chronic Effects

Disapproval of cannabis stems from a concern about the alleged consequences of its use, particularly over a prolonged period of time. It has been claimed that cannabis use leads to antisocial and criminal behavior, is a stepping stone to dependence on heroin and other more harmful drugs, and produces psychosis, `amotivational syndrome', and many physical disorders (Davison and Neale, 1978). The scientific findings regarding the asserted sequelae of long-term cannabis use are summarized in this section.

Whereas the acute effects of cannabis are relatively well established, consensus concerning the consequences of long-term use is lacking among researchers. This is particularly true regarding THC's supposed psychopathological effects. The major reason is that previous epidemiological and field studies have been unsuitable for determining the cause-effect relationship. Cross-sectional comparisons of users and nonusers fail to reveal whether drug use is the cause of psychopathology or whether both are the consequence of the same underlying problem. Although cannabis is most commonly taken for pleasure, the drug is also used as a facet of adolescent experimentation, to demonstrate independence or rebelliousness, cope with anxiety, or as 'self-medication' for early symptoms of mental illnesses such as depression or schizophrenia (Grinspoon and Bakalar, 1992). Prudence is therefore warranted when attributing a psychopathological or physical syndrome to chronic cannabis use when only an association between the two has been proven.

The cause-effect problem can be clearly illustrated by a most important longitudinal investigation of the relation between psychological characteristics and drug use (Shedler and Block, 1990). The investigators followed 101 San Francisco children from ages 3 to 18, and assessed many psychological measures at ages 3, 4, 5, 7, 11, 14, and 18. The parent-child interaction was measured at age 5. Information regarding their drug use was collected at age 18. By then 68% had used marijuana at least once and 39% were currently using it at least once a month, 21%, at least once a week. Large minorities had also used other drugs. Most of the sample could be divided into three groups: abstainers (29), experimenters (36), and frequent users (20). Sixteen did not fit into any category and were largely ignored in the authors' conclusions. The identifiable groups did not differ in socioeconomic status or IQ. The picture of frequent users that emerged was "one of a troubled adolescent, an adolescent who is interpersonally alienated, emotionally withdrawn, and manifestly unhappy, and who expresses his or her maladjustment through undercontrolled, overtly antisocial behavior." Abstainers were relatively "tense, overcontrolled, emotionally constricted . . . socially isolated and lacking in interpersonal skills," and experimenters appeared as the best-adjusted in the sample (The investigators hastened to add that these findings should not be misinterpreted as indicating that drug use might somehow improve an adolescent's psychological health). Solely on the basis of these observations one might conclude that frequent cannabis use leads to personal and social maladjustment. Yet the psychological differences between frequent drug users, experimenters, and abstainers could be traced to the earliest years of childhood and related to the quality of parent-child interactions. At seven and eleven years of age, future frequent users got along poorly with other children, were insecure, and showed numerous signs of emotional distress. Abstainers, at the same ages, were relatively overcontrolled, shy, fearful, and morose. At five years of age, mothers of both abstainers and frequent users were relatively cold and unresponsive. While pressuring their children to perform well, they gave them little encouragement. Fathers of frequent users were not different from those of experimenters, but fathers of abstainers were impatient, domineering, and squelched spontaneity and creativity. The investigators concluded from their observations that "(a) problem drug use is a symptom, not a cause, of personal and social maladjustment, and (b) the meaning of drug use can be understood only in the context of an individual's personality structure and developmental history." Of course, these findings should be replicated before final conclusions are drawn but they highlight a very important point that is too easily overlooked, i.e. simple associations between phenomena are never conclusive evidence for alleged cause-effect relationships.

Early reports emanating from law enforcement authorities and the story of the hashshashim (Section 2.1) led to another questionable association, i.e. that between cannabis use and violent and criminal behavior. It appears true that heavy cannabis use was more prevalent among criminals, before the 1960s when it was mainly confined to dissident minorities, but it does not necessarily follow that cannabis was the cause of their criminal acts. Though symptoms of bhanginduced toxic psychosis may include hostile feelings (Chaudry et al., 1991), cannabis failed to increase hostility in experimental studies involving healthy, occasional cannabis users (Salzman et al., 1976; Marks and Pow, 1989). Abel (1977) reviewed the relationship between cannabis use and violence, and concluded that cannabis does not precipitate violence in the vast majority of users though it may in unstable individuals. The consensus today is that, if anything, cannabis seems to inhibit violent and aggressive behavior (National Commission on Marihuana and Drug Abuse, 1972; Hollister, 1986; Grinspoon and Bakalar, 1992).

Another association that has been wrongly interpreted as showing a causal relationship is between the use of cannabis and other, more harmful, drugs such as heroin and cocaine. Some assume that the former is a stepping stone to the latter. While it is true that many heroin and cocaine addicts began their illicit drug 'career' with cannabis, the majority of cannabis users do not take these other drugs (Davison and Neale, 1978; Grinspoon and Bakalar, 1992). Similarly, most cannabis users first took licit drugs such as alcohol and nicotine by smoking tobacco (Ellickson et al., 1992), but the majority of alcohol and tobacco users do not smoke cannabis. Thus, if cannabis can be considered a stepping stone to more harmful drugs, alcohol and tobacco would fulfil that function for cannabis. One deduction of the stepping stone theory is that fewer people would be addicted to opiates if cannabis were not available. There is, however, little evidence for this hypothesis. Although cannabis users are more likely to become involved with other illicit drugs than nonusers, it seems very unlikely that THC itself causes the craving for other drugs. A more likely explanation is that users of any given drug generally show greater interest in experimenting with other drugs. Furthermore, the common subcultural scene in which both cannabis and other illicit drugs are available from the same source makes it easier for users of the former to become involved with the latter (Johnson, 1973; Cohen, 1975).

One of the major reasons for many people's disapproval of cannabis has been the possible development of an `amotivational syndrome'. This syndrome is characterized by a state of apathy and bluntness, along with a loss of interest in personal appearance and conventional goals. This personality disorder was manifest among frequent users as described by Shedler and Block (1990). As before, there is no reason to preclude a causal relationship. Neither is there a compelling reason to accept its existence. Many authorities have concluded that there is no convincing evidence that the amotivational syndrome is a direct consequence of marijuana use (e.g. Campbell, 1976; Hollister, 1986; Grinspoon and Bakalar, 1992).

Another alleged sequela of repeated cannabis use is the possible occurrence of 'flashbacks'. Flashbacks are spontaneous recurrences of a drug's effects, during a drug-free period, that are similar to those experienced earlier while under its influence. Their duration is usually much shorter, from seconds to minutes, but there are also reports of flashbacks that persisted for weeks or months and required psychiatric treatment (see review by Fischer and Tdschner, 1991). The etiology of flashbacks is still unknown but it is well established that they can occur in hallucinogen users. Most inquiries into the incidence of flashbacks among LSD users have determined that about 30% experience them at least once (Yager et al., 1983), but more extreme values in both directions have been reported: Fischer and Taschner (1991) surveyed the literature to show that incidence rates vary between 15 and 75%. This wide variation is probably due to the fact that various investigators apply different definitions of flashbacks and usually rely on the users' recall of their occurrence. The same and other problems arise when attempting to define the occurrence and frequency of flashbacks in cannabis users. The fraction of cannabis users who report flashbacks are mainly the greatest consumers. As mentioned above, their subjective reactions to cannabis contain a large placebo component based on expectations. Similarly, they might selectively attend to experiences of naturally occurring altered physiological and psychological states and attribute them to a spontaneous recurrence of previous intoxication (Heaton and Victor, 1976). Another complicating factor is that the majority of cannabis users who report flashbacks also have previous experience with much stronger hallucinogenic drugs, such as LSD and mescaline (Tunving, 1985). And, as a rule, flashbacks may well occur in these cases. There are some reports showing that flashbacks occur in individuals whose sole illicit drug use is cannabis, but, according to Fischer and Taschner, they are unusual enough to warrant a thorough diagnostic evaluation in every case.

As mentioned in the previous section, cannabis may trigger immediate organic psychosis and precipitate schizophreniform episodes. Though evidence for chronic organic reactions is sparse (Thornicroft, 1990), chronic cannabis use is associated with an increased risk of developing schizophrenia in the subsequent fifteen years (Andreasson et al., 1987). It has, however, also been suggested that cannabis use might be more prevalent in people with greater pre-existing social and psychological vulnerability, and that chronic users might have become mentally ill in any case (Johnson et al., 1988; Negrete, 1989). Mathers and Ghodse (1992) compared 61 newly admitted in-patients with psychotic symptoms and cannabis-positive urine analysis to 43 control patients with psychotic symptoms but with drug-free urine analysis. Interestingly, a far greater proportion of cannabis-positive patients had histories of other drug use than the controls. Patients were interviewed within one week of admission and again at one and six months, using the 140-item Present State Examination. At the first assessment, the two groups differed on only five items: changed perception, thought insertion, nonverbal auditory hallucinations, delusions of control, and of grandiose ability. Although the number of discriminating items were no more than expected on the basis of chance alone, they are consistent with acute cannabis intoxication. Only one item (delayed sleep) discriminated between the groups at one month, and none at six months. The authors concluded that short-lived psychotic episodes can occur in clear consciousness after cannabis intoxication, but the development of chronic cannabis-induced psychosis is unlikely. They as well as Thornicroft (1990) discouraged psychiatrists' use of the diagnostic label 'cannabis psychosis' because it may delay the correct alternative diagnosis, usually paranoid schizophrenia. Thus, it seems unlikely that chronic use of cannabis can produce a functional psychosis de novo in a stable individual. But it is quite likely that cannabis provokes, exacerbates, or prolongs pre-existing mental disturbances (Tunving, 1985).

As mentioned above, working memory impairment is one of the most consistently reported acute effects of THC. This raises the logical question whether chronic use of cannabis is associated with similar, more persistent, impairments. Early studies conducted in Jamaica, Greece, Costa Rica, and India failed to reveal any significant impairment of cognitive functions in chronic cannabis users (Dornbush and Kokkevi, 1976; Stefanis etal., 1976; Coggins, 1976; Venkoba Rao et al., 1975), but one Egyptian study did (Soueif, 1975). Most of these studies, however, have been criticized for many different reasons including small sample sizes, inadequate sampling techniques, lack of test standardization, and acute intoxication (Hollister, 1986; Solowij et al., 1991). Mendhiratta et al. (1978) compared 25 charas smokers and 25 bhang drinkers to 25 nonusers of cannabis, and found that users scored worse on backward digit span, pencil tapping, time and size estimation, recognition and reproduction of designs, and were slower in producing associations to words. After a lapse of ten years, they re-evaluated 60% of the subjects (i.e. 30 users and 15 nonusers) and found similar impairments, except in recognition and size and time estimation, which were not measured (Mendhiratta et al., 1988). Varma et al. (1988) compared 26 long-term heavy users with 26 controls who were matched in terms of age, education, and occupation. A comprehensive battery of tests was applied to measure cognitive, perceptual-motor, and personality variables. Performance impairments of users relative to nonusers were similar to those previously found by Mendhiratta et al. Varma et al. also measured ten different memory functions but found a significant difference between the groups in only one; i.e. users had poorer recent memory than nonusers. They concluded that "the differences, if any, between users and nonusers in terms of cognitive functions pertain to perceptuo-motor kasks." Users were less capable in personal, social, and vocational functioning, and showed higher psychoticism and neuroticism, but their relative disabilities failed to reach clinically significant levels. Varma et al. found no differences between the groups with respect to intelligence but Le6n-CarriOn (1990) reported that 23 daily users of cannabis had lower scores on nine of the fourteen subscales of the Wechsler Adult Intelligence Scale than 24 controls coming from the same socioeconomic and cultural strata. Finally, Solowij et al. (1991) compared auditory reaction times and evoked, event-related EEG potentials of nine chronic cannabis users with nine nonusers who were matched on age, sex, and years of education. Subjects participated in four conditions in which they were instructed to attend to a particular location (left or right) and pitch (1047 or 1319 Hz). Stimuli were either of short (51 ms) or long (102 ms) duration. Subjects were required to discriminate between them and respond to the latter. Users responded as rapidly as nonusers but with fewer correct and more false detections. The negative-going phase of early evoked potentials after short-duration stimuli was greatly enhanced in users relative to nonusers. The authors interpreted this as showing that the users engaged in superfluous pitch processing and were therefore less able to selectively attend to the duration of stimuli. These studies demonstrate that chronic use of cannabis may be associated with psychomotor and cognitive deficits, although, at the same time, deficits in chronic users seem to be less pervasive than those occurring during acute intoxication (Block et al., 1992). Yet none of the forementioned examined the possibility that users and nonusers differed on the evaluated variables, or other characteristics, prior to the former group's experience with cannabis. Prospective studies are still required for establishing the causal role of cannabis in producing these deficits.

It appears unlikely that gross neurological abnormalities develop as a result of chronic cannabis use. Cerebral blood flow is lower in chronic cannabis users, without significant regional blood flow abnormalities, and tends to return to normal with abstinence (Tunving et al., 1986). Chronic users' EEG patterns are similar to those from sedative addicts or normal drowsy individuals; i.e. all show increased alpha-activity and reduced beta-activity with a general slowing of activity across the EEG frequency band (Tunving, 1985). However, there are no signs of neurotoxicity, cerebral atrophy, or diminished cerebral function in chronic marihuana users (Farre-Albaladejo, 1989; Fehr and Kalant, 1983; Co et al., 1977).

The discovery of cannabinoid receptors in human leukocytes corroborates earlier observations that THC can impair immune responses by inhibiting T-lymphocyte proliferation and function, and suppressing 7-interferon production (Hollister, 1986; Bouaboula et al., 1993). While one is tempted to deduce from these findings that chronic cannabis users are more susceptible to opportunistic infections or malignancy, Hollister reported that the confirming clinical evidence is still lacking. Yet the potential immunosuppressive effect of THC is an important issue, even more so since the U.S. Food and Drug Administration (FDA) approved of dronabinol, a synthetic THC analog, for treatment of anorexia in AIDS patients (below). Drugs that are potentially immunosuppressive pose a grave risk to these patients. Dronabinol's net effect upon mortality in this population should therefore be carefully monitored.

Impaired pulmonary function is perhaps of greatest significance among the potential physical sequelae of long-term cannabis use. Heavy hashish use by soldiers has been associated with chronic bronchitis, chronic cough, and precancerous mucosal changes; i.e. squamous metaplasia (Henderson et al., 1972). Marijuana smoking may produce mild but significant airway obstruction (Tashkin et al., 1976). Compared with tobacco smoking, it delivers greater amounts of the carcinogenic products, tar and benzopyrene to the mouth and leads to greater retention of the inhaled tar in the lung (Wu et al., 1988; Tashkin etal., 1991). Furthermore, it inhibits alveolar macrophages that contribute to the pulmonary antibacterial defense system (Sherman et al., 1991). On the other hand, heavy use of cannabis is generally defined as smoking at least once per day while, for tobacco smokers, it usually means at least 20 times per day. The total daily inhaled amount of tar is therefore probably much less in heavy smokers of cannabis than of tobacco. Nevertheless, chronic cannabis smoking has the potential to produce bronchitis, emphysema, or pulmonary carcinoma. Furthermore, the cardiovascular and respiratory depressant effects of a single dose are potentially dangerous for patients with pre-existing cardiovascular or pulmonary disease.
Changes in reproductive functions have been a source of controversy ever since Kolodny et al.'s (1974) widely publicized report. It indicated that chronic male marijuana users had depressed testosterone levels and also that follicle stimulating hormone levels were lower in heavy than light users. Further investigations showed depressed levels of other gonadotropins in chronic THC users, specifically luteinizing hormone and prolactin. But, Block et al. (1991) noted that they were outnumbered by studies showing no significant difference in any hormone's concentration between groups of users and nonusers. They went on to measure testosterone, luteinizing hormone, follicle stimulating hormone, and prolactin concentrations in 27 frequent, 18 moderate, and 30 infrequent marijuana users, and compared them with the values obtained from 74 nonusers. Also measured were cortisol concentrations to control for possible reactions to the stress of venipuncture. No significant intergroup differences for any of the five hormones were found in either men or women.

While cannabis' effects on hormone levels seem to be minor, those on other reproductive functions appear to be of greater concern. Adverse effects on males' sperm have been repeatedly reported. These include decreased sperm count and motility, and increased morphological abnormalities (Kolodny et al., 1974; Hembree et al., 1979). These early findings are corroborated by the recent discovery of cannabinoid receptors, though sparse, in human testis (Gerard etal., 1991). In addition, THC crosses the placenta (Idanpaan-Heikkila et al., 1969) and may retard fetal growth (Zuckerman et al., 1989; Day et al., 1992). Newborns prenatally exposed to cannabis had, compared to controls, decreased mean arm circumference and nonfat volume of the arm, but normal fat stores (Frank et al., 1990). The fetal growth retardation followed a symmetric pattern and no abnormalities were found in weight/length or arm circumference/head circumference ratios, which would have indicated a malnutritional cause. Because THC decreases fetal oxygenation (e.g. Clapp et al., 1987), Frank et al. were led to hypothesize that maternal-fetal hypoxia produced the retardation of fetal growth. One study reported that offspring of women who chronically used marijuana during pregnancy were more likely to have abnormalities comparable to the fetal alcohol syndrome than offspring of nonusers (Hingson et al., 1982), but another showed that infant morphological abnormalities could not be explained by the mothers' marijuana use (Day et al., 1992). Exposure to cannabinoids via the mother's milk during the first month postpartum was associated with decreased motor development at one year of age (Astley and Little, 1990).

Canadian investigators recorded the pregnancies and deliveries of approximately 700 women residing in the Ottawa region over a six year period. They also failed to find differences between users and nonusers with regard to pregnancy outcome measures, including miscarriage rates, type of presentation at birth, Apgar status, and frequency of complications or major physical anomalies of the newborn infants (Fried et al., 1983, 1984). Behavioral evaluations revealed that prenatally exposed newborns had increased tremors and startles, and poorer habituation to visual stimuli (Fried and Makin, 1987). At one, two, and three years of age, no adverse effects of prenatal THC exposure were found (Fried, 1989a). One of Fried's (1989b) alternative explanations was that "the drug's effects are transitory and . . . the nervous system has 'caught up' sufficiently in the marijuana offspring to be undifferentiated at a behavioral level from control subjects." Yet at four years of age, children of mothers who smoked marijuana during pregnancy yielded significantly lower scores in verbal and memory domains than the controls (Fried and Watkinson, 1990). At six years of age, they committed more omission errors in a vigilance task and had higher ratings by the mothers on an impulsive/hyperactive scale (Fried etal., 1992). It may be, as Fried (19896) stated earlier: "the long-term effects of maternal marijuana use are very subtle and the facets of behavior that are affected manifest themselves under more complex situations than can be examined in a very young child." Opposed to these results are those of a Jamaican study which failed to reveal any adverse effects of prenatal marijuana exposure on 4- and 5-year old children's abilities (Hayes et al., 1991).

Most studies controlled at least for some potential confounding factors, such as socioeconomic status, other drug use, and home environment. There are, however, other variables that are more difficult to control but may influence newborn and later development, including diet during pregnancy and the postnatal parent-child relationship (Fried, 1989a; Zuckerman and Bresnahan, 1991). These investigators suggested that poor diet during pregnancy and deficient child caring may potentiate the adverse effects of prenatal drug exposure. Conversely, children might be protected from the drug's effects in utero if their mothers' diet during pregnancy is optimal; and, might recover from adverse effects, if any, when an intervention program addressing the child's developmental needs is provided in combination with a resolution of the mother's, and, if pertinent, also the father's, drug dependence.

In conclusion, chronic use of cannabis has been associated with many adverse effects, including the psychological, physical, and teratological. While the acute effects of cannabis were relatively simple to establish, it has proven very difficult to find evidence for the alleged long-term consequences of cannabis use, particularly the psychopathological. This is not surprising. Long-term effects generally appear only after many years of sustained drug use, and, in the meantime, many other factors confound the observed association between cannabis use and adverse effect. Even when a true association is found, it's still open to question whether cannabis use was the cause or simply an effect. Only prospective studies including very large samples seem capable to demonstrate long-term effects of drugs in general, and cannabis in particular. Excellent examples of these type of studies were provided by Shedler and Block (1990) and the Ottawa group. On the basis of the available evidence, it appears that cannabis does not produce antisocial or criminal behavior, is not a stepping stone to dependence on more harmful drugs, and produces neither psychosis nor amotivational syndrome in normal, stable individuals. It seems that heavy use of cannabis can be just one among the many symptoms of maladjustment or psychopathology. At most, cannabis may exacerbate an individual's vulnerability and provoke pre-existing or aggravate existing psychiatric illness. On the other hand, chronic cannabis use has been associated with some cognitive and psychomotor deficits. While it remains to be determined whether these effects are solely attributable to chronic drug use, it would be premature to draw definite conclusions regarding the psychological sequelae of long-term cannabis use. Perhaps of greater importance are the physical consequences. Cannabis has the potential to suppress the immune system, produce respiratory disease, and increase the risk of lung cancer. Furthermore, it may impair reproductive functions, retard fetal growth, and possibly the postnatal development of children born (and reared) by cannabis-dependent mothers.

2.5.4 Tolerance and Dependence

An important question with relation to cannabis, as to all other psychoactive drugs, is whether long-term use produces tolerance and/or dependence. This has important implications for both heavy recreational users and patients who use THC therapeutically (below). Tolerance to some drug effects may be desired by one group but not the other. The possible experience of withdrawal symptoms after abrupt drug discontinuation is disliked by all users.

Tolerance means that a drug's effects diminish with repeated administration of similar doses; and, that greater amounts must be administered to produce the effects initially experienced. Most psychoactive drugs including THC have multiple effects. Tolerance generally does not develop to all of them at the same rate so it should be determined for each effect separately. Dependence can be defined as (1) preoccupation with drug acquisition, (2) compulsive use, and (3) relapse to or recurrent use (e.g. Miller and Gold, 1989). Physical or physiological dependence on a drug is said to have developed when a characteristic abstinence syndrome is observed when use of that drug is discontinued; and, when the withdrawal symptoms can be reversed by readministration of the drug. Physical dependence has previously been referred to by the term addiction. The concept of psychological dependence, previously referred to by the term drug habituation, was proposed to describe drug dependence without the occurrence of withdrawal reactions when drug use is terminated. Individuals with a psychological dependence on a drug may, if the drug is withdrawn, exhibit reactions that resemble those of a withdrawal syndrome, e.g. restlessness and irritability, but should not be regarded as such unless tolerance has also developed (Davison and Neale, 1978). Although tolerance and dependence are closely related, the former does not necessarily produce the latter.

Diminished drug-induced tachycardia and psychological 'high' have been observed by several investigators after repeated marijuana smoking within a relatively short period, i.e. 2 to 18 hours (Cochetto et al., 1981; Heishman et aL, 1990; Chait, 1990). These studies were, however, not specifically designed to examine the possible development of tolerance. Those that were suggested tolerance without conclusive evidence because ethical considerations limited the subjects' consumption of THC in repeated doses (Hollister, 1986). Jones and his colleagues conducted a series of outstanding studies to examine tolerance and withdrawal symptoms following prolonged oral administration of THC. Jones et aL (1981) summarized their results by concluding, "both tolerance and physical dependence develop after surprisingly short periods of cannabis or THC administration when the conditions are optimal", i.e. when dose, dosing frequency, duration, and route of administration are such that sustained blood/brain levels of THC are achieved. Their volunteers, all experienced marijuana users, lived on a hospital research ward for two to six weeks. They usually received placebo treatment for three to seven days after admission, followed by a 5— to 21—day period of active treatment with 10 to 30 mg oral doses of cannabis extract or THC, given every three or four hours, 24 hours a day. Then, under double-blind conditions, treatment was abruptly switched to placebo for four to eight days. Tolerance to THC's subjective effects developed rapidly. For example, a 50% reduction in intoxication ratings was observed after four days exposure to 10 mg doses. More than 80% reduction was achieved in the late stage of drug administration. Initial physiological signs of intoxication such as tachycardia and decreased skin temperature also gradually disappeared, with bradycardia developing at the end of the treatment period. Tolerance to orthostatic hypotension developed within 12 to 24 hours. Initial decreases in intraocular pressure, salivary flow, and REM sleep partially or completely recovered within ten days of sustained drug administration, and rebounded above pre-drug levels after abrupt cessation. Tolerance did not develop to the drug's appetite-enhancing effect and body weight increased progressively. Tolerance was further examined while subjects received constant oral doses of 20 mg every three hours or 30 mg every four hours and were challenged daily with superimposed doses of 20 mg via marijuana smoking or 2 mg/70 kg via injection. Smoked or injected THC doses were also administered before the oral treatment period to measure their acute effects, and, after withdrawal of oral treatment, to determine how long tolerance• persisted. Subjective, heart rate, skin temperature, and intraocular pressure reactions to the challenge diminished after only a few days of oral THC administration, but initial reactions rapidly returned when placebo was substituted for drug in oral dosing. Another interesting finding was that tolerance developed more rapidly while the subjects were constantly treated with the lower, more frequent oral doses than the opposite. According to Jones et al., this suggests that the rate of tolerance acquisition is mainly a function of the time subjects are in the rapid a-phase of TI IC elimination from plasma rather than the 13-phase with the slower elimination.

Tolerance to a drug's effect generally implies that progressively larger amounts of the drug must be taken to reproduce its initial effects. Yet Cappell and Pliner (1974) found no difference between frequent and infrequent users' sel f-administration of marijuana cigarettes to achieve a 'nice high'. This indicates that the regularity of marijuana smoking by those who described their use as 'frequent' is still not sufficient for acquiring substantial tolerance. If marked tolerance to THC's subjective as well as cognitive and psychomotor effects were common, there would be implications for experimental research: subjects whose smoking regularity differed widely could not be considered as members of the same population and conclusions would necessarily be confined to that subpopulation providing the results. However, the general lack of tolerance exhibited by recreational THC users and the usual periods of abstinence required by most experimental protocols prior to testing encourage the belief that acute effects are not confounded by that factor.

The nature of tolerance may be either pharmacodynamic or pharmacokinetic. The former, also called 'functional' tolerance, is due to localized changes at the drug's site of action. The latter, also called `dispositional' tolerance, is a result of either increased drug metabolism or other factors that reduce its plasma concentration after repeated dosing. Hunt and Jones (1980) administered "C-labelled THC by i.v. injection in six subjects both before and at the end of the oral THC administration period, and found only a slight increase in drug metabolism, too little to account for the observed degree of tolerance. Furthermore, there seems to be no difference in the rate of clearance of THC between chronic and naive users (Agurell et al., 1986). The available evidence therefore indicates that tolerance to THC is functional rather than dispositional in nature.

In Jones' studies, withdrawal signs and symptoms began to appear five to six hours after the last oral THC administration. Physiological signs such as salivary flow, intraocular pressure, REM sleep, finger tremor, and sweating were increased, body weight decreased.

Subjective symptoms included irritability, restlessness, insomnia, anorexia, and nausea. Both physiological and subjective withdrawal phenomena were alleviated for two to three hours by smoking a marijuana cigarette containing 20 mg THC. Generally, symptoms were most intense at about eight to twelve hours after THC cessation, diminished thereafter, and were not measurable three to four days later. The exception was disturbed sleep: some subjects reported insomnia for as long as several weeks. Other studies, reviewed by Hollister (1986) and Compton et al. (1990) , wherein THC was inhaled or where smaller doses of THC were administered orally, found fewer and milder symptoms of a similar nature during withdrawal. There are apparently no reports of deaths during withdrawal but many describe psychological dependence on the drug. Therefore, Compton et al. concluded that psychological dependence is more probable and of greater importance than physical dependence. This opinion was endorsed by Shedler and Block (1990). Their findings suggested that dependence upon cannabis is a symptom rather than a cause of social and personal maladjustment. Grinspoon and Bakalar (1992) articulated this view as follows: "becoming attached to cannabis is not so much a function of any inherent psychopharmacologic property of the drug as it is emotionally driven by the underlying psychopathology (p. 243)."

Both tolerance and withdrawal depend upon the THC dose taken at one time, and the frequency and duration of dosing. These phenomena probably occur in a coupled series; substantial tolerance followed by strong withdrawal symptoms. But neither is likely to become practically significant until large THC doses are consumed more than once daily over a prolonged period of time. According to Hollister, most users do not follow such a dosing regimen. So neither tolerance nor physical dependence has been a major issue in social cannabis use. This may change with the recent rise in cannabis potency. As cannabis smokers seem not to adjust their smoking pattern to the drug's potency (see Section 2.5.2), instances of physical dependence may become more prevalent in the near future.

2.5.5    Therapeutic Applications

The most promising therapeutic use of THC is for treatment of nausea and vomiting associated with cancer chemotherapy (Chang et al., 1979; Vinciguerra et al., 1988; Randall, 1990). Two synthetic cannabinoids, nabilone and levonantradol, are also effective antiemetics (Vincent et al., 1983). Dronabinol, formulated in sesame oil and encapsulated in round soft gelatin capsules for oral administration, was approved in 1985 for antiemetic indications by the FDA. The drug is now marketed under the name Marina (Unimed Inc., Somerville, NJ).

THC has also been successfully applied in some cases of anorexia nervosa (Zinberg, 1979), and clinical trials have demonstrated that patients suffering from AIDS may also benefit from the drug's appetite stimulating effects (Plasse et al., 1991). In 1993, the FDA approved of a supplemental drug application for dronabinol for use in AIDS patients who suffer anorexia and weight loss.

THC decreases intraocular pressure by decreasing fluid formation and increasing fluid outflow. It has been successfully applied in the treatment of glaucoma or ocular hypertension (Merrit et al., 1980; Grinspoon and Bakalar, 1993).

A variety of THC indications have been suggested besides those already accepted by drug regulatory authorities. Although cannabinoids do not bind at opiate receptors, levonantradol and its active metabolite desacetyllevonantradol
were found to be potent analgetics in rodents (Howlett et al., 1990). THC and CBD posses anticonvulsant properties which may be of therapeutic value (Reiman, 1982; Hollister, 1986). They exacerbate hypokinesia and resting tremor in Parkinsonian patients but are beneficial for some forms of dystonia, tremor, and spasticity (Petro and Ellenberger, 1981; Consroe etal., 1986; Meinck et a1., 1989). CBD attenuates the anxiogenic effect of THC and possesses anxiolytic properties in individuals exposed to a stressful situation (Zuardi et al., 1982, 1993). Other alleged indications for cannabinoids include bronchitis, asthma, insomnia, hypertension, abstinence syndromes, migraine, and alcoholism. But approval of cannabinoids for any of these indications is, at present, most unlikely (Hollister, 1986).

2.6 Summary and Conclusions

Man's interest in Cannabis sativa extends backward for more than ten thousand years. Cannabis has been used as an ingredient in clothes and rope, and as a medicine for a variety of discomforts and diseases. The principal interest in cannabis, however, has been in its euphoric and mind-altering properties. In Western societies, recreational use of cannabis for that reason began in the Middle Ages, but it was not until the 1960s that cannabis' popularity became widespread, not only in lower social classes and minority groups but throughout all levels in society. As a consequence, scientific research on cannabis increased dramatically ever since.

Marijuana and hashish are the most widely used cannabis preparations in Western societies. The former is made from the dried leaves and flowering tops of the plant, which are cut, dried, chopped, and incorporated into cigarettes. The latter is prepared from the resinous exudate itself and is often mixed with tobacco before smoking. The drug can also be chewed, prepared as tea, or eaten in baked goods.

Cannabis sativa contains more than 400 chemical compounds. More 60, the cannabinoids, are specific to the plant and are present in the sticky golden resin that covers the flowers and top leaves. The cannabinoid primarily responsible for the drug's physiological and psychological effects is Y-tetrahydrocannabinol (THc). When entering the systemic circulation, THC is rapidly distributed into highly perfused fatty tissues, including the brain. Next, it slowly diffuses from tissue into blood, is metabolized and excreted in urine and feces. The distribution (a) phase has a half-life of 30 minutes, the elimination (13) phase, several days. As a result, accumulation of THC occurs with repeated drug administration, though only in neutral fat and liver and not in blood and brain. Due to their slow elimination, THC's metabolites can be present in urine for weeks after the last drug intake. Recently, cannabinoid receptors have been identified in human brain, testis, spleen, tonsils and leukocytes. Receptor concentrations are most dense in certain brain areas, including the cerebral cortex, hippocampus, cerebellum, and projection nuclei of the basal ganglia. Furthermore, a putative endogenous ligand has been discovered. These findings, and others to come, will certainly contribute to a better understanding of the cannabinoids' mechanism of action.
The acute subjective effects of cannabis smoking appear within 15 minutes and generally last for about two to four hours. When cannabis is ingested, THC's onset of action is some hours delayed and its effects generally last for 5 to 12 hours. The most reliable physiological signs of cannabis intoxication include conjunctivitis and tachycardia. The principal psychological effect is an euphoric state which is consistently accompanied by distorted time perception and disruption of memory processes. The user may experience temporal disintegration, manifested as difficulty concentrating and intrusion of thoughts, and depersonalization. Dysphoric reactions may occur in inexperienced users or anyone after consuming very high doses of THC and include anxiety, sometimes accompanied by paranoid thoughts, and panic reactions. Acute depressive episodes and toxic psychosis have been occasionally reported, mainly in users with some underlying depression or a history of psychosis.

The timing and magnitude of the acute effects are dependent on the THC dose and route of administration. Other factors that may influence the acute effects of THC, particularly when consuming low or moderate doses, include the social setting and the user's expectations based on previous experience. The smoking technique may affect physiological signs of intoxication, but has little effect on subjective feelings or performance.

It has proven very difficult to find convincing evidence for the alleged longterm sequelae of cannabis use. Heavy chronic users of cannabis have been shown to be socially and personally maladjusted, but a longitudinal study indicated that problem drug use is more likely a symptom rather than the cause of maladjustment. While cannabis use may provoke pre-existing or aggravate existing psychiatric illness, there is no convincing evidence that it produces functional psychosis de novo in a stable individual. Neither is there much reason to believe that cannabis use per se leads to criminal behavior, dependence on other, more harmful drugs, or amotivational syndrome. Yet chronic cannabis use was associated with some cognitive and psychomotor deficits. It remains to be determined whether these effects are solely attributable to cannabis but they are probably the most obvious psychological sequelae of chronic cannabis use. The physical sequelae seem to be more pervasive. Chronic use of cannabis suppresses the immune system and may produce respiratory disease and lung cancer. Furthermore, it reduces sperm quality and may retard both fetal growth and postnatal development.

Both tolerance and dependence followed by withdrawal symptoms are a function of the dosing regimen. Neither is likely to reach clinically significant levels until large THC doses are consumed more than once daily over a prolonged period of time. Tolerance appears more functional than dispositional and is rapidly reversible with abstinence. Withdrawal symptomatology is mild and transient, and is characterized by irritability, restlessness, insomnia, anorexia, nausea, salivation, sweating, tremor, increased intraocular pressure and increased REM sleep.

The most promising therapeutic applications of THC include the treatment of nausea and vomiting in cancer patients undergoing chemotherapy, and anorexia associated with weight loss in AIDS patients. Cannabinoids have also been shown to possess analgesic and anticonvulsant properties, and may be successfully applied in the treatment of glaucoma and some forms of dystonia, tremor, and spasticity.

In conclusion, cannabis is a notorious drug but also a fascinating pharmacological entity. Throughout history many people have shown interest in the drug for many different reasons. Lately that interest has been shared by scientists resulting in a huge number of publications. Yet there is still a surprising paucity of sound information regarding the long-term effects of cannabis. Consequently, many people's viewpoints are still largely based upon personal experience or prejudice, or both, which has resulted in extreme polarization.
Similar opposing opinions prevail regarding cannabis' effects on driving performance and traffic safety. Some apparently believe that cannabis users are among the most hazardous drivers on the road whereas others claim that cannabis produces more cautious driving and is therefore no hazard at all. The debate might continue indefinitely were one unable to objectively measure TI-IC's effects on its users' actual driving performance. Fortunately, it was possible for us to do so in a series of acute dosing studies conducted in environments that gradually approached full ecological validity. Before describing them, however, it seems wise to consider what previous research has shown concerning the influence of cannabis smoking on driving. This is done in the following chapter with a literature review focused upon that specific issue.

< Prev		Next >