Back to top

Report to the Director, National Institutes of Health

Daily Dose 2010-11-10 0 comments

By the Ad Hoc Group of Experts – Excerpt

Over the past 18 months there has been wide-ranging public discussion on the potential medical uses of marijuana, particularly smoked marijuana. To contribute to the resolution of the debate, the National Institutes of Health (NIH) held a 2-day scientific meeting on February 19-20, 1997, to review the scientific data concerning the potential therapeutic uses for marijuana and the need for and feasibility of additional research.

Central to the current debate about the therapeutic uses of marijuana is the claim that smoked marijuana offers therapeutic advantages over the currently available oral form (dronabinol capsules) of its most active ingredient, delta-9-tetrahydrocannabinol (9-THC), for a wide variety of conditions. As the therapeutic claims surrounding marijuana are wide-ranging, 10 separate NIH Institutes (with interest in the relevant areas) selected a group of eight experts with broad experience in clinical studies and therapeutics (and none of whom had a predetermined position on the medical utility of marijuana) to examine the data from the published scientific literature presented by speakers in the various therapeutic fields. The Ad Hoc Group of Experts also considered public comments including those of patients and advocacy groups as well as written material submitted to the Group after the meeting. The Expert Group was asked to focus on four questions:

Question 1 – What research has been done previously and what is currently known about the possible medical uses of marijuana?

Question 2 – What are the major unanswered scientific questions?

Question 3 – What are the diseases or conditions for which marijuana might have potential as a treatment and that merit further study?

Question 4 – What special issues have to be considered in conducting clinical trials of the therapeutic uses of marijuana?

Each presentation of data by a speaker was followed by a question-and-answer session by the Expert Group. There was no requirement that individuals on the Group agree or express a consensus view, although they were free to do so if they so desired. A second day was provided for public comment and further discussion by the Expert Group.

This report is a compilation of the opinions of the Expert Group. Speakers reviewed the literature on the potential efficacy of cannabinoids, including smoked marijuana, in the areas of analgesia, neurological and movement disorders, nausea and vomiting associated with cancer chemotherapy, glaucoma, and appetite stimulation/cachexia. A review of selected aspects of the general clinical pharmacology of marijuana precedes the disorder-specific commentary.

The discovery of receptors in the central nervous system (CNS) for cannabinoid compounds, and the presence of an endogenous ligand for these receptors, is of importance to the debate concerning the potential therapeutic uses of marijuana. This discovery supports a recommendation for more basic research to discover the functional roles of the cannabinoid receptors as a key underpinning for possible therapeutic applications. Such an approach allows the bridging of knowledge from molecular neurobiology to animal studies to human clinical trials.

The scientific process should be allowed to evaluate the potential therapeutic effects of marijuana for certain disorders, dissociated from the societal debate over the potential harmful effects of nonmedical marijuana use. All decisions on the ultimate usefulness of a medical intervention are based on a benefit/risk calculation, and marijuana should be no exception to this generally accepted principle.

The availability of THC in capsule form does not fully satisfy the need to evaluate the potential medical utility of marijuana. The Expert Group noted that, although delta-9-tetrahydrocannabinol (THC, dronabinol, Marinol¬ģ, or¬†9-THC) is the principal psychoactive component of the cannabis leaf, there may be other compounds in the leaf that have useful therapeutic properties. Furthermore, the bioavailability and pharmacokinetics of THC from smoked marijuana are substantially different than those of the oral dosage form. These are the rationales for studying the pharmacological actions of other constituents of the cannabis leaf, as well as determining whether a differential benefit occurs with smoked marijuana rather than oral dronabinol.

The Expert Group noted that even for conditions where good therapies are available, some patients develop adverse reactions or are nonresponders. The needs of this subset of nonresponders must be considered in the deliberations on testing marijuana as a possible therapeutic agent.

The Expert Group also noted that risks associated with marijuana, especially smoked marijuana, must be considered not only in terms of immediate adverse effects on the lung; e.g., bronchi and alveoli, but also long-term effects in patients with chronic diseases. Additionally, age, immune status, the development of intercurrent illnesses, and concomitant diseases should be considered in the determination of the risk calculation. The possibility that frequent and prolonged marijuana use might lead to clinically significant impairments of immune system function is great enough that relevant studies should be part of any marijuana medication development research, particularly when marijuana will be used by patients with compromised immune systems. Concerns were expressed by members of the Expert Group on the use of smoked marijuana because of the combustion byproducts, particularly when marijuana would be used for conditions requiring chronic therapy. Hence, a recommendation was made for the development of insufflation/inhalation devices or dosage forms capable of delivering purer THC or cannabinoids to the lungs free of dangerous combustion byproducts.

The major conclusions in each therapeutic area are summarized below.


No clinical trials involving smoked marijuana have been performed in patients with naturally occurring pain. Two adequate and well-controlled studies in cancer pain compared graded doses of oral 9-THC to placebo, and one of these included graded doses of codeine as a control. Although there was evidence of analgesic efficacy, the studies indicate there is a narrow therapeutic margin between the doses that produce useful analgesia and those producing unacceptable adverse CNS effects.

Neurological and Movement Disorders

Numerous preclinical and clinical studies of the use of cannabinoids in neurological and movement disorders have been reported as accounts of animal experiments, clinical anecdotes, surveys, and clinical studies.

Evidence that marijuana relieves spasticity produced by multiple sclerosis (MS) and partial spinal cord injury is largely anecdotal. Large-scale trials or controlled studies to compare marijuana or THC with currently available therapies have not been performed. There is no published evidence that cannabinoids are superior or equivalent to available therapies.

Preclinical evidence suggests a possible role for cannabinoids in the treatment of the epilepsies, particularly generalized and partial tonic-clonic seizures. There is scant information on the use of marijuana or other cannabinoids for the actual treatment of epilepsy.

Individual case studies have reported some benefit of smoked marijuana in treatment of dystonic states. Smoked marijuana or oral THC administrations for Parkinson’s disease or Huntington’s chorea have not been effective.

Cannabinoids have shown efficacy as immune modulators in animal models of neurological conditions such as experimental allergic encephalomyelitis (EAE) and neuritis. These data suggest that cannabinoids might modify the presumed autoimmune cause of a disease such as MS. However, long-term risks of smoked marijuana need to be quantified when considering chronic therapy for neurological conditions.

Nausea and Vomiting Associated With Cancer Chemotherapy

There is a large body of literature on the effects of cannabinoids on chemotherapy-induced nausea and vomiting. Most of the clinical trials used oral dronabinol rather than smoked marijuana. The oral THC studies showed this dosage form to be superior to placebo and generally equivalent or superior to prochlorperazine, but inferior to metoclopramide. Only one study compared smoked marijuana and dronabinol in a crossover design. Of the 20 patients studied, 9 had no preference, 7 preferred dronabinol, and 4 preferred smoked marijuana.

Since the approval of dronabinol in the mid 1980s for the relief of nausea and vomiting associated with cancer chemotherapy, more effective antiemetics have been developed, such as ondansetron, granisetron, and dolasetron, each combined with dexamethasone. The relative efficacy of cannabinoids versus these newer antiemetics has not been evaluated. Smoked marijuana was tested in one trial in patients who previously had no benefit from older antiemetic agents. Nearly one-quarter of patients who initially agreed to participate later declined citing bias against smoking, the harshness of smoke, and preference for dronabinol. Among the remaining 56 patients, 78 percent rated smoked marijuana very effective or moderately effective. Sedation was seen in 88 percent and dry mouth in 77 percent. It is not known whether smoked marijuana would benefit patients refractory to the current generation of antiemetic therapy.


Smoked marijuana has been shown to lower intraocular pressure (IOP) in subjects with normal IOP and patients with glaucoma. The duration of the pressure-lowering effect is 3 to 4 hours. Single-administration studies have reported blood pressure falls concurrently with the IOP lowering, raising concern that blood flow to the optic nerve could be compromised. Mitigating this concern are data suggesting that tolerance may develop to cardiovascular effects. Efforts to avoid or reduce side effects led to the development of a topical dosage form of THC. Topically applied THC did not lower IOP.

The mechanism of all IOP-lowering drugs currently used to treat glaucoma is known with the exception of marijuana. The interactive effect of marijuana with currently available IOP-lowering agents is not known but is evaluable. Elucidation of the mechanism of action of marijuana’s IOP-lowering effect is crucial to its potential utilization for treatment of glaucoma; a unique mechanism of action might provide additive benefit whereas a mechanism identical to an available medication would suggest an unfavorable benefit/risk ratio.

Appetite Stimulation/Cachexia

Clinical studies and survey data in healthy populations have shown a strong relationship between marijuana use and increased eating. Marijuana is reported to increase food enjoyment and the number of times individuals eat per day. Mechanistic studies of marijuana on taste and satiety have shown that it does not affect taste or produce a collapse of normal satiety mechanisms. Food intake associated with marijuana use is influenced by the social setting.

There are no controlled studies of marijuana in the AIDS-wasting syndrome, nor have there been any systematic studies of the effects of smoked marijuana on immunological status in HIV-infected patients. Smoking (tobacco, marijuana, or crack cocaine) has been shown to increase the risk of developing bacterial pneumonia in HIV-positive immune-compromised patients. Dronabinol has been shown to increase appetite and produce weight gain in AIDS and cancer patients, although the weight gain is not in lean body mass. Dronabinol is approved for the treatment of anorexia in patients with AIDS-associated weight loss.

Question 3: Which Diseases or Conditions Merit Further Study?

Concerning Question 3, there were varying degrees of enthusiasm to pursue smoked marijuana for several indications. This enthusiasm was tempered by the fact that, for many of these disorders, effective alternative treatments are already available. Given the general consensus among the experts that the number, design and documentation of studies performed to date with smoked marijuana did not provide definitive answers, it was difficult to compare marijuana with products that had received regulatory approval under more rigorous experimental conditions. This does not mean, however, that the issue should be foreclosed. It simply means that in order to evaluate various hypotheses concerning the potential utility of marijuana in various therapeutic areas, more and better studies would be needed. In the words of Dr. William Beaver, Professor of Pharmacology and Anesthesia, Georgetown University School of Medicine, who chaired the workshop, “For at least some potential indications, marijuana looks promising enough to recommend that there be new controlled studies done.” The indications in which varying levels of interest was expressed are the following:

  • Appetite stimulation/cachexia(1)
  • Nausea and vomiting following anticancer therapy(2)
  • Neurological and movement disorders
  • Analgesia
  • Glaucoma

Accordingly, the NIH should consider relevant administrative mechanisms to facilitate grant applications in each of these areas. Whether or not the NIH is the primary source of grant support for a proposed bona fide clinical research study, if that study meets U.S. regulatory standards (U.S. Food and Drug Administration (FDA) protocol approval and Drug Enforcement Administration (DEA) controlled substances registration) the study should receive marijuana and/or matching placebo supplied by the National Institute on Drug Abuse (NIDA). In this way, a new body of studies may emerge to test the various hypotheses concerning marijuana.

The last question, Question 4, concerning the special issues involved in conducting clinical trials with marijuana, was particularly difficult. There was considerable discussion and debate as to whether smoked marijuana (with the inherent health risks of smoking) would need to demonstrate clear superiority or some unique benefit compared with other medications currently available for these conditions. The Expert Group concluded that smoked marijuana should be held to standards equivalent to other medications for efficacy and safety considerations. Moreover, there might be some patient populations; e.g., cancer patients experiencing nausea and vomiting during chemotherapy, for whom the inhalation route might offer advantages over the currently available capsule formulation. This raises many issues concerning the best mode of administration. Generally accepted pharmacotherapy development schema would favor finding routes of administration under which dosing could be more tightly controlled and easily titrated. Smoking plant material poses difficulties in standardizing testing paradigms, and components of the smoke are hazardous, especially in the immunocompromised patient. Additionally, practical problems exist. Given the no-smoking policy of hospitals and public facilities, it would be difficult to imagine the utility of smoked marijuana in these settings. Therefore, the experts generally favored the development of alternative dosage forms, including an inhaler dosage form into which a controlled unit dose of THC could be placed and volatilized. Other problems noted were the difficulty in attempting to match placebo control against smoked marijuana (especially for those with previous marijuana experience), and the fact that under U.S. law, researchers will need to obtain DEA registration to handle marijuana, which is currently a Schedule I controlled substance (see Appendix).

In summary, the testing of smoked marijuana to evaluate its therapeutic effects is a difficult, but not impossible, task. Until studies are done using scientifically acceptable clinical trial design and subjected to appropriate statistical analysis, the questions concerning the therapeutic utility of marijuana will likely remain much as they have to date–largely unanswered. To the extent that the NIH can facilitate the development of a scientifically rigorous and relevant database, the NIH should do so.

On February 19 and 20, 1997, the National Institutes of Health (NIH) held a meeting concerning the potential medical uses of marijuana. Recent (November 1996) ballot initiatives in California and Arizona had sparked a public health and policy debate on the medical utility of marijuana and the desirability of allowing healthcare providers to prescribe, and patients to receive, marijuana for medicinal purposes.

For some years the principal psychoactive ingredient of marijuana, delta-9-tetrahydrocannabinol (9-THC), has been available to healthcare providers in an oral form as dronabinol (trade name Marinol) for the treatment of emesis associated with cancer chemotherapy and for appetite stimulation in the treatment of AIDS wasting syndrome. The current debate centers primarily on the potential for other treatment indications and the claims that, when smoked, marijuana offers therapeutic advantages over the currently available oral form. As the Federal Government’s principal biomedical research agency, the NIH believed that the public debate could benefit from an impartial examination of all the data available to date concerning these issues. As the claims for benefits were wide ranging, 10 major components of the NIH participated in the planning for the conference.

The NIH planning group focused the meeting on the following four questions concerning marijuana as a potential therapeutic agent:

Question 1 – What research has been done previously and what is currently known about the possible medical uses of marijuana?

Question 2 – What are the major unanswered scientific questions?

Question 3 – What are the diseases or conditions for which marijuana might have potential as a treatment and that merit further study?

Question 4 – What special issues have to be considered in conducting clinical studies of the therapeutic uses of marijuana?

The meeting was formatted as a scientific workshop. It was not an attempt to render a consensus. Therefore, it was structured so that speakers with experience in the relevant therapeutic areas would present to a group of eight expert consultants who possessed broad expertise in clinical studies and therapeutics and who had no public positions on the potential use of marijuana as a therapeutic agent. Each presentation was followed by a session for questions and answers from the Expert Group. The second day was allotted for the public to present their views and for discussion by the Expert Group. This report represents a compilation of the views of the Expert Group. Since this report was not intended as a general review of the literature on marijuana and THC, only a few selected references from among the thousands that exist are cited. Each of the members in the Expert Group chose those references relevant to their own contributions to the report.

The Pharmacology of Natural Products

It is important to keep in mind that marijuana is not a single drug. Marijuana is a mixture of the dried flowering tops and leaves from the plant cannabis sativa (Agurell et al. 1984; Graham 1976; Jones 1987; Mechoulam 1973). Like most plants, marijuana is a variable and complex mixture of biologically active compounds (Agurell et al. 1986; Graham 1976; Mechoulam 1973). Characterizing the clinical pharmacology of the constituents in any pharmacologically active plant is often complicated, particularly when the plant is smoked or eaten more or less in its natural form. Marijuana is not unusual in this respect. Cannabis sativa is a very adaptive plant, so its characteristics are even more variable than most plants (Graham 1976; Mechoulam 1973). Some of the seeming inconsistency or uncertainty in scientific reports describing the clinical pharmacology of marijuana results from the inherently variable potency of the plant material used in research studies. Inadequate control over drug dose when researching the effects of smoked and oral marijuana, together with the use of research subjects who vary greatly in their past experience with marijuana, contribute differing accounts of what marijuana does or does not do.

The Plant

Marijuana contains more than 400 chemicals. Approximately 60 are called cannabinoids; i.e., C21 terpenes found in the plant and their carboxylic acids, analogs, and transformation products (Agurell et al. 1984, 1986; Mechoulam 1973). Most of the naturally occurring cannabinoids have been identified. Cannabinoids appear in no other plant. Cannabinoids have been the subject of much research, particularly since the mid 1960s when Mechoulam and his colleagues first isolated delta-9-tetrahydrocannabinol (9-THC) (Mechoulam 1973; Mechoulam et al. 1991). THC in the scientific literature is termed 9-THC or 1-THC depending on whether the pyran or monoterpinoid numbering system is used.

Cannabinoids of Importance

THC, the main psychoactive cannabinoid in marijuana, is an optically active resinous substance. THC is not soluble in water but is extremely lipid soluble (Agurell et al. 1984, 1986; Mechoulam 1973). Varying proportions of other cannabinoids, mainly cannabidiol (CBD) and cannabinol (CBN), are also present in marijuana, sometimes in quantities that might modify the pharmacology of THC or cause effects of their own. CBD is not psychoactive but has significant anticonvulsant, sedative, and other pharmacologic activity likely to interact with THC (Adams and Martin 1996; Agurell et al. 1984, 1986; Hollister 1986a).

The concentration of THC and other cannabinoids in marijuana varies greatly depending on growing conditions, plant genetics, and processing after harvest (Adams and Martin 1996; Agurell et al. 1984; Graham 1976; Mechoulam 1973). In the usual mixture of leaves and stems distributed as marijuana, concentration of THC ranges from 0.3 percent to 4 percent by weight. However, specially grown and selected marijuana can contain 15 percent or more THC. Thus, a marijuana cigarette weighing 1 gram (g) might contain as little as 3 milligrams (mg) of THC or as much as 150 mg or more.

Potency of Tetrahydrocannabinol

THC is quite potent when compared to most other psychoactive drugs. An intravenous (IV) dose of only a milligram or two can produce profound mental and physiologic effects (Agurell et al. 1984, 1986; Fehr and Kalant 1983; Jones 1987). Large doses of THC delivered by marijuana or administered in the pure form can produce mental and perceptual effects similar to drugs usually termed hallucinogens or psychomimetics. However, the way marijuana is used in the United States does not commonly lead to such profound mental effects. Despite potent psychoactivity and pharmacologic actions on multiple organ systems, cannabinoids have remarkably low lethal toxicity. Lethal doses in humans are not known. Given THC’s potency on some brain functions, the clinical pharmacology of marijuana containing high concentrations of THC, for example greater than 10 percent, may well differ from plant material containing only 1 or 2 percent THC simply because of the greater dose delivered.

Some Limitations of Previous Marijuana Research

Unfortunately, much of what is known about the human pharmacology of smoked marijuana comes from experiments with plant material containing about 2 percent THC or less, or occasionally up to 4 percent THC. In addition, human experiments typically are done in laboratory settings where only one or two smoked doses were administered to relatively young, medically screened, healthy male volunteers well experienced with the effects of marijuana. Females rarely participated in past marijuana research because of prohibitions (now removed) against their inclusion. Thus the clinical pharmacology of single or repeated smoked marijuana doses given to older people or to people with serious diseases has hardly been researched at all in a controlled laboratory or clinic setting. Some of the very few reports of experiments that have included older or sicker people, particularly patients less experienced in using marijuana, suggest the profile of adverse effects may differ from healthy student volunteers smoking in a laboratory experiment (Hollister 1986a, 1988a).

THC administered alone in its pure form is the most thoroughly researched cannabinoid. Much of what is written about the clinical pharmacology of marijuana is actually inferred from the results of experiments using only pure THC. Generally, in experiments actually using marijuana, the assumed dose of marijuana was based only on the concentration of THC in the plant material. The amounts of cannabidiol and other cannabinoids in the plant also vary so that pharmacologic interactions modifying the effects THC may occur when marijuana is used instead of pure THC. Only rarely in human experiments using marijuana was the content of CBD or other cannabinoids specified or the possibility of interactive effects between THC and other cannabinoids or other marijuana constituents actually measured.

The result of this research strategy is that a good deal is known about the pharmacology of THC, but experimental confirmation that the pharmacology of a marijuana cigarette is indeed entirely or mainly determined by the amount of THC it contains remains to be completed. The scientific literature contains occasional hints that the pharmacology of pure THC, although similar, is not always the same as the clinical pharmacology of smoked marijuana containing the same amount of THC (Graham 1976; Harvey 1985; Institute of Medicine 1982). Proponents of therapeutic applications of marijuana emphasize possible but not well documented or proven differences between the effects of the crude plant and pure constituents like THC (Grinspoon and Bakalar 1993).

Route-Dependent Pharmacokinetics

Route of administration determines the pharmacokinetics of the cannabinoids in marijuana, particularly absorption and metabolism (Adams and Martin 1996; Agurell et al. 1984, 1986). Typically, marijuana is smoked as a cigarette (a joint) weighing between 0.5 and 1.0 g, or in a pipe in a way not unlike tobacco smoking. Marijuana can also be baked in foods and eaten, or ethanol or other extracts of plant material can be taken by mouth. Some users claim marijuana containing adequate THC can be heated without burning and the resulting vapor inhaled to produce the desired level of intoxication. This has not been studied under controlled conditions. Pure preparations of THC and other cannabinoids can be administered by mouth, by rectal suppository, by IV injection, or smoked. IV injection of crude extracts of marijuana plant material would be quite toxic, however.

Marijuana Smoking and Oral Administration

Smoking plant material is a special way of delivering psychoactive drugs to the brain. Smoking has different behavioral and physiologic consequences than oral or IV administration. What is well known about tobacco (nicotine) and coca (cocaine) clinical psychopharmacology and toxicity illustrates this point all too well. When marijuana is smoked, THC in the form of an aerosol in the inhaled smoke is absorbed within seconds and delivered to the brain rapidly and efficiently as would be expected of a very lipid-soluble drug. Peak venous blood levels of 75 to 150 nanograms per milliliter (ng/mL) of plasma appear about the time smoking is finished (Agurell et al. 1984, 1986; Huestis et al. 1992a, 1992b). Arterial concentrations of THC have not been measured but would be expected to be much higher initially than venous levels, as is the case with smoked nicotine or smoked cocaine.

Oral ingestion of THC or marijuana is quite different than smoking. Maximum THC and other cannabinoid blood levels are only reached 1 to 3 hours after an oral dose (Adams and Martin 1996; Agurell et al. 1984, 1986). Onset of psychoactive and other pharmacologic effects is rapid after smoking but much slower after oral doses.

Marijuana Smoking Behavior and Dose Control

As with any smoked drug (e.g., nicotine or cocaine), characterizing the pharmacokinetics of THC and other cannabinoids from smoked marijuana is a challenge (Agurell et al. 1986; Heishman et al. 1989; Herning et al. 1986; Heustis et al. 1992a). A person’s smoking behavior during an experiment is difficult for a researcher to control. People differ. Smoking behavior is not easily quantified. An experienced marijuana smoker can titrate and regulate dose to obtain the desired acute psychological effects and to avoid overdose and/or minimize undesired effects. Each puff delivers a discrete dose of THC to the body. Puff and inhalation volume changes with phase of smoking, tending to be highest at the beginning and lowest at the end of smoking a cigarette. Some studies found frequent users to have higher puff volumes than did less frequent marijuana users. During smoking, as the cigarette length shortens, the concentration of THC in the remaining marijuana increases; thus, each successive puff contains an increasing concentration of THC.

One consequence of this complicated process is that an experienced marijuana smoker can regulate almost on a puff-by-puff basis the dose of THC delivered to lungs and thence to brain. A less experienced smoker is more likely to overdose or underdose. Thus a marijuana researcher attempting to control or specify dose in a pharmacologic experiment with smoked marijuana has only partial control over drug dose actually delivered. Postsmoking assay of cannabinoids in blood or urine can partially quantify dose actually absorbed after smoking, but the analytic procedures are methodologically demanding, and only in recent years have they become at all practical.

After smoking, venous blood levels of THC fall precipitously within minutes, and an hour later they are about 5 to 10 percent of the peak level (Agurell et al. 1986; Huestis et al. 1992a, 1992b). Plasma clearance of THC is quite high, 950 milliliters per minute (mL/min) or greater; thus approximating hepatic blood flow. However, the rapid disappearance of THC from blood is largely due to redistribution to other tissues in the body rather than simply because of rapid cannabinoid metabolism (Agurell et al. 1984, 1986). Metabolism in most tissues is relatively slow or absent. Slow release of THC and other cannabinoids from tissues and subsequent metabolism makes for a very long elimination half-time. The terminal half-life of THC is estimated to be from about 20 hours to as long as 10 to 13 days, though reported estimates vary as expected with any slowly cleared substance and the use of assays with varied sensitivity.

Cannabinoid metabolism is extensive with at least 80 probably biologically inactive but not completely studied metabolites formed from THC alone (Agurell et al. 1986; Hollister 1988a). 11-hydroxy-THC is the primary active THC metabolite. Some inactive carboxy metabolites have terminal half-lives of 50 hours to 6 days or more and thus serve as long persistence markers of prior marijuana use by urine tests. Most of the absorbed THC dose is eliminated in feces and about 33 percent in urine. THC enters enterohepatic circulation and undergoes hydroxylation and oxidation to 11-nor-9-carboxy-delta-9-THC (9-COOH-9-THC). The glucuronide is excreted as the major urine metabolite along with about 18 nonconjugated metabolites. Frequent and infrequent marijuana users are similar in the way they metabolize THC (Agurell et al. 1986; Kelly and Jones 1992).

Route of Use Bioavailability and Dose

THC bioavailability, i.e., the actual absorbed dose as measured in blood, from smoked marijuana varies greatly among individuals. Bioavailability can range from 1 percent to 24 percent with the fraction absorbed rarely exceeding 10 percent to 20 percent of the THC in a marijuana cigarette or pipe (Agurell et al. 1986; Hollister 1988a). This relatively low and quite variable bioavailability results from significant loss of THC in sidestream smoke, from variation in individual smoking behaviors, from incomplete absorption from inhaled smoke, and from metabolism in lung and cannabinoid pyrolysis. A smoker’s experience is probably an important determinant of dose actually absorbed (Herning et al. 1986; Johansson et al. 1989). Much more is known about the dynamics of tobacco (nicotine) smoking. Many of the same pharmacokinetic considerations apply to marijuana smoking.

Oral bioavailability of THC, whether given in the pure form or as THC in marijuana, also is low and extremely variable, ranging between 5 percent and 20 percent (Agurell et al. 1984, 1986). Great variation can occur even when the same individual is repeatedly dosed under controlled and ideal conditions. THC’s low and variable oral bioavailability is largely a consequence of large first-pass hepatic elimination of THC from blood and due to erratic absorption from stomach and bowel. Because peak effects are slow in onset and variable in intensity, typically at least an hour or two after an oral dose, it is more difficult for a user to titrate dose than with marijuana smoking. When smoked, THC’s active metabolite 11-hydroxy-THC probably contributes little to the effects since relatively little is formed, but after oral doses the amounts of 11-hydroxy-THC metabolite may exceed that of THC and thus contribute to the pharmacologic effects of oral THC or marijuana.

Mental and Behavioral Effects

Common Acute Effects

Usually the mental and behavioral effects of marijuana consist of a sense of well-being (often termed euphoria or a high), feelings of relaxation, altered perception of time and distance, intensified sensory experiences, laughter, talkativeness, and increased sociability when taken in a social setting. Impaired memory for recent events, difficulty concentrating, dreamlike states, impaired motor coordination, impaired driving and other psychomotor skills, slowed reaction time, impaired goal-directed mental activity, and altered peripheral vision are common associated effects (Adams and Martin 1996; Fehr and Kalant 1983; Hollister 1988a; Institute of Medicine 1982; Tart 1971).

With repeated exposure, varying degrees of tolerance rapidly develops to many subjective and physiologic effects (Fehr and Kalant 1983; Jones 1987). Thus, intensity of acute effects is determined not only by THC dose but also by past experience, setting, expectations, and poorly understood individual differences in sensitivity. After a single moderate smoked dose most mental and behavioral effects are easily measurable for only a few hours and are usually no longer measurable after 4 to 6 hours (Hollister 1986a, 1988a). A few published reports describe lingering cognitive or behavioral changes 24 hours or so after a single smoked or oral dose (Fehr and Kalant 1983; Institute of Medicine 1982; Yesavage et al. 1985). Venous blood levels of THC or other cannabinoids correlate poorly with intensity of effects and character of intoxication (Agurell et al. 1986; Barnett et al. 1985; Huestis et al. 1992a).

Adverse Mental Effects

Large smoked or oral marijuana doses or even ordinary doses taken by a sensitive, inexperienced, or predisposed person can produce transient anxiety, panic, feelings of depression and other dysphoric mood changes, depersonalization, bizarre behaviors, delusions, illusions, or hallucinations (Adams and Martin 1996; Fehr and Kalant 1983; Hollister 1986a, 1988a; Institute of Medicine 1982). Depending on the mix of symptoms and behaviors, the state has been termed an acute panic reaction, toxic delirium, acute paranoid state, or acute mania. The unpleasant effects are usually of sudden onset, during or shortly after smoking, or appear more gradually an hour or two after an oral dose, usually last a few hours, less often a few days, and completely clear without any specific treatment other than reassurance and a supportive environment. A subsequent marijuana dose, particularly a lower one, may be well tolerated. In a large survey of regular marijuana users, 17 percent of young adult respondents reported experiencing at least one of the preceding symptoms during at least one occasion of marijuana use, usually early in their use (Tart 1971).

Whether marijuana can produce or trigger lasting mood disorders (depression or mania) or schizophrenia is less clearly established (Fehr and Kalant 1983; Gruber and Pope 1994; Hollister 1986a, 1988a; Institute of Medicine 1982). A psychotic state with schizophrenic-like and manic features lasting a week or more has been described. Marijuana can clearly worsen schizophrenia. Chronic marijuana use can be associated with behavior characterized by apathy and loss of motivation along with impaired educational performance even without obvious behavioral changes (Pope and Yurgelun-Todd 1996; Pope et al. 1995). The explanation and mechanisms for this association are still not well established.

Cardiovascular and Autonomic Effects

A consistent, prominent, and sudden effect of marijuana is a 20 to 100 percent increase in heart rate lasting up to 2 to 3 hours (Hollister 1986a, 1988a; Jones 1985). After higher smoked or oral doses postural hypotension and associated faintness or dizziness can occur upon standing up from a supine or prone position. Tolerance to these effects appears after only a few days of two to three times per day dosing (Benowitz and Jones 1981; Jones 1985). Typical is a modest increase in supine blood pressure. Cardiac output can increase 30 percent when supine. Peripheral vascular resistance decreases with the greatest drop in resistance in skeletal muscles. Skin temperature drops are large; 4 to 6 degrees centigrade, even after a modest smoked dose and roughly parallel to plasma norepinephrine increases. With a few days of repeated exposure to frequent doses of oral THC or marijuana extract, supine blood pressure falls, the sometimes marked initial orthostatic hypotension disappears, blood volume increases, and heart rate slows (Benowitz and Jones 1981). Thus like other system effects, the intensity and character of many hemodynamic effects of single smoked doses in humans are a function of recent marijuana exposure, dose, and even body position.

The cardiovascular effects of smoked or oral marijuana have not presented any health problems for healthy and relatively young users. However, marijuana smoking by older patients, particularly those with some degree of coronary artery or cerebrovascular disease, is likely to pose greater risks because of the resulting increased cardiac work, increased catecholamines, carboxyhemoglobin, and postural hypotension (Benowitz and Jones 1981; Hollister 1988a). Such issues have not been well addressed in past marijuana research.

Respiratory System Effects

Pulmonary effects associated with marijuana smoking include transient bronchodilation after acute exposure. Chronic bronchitis and pharyngitis are associated with repeated exposure with an increased frequency of pulmonary illness. With chronic marijuana smoking, large-airway obstruction is evident on pulmonary function tests, and cellular inflammatory histopathological abnormalities appear in bronchial epithelium (Adams and Martin 1996; Hollister 1986a). These effects appear to be additive to those produced by tobacco smoking.

Endocrine System

Endocrine system effects include a moderate depression of spermatogenesis and sperm motility and a decrease in plasma testosterone in males. Prolactin, FSH, LH, and GH levels are decreased in females. Although suppressed ovulation and other ovulatory cycle changes occur in nonhuman primates, a study of human females smoking marijuana in a research hospital setting did not find hormone or menstrual cycle changes like those in the monkeys given THC (Mendelson and Mello 1984; Mendelson et al. 1984a). Relatively little research has been done on experimentally administered marijuana effects on human female endocrine and reproductive system function.

Immune System

THC and other cannabinoids in marijuana have immunosuppressant properties producing impaired cell-mediated and humoral immune system responses. A large literature describes the results of experiments with animal and animal tissue in in vivo and in vitro model systems. THC and other cannabinoids suppress antibody formation, cytokine production, leukocyte migration and natural killer-cell activity. Cannabinoids decrease host resistance to infection from bacterial and viral infection in animals. Marijuana smokers show evidence of impaired immune function: for example, decreased leukocyte blastogenesis in response to mitogens. Marijuana smokers, when compared to nonmarijuana smokers, have more respiratory illness (Polen et al. 1993).

The cannabinoids have been characterized as immunomodulators because although they generally suppress, they occasionally enhance some immune responses (Friedman et al. 1995). Reviews of marijuana immune system effects have characterized the effects as complicated or conflicting or controversial (Adams and Martin 1996; Hollister 1988b). The clinical significance or relevance of these findings remains uncertain. Much of the complexity and controversy results from the use of mostly in vitro animal models, or in vitro animal and human cell cultures, or in vivo animal studies. Generally in most studies the cannabinoid doses or concentrations used have been quite high when compared to reasonable levels of exposure in human marijuana smoking.

Suppressed or impaired immune mechanisms would likely have negative effects on health by increasing susceptibility to infection or to tumors. People with compromised immune systems or existing malignancies may be at higher risk than healthy people. For example, the risk of developing AIDS may be higher with HIV infection, with a higher risk for infection by opportunistic bacteria, fungi, or viruses. On the other hand, some have suggested that the immunosuppressive effects of cannabinoids might be useful clinically; for example, in treating multiple sclerosis, mostly reasoning from theoretical assumptions or experimental disease models in animals.

In summary, there is good evidence that THC and other cannabinoids can impair both cell-mediated and humoral immune system functioning, leading to decreased resistance to infection by viruses and bacteria. However, the health relevance of these findings to human marijuana use remains uncertain. Conclusive evidence for increased malignancy, or enhanced acquisition of HIV, or the development of AIDS, has not been associated with marijuana use.

There is a need for further research, particularly in circumstances where long-term administration of marijuana might be considered for therapeutic purposes; for example, in individuals who are HIV-positive or who have tumors, malignancies, or diseases where immune system function may be important in the genesis of the disease. Clinical studies with smoked marijuana in patients with compromised immune systems may offer a sensitive index of adverse immune system effects associated with cannabinoid exposure. Direct measures of viral load and other sensitive indices of immune system function are now more practical than in past years when most of the cannabinoid immune system research was carried out. The possibility that frequent and prolonged marijuana use might lead to clinically significant impairments of immune system function is great enough that such studies should be part of any marijuana medication development research, particularly when marijuana will be used by patients with compromised immune systems.

Tolerance and Physical Dependence

After repeated smoked or oral marijuana doses, marked tolerance is rapidly acquired (after a day or two) to many marijuana effects, e.g., cardiovascular, autonomic, and many subjective effects. After exposure is stopped, tolerance is lost with similar rapidity (Jones et al. 1981). Measurable tolerance or tachyphalaxis is evident for some hours after smoking even a single marijuana cigarette.

Withdrawal symptoms and signs appearing within hours after cessation of repeated marijuana use have been occasionally reported by patients in clinical settings (Duffy and Milin 1996; Mendelson et al. 1984b). A withdrawal syndrome was reliably produced by as little as 5 days of modest but frequent oral doses of THC or marijuana extract in double-blind, placebo-controlled experiments (Jones et al. 1981). THC decreased or relieved the symptoms. Typical symptoms and signs were restlessness, insomnia, irritability, salivation, tearing, nausea, diarrhea, increased body temperature, anorexia, weight loss, tremor, sweating, sleep brainwave rapid eye movement rebound, and subjective sleep disturbance. Increased dreaming contributing to the sleep disturbance sometimes persisted for weeks, but the other signs and symptoms were gone or markedly diminished within 48 hours after the last oral marijuana dose.

Drug Interactions With Marijuana

Tobacco, ethanol, and other psychoactive and therapeutic drugs commonly consumed together with marijuana share metabolic pathways with cannabinoids, so metabolic interactions are likely. Both THC and CBD inhibit the metabolism of drugs metabolized by hepatic mixed-function oxidase enzymes (Benowitz and Jones 1977; Benowitz et al. 1980; Hollister 1986b).

The absorption or clearance of other drugs taken with marijuana may be slowed or hastened depending on timing and sequence of drug ingestion and past exposure. For example, ethanol consumed just after smoking a marijuana cigarette produces a much lower peak blood level than the same dose of ethanol taken an hour before marijuana smoking because THC slows gastric emptying time, thus slowing absorption of ethanol.

THC is highly bound to plasma proteins (97 percent to 99 percent) and thus is likely to interact with other highly bound drugs because of competition for binding sites on plasma proteins.

Finally, there is experimental evidence for drug interactions at the functional (neural) adaptation level (Adams and Martin 1996).

By those and possibly by other mechanisms, recent or concurrent THC or CBD exposure measurably alters the pharmacokinetics and/or effects of ethanol, barbiturates, nicotine, amphetamines, cocaine, phencyclidine, opiates, atropine, and clomipramine (Fehr and Kalant 1983; Institute of Medicine 1982). Marijuana use is likely to alter the pharmacology of some concurrently used therapeutic drugs, e.g., cancer chemotherapeutic agents or anticonvulsants.

Cannabinoid Receptors

Mechanisms of psychoactive cannabinoid action were long suspected to be through interactions of/with lipid components of cell membranes (Adams and Martin 1996; Hollister 1988a). The discovery of cannabinoid receptors in the human brain in the late 1980s led to renewed interest in the pharmacology and potential therapeutic uses of cannabinoids (Adams and Martin 1996; Herkenham 1992). The mechanisms of action of THC are now assumed to be mainly receptor mediated. So far, it still is a relatively simple receptor family (CB 1 and CB 2). Receptors are abundant in brain areas concerned with memory, cognition, and motor coordination. An endogenous ligand, a fatty acid derivative named anandamide, has been identified but not yet studied in humans (Thomas et al. 1996). A specific THC antagonist, SR141716A, provokes intense withdrawal signs and behaviors in rodents that have been exposed to THC for even relatively brief periods (Adams and Martin 1996). The clinical pharmacology of the antagonist has not been studied in humans.


Adams, I.B., and Martin, B.R. Cannabis: Pharmacology and toxicology in animals and humans. Addiction 91(11):1585-1614, November 1996.

Agurell, S., Dewey, W.L., and Willett, R.E., eds. The Cannabinoids: Chemical, Pharmacologic, and Therapeutic Aspects. New York: Academic Press, 1984.

Agurell, S.; Halldin, M.; Lindgren, J.E.; Ohlsson, A.; Widman, M.; Gillespie, H.; and Hollister, L. Pharmacokinetics and metabolism of delta 1-tetrahydrocannabinol and other cannabinoids with emphasis on man. Pharmacol Rev 38(1):21-43, March 1986.

Barnett, G.; Licko, V.; and Thompson, T. Behavioral pharmacokinetics of marijuana. Psychopharmacology 85(1):51-56, 1985.

Benowitz, N.L., and Jones, R.T. Effect of delta-9-tetrahydrocannabinol on drug distribution and metabolism: Antipyrine, pentobarbital and ethanol. Clin Pharmacol Ther 22(3):259-268, 1977.

Benowitz, N.L., and Jones, R.T. Cardiovascular and metabolic considerations in prolonged cannabinoid administration in man. J Clin Pharmacol 21:214S-223S, 1981.

Benowitz, N.L.; Nguyen, T.; Jones, R.T.; Herning, R.I.; and Bachman, J. Metabolic and psychophysiologic studies of cannabidiol-hexobarbital interaction. Clin Pharmacol Ther 28:115-120, 1980.

Duffy, A., and Milin, R. Case study: Withdrawal syndrome in adolescent chronic cannabis users. J Am Acad Child Adolesc Psychiatry 35(12):1618-1621, December 1996.

Fehr, K., and Kalant, H., eds. ARF/WHO Scientific Meeting on Adverse Health and Behavioral Consequences of Cannabis Use (1981: Toronto, Canada) Cannabis and Health Hazards: Proceedings of an ARF/WHO Scientific Meeting on Adverse Health and Behavioral Consequences of Cannabis Use. Toronto, Canada: Addiction Research Foundation, 1983.

Friedman, H.; Klein, T.W.; Newton, C.; and Daaka, Y. Marijuana, receptors and immunomodulation. Adv Exp Med Biol 373:103-113, 1995.

Graham, J.D.P., ed. Cannabis and Health. New York: Academic Press, 1976.

Grinspoon, L., and Bakalar, J.B. Marihuana, the Forbidden Medicine. New Haven: Yale University Press, 1993.

Gruber, A.J., and Pope, H.G. Cannabis psychotic disorder: Does it exist? Am J Addict v3 (n1):72-83, Winter 1994.

Harvey, D.J., ed.¬†Satellite Symposium on Cannabis (3rd: 1984: Oxford, England) Marihuana ’84: Proceedings of the Oxford Symposium on Cannabis. Washington, DC: IRL Press, 1985.

Heishman, S.J.; Stitzer, M.L.; and Yingling, J.E. Effects of tetrahydrocannabinol content on marijuana smoking behavior, subjective reports, and performance.Pharmacol Biochem Behav 34(1):173-179, September 1989.

Herkenham, M. Cannabinoid receptor localization in brain: Relationship to motor and reward systems. In: Kalivas, P.W., and Samson, H.H., eds. The neurobiology of drug and alcohol addiction. Ann N Y Acad Sci 654:19-32, 1992.

Herning, R.I.; Hooker, W.D.; and Jones, R.T. Tetrahydrocannabinol content and differences in marijuana smoking behavior. Psychopharmacology 90(2):160-162, 1986.

Hollister, L.E. Health aspects of cannabis. Pharmacol Rev 38(1):1-20, March 1986a.

Hollister, L.E. Interactions of cannabis with other drugs in man. In: Braude, M.C., and Ginzburg, H.M., eds. Strategies for Research on the Interactions of Drugs of Abuse. National Institute on Drug Abuse Research Monograph 68. DHHS Pub. No. (ADM)86-1453. Washington, DC: Supt. of Docs., U.S. Govt. Print. Off., 1986b. pp. 110-116.

Hollister, L.E. Cannabis–1988. (Literature review).¬†Acta Psychiatr Scand (Suppl) 78(345):108-118, 1988a.

Hollister, L.E. Marijuana and immunity. J Psychoactive Drugs 20(1:):3-8, January-March 1988b.

Huestis, M.A.; Henningfield, J.E.; and Cone, E.J. Blood Cannabinoids. 1. Absorption of THC and formation of 11-OH-THC and THC COOH during and after smoking marijuana. J Anal Toxicol 16(5):276-282, September-October 1992a.

Huestis, M.A.; Sampson, A.H.; Holicky, B.J.; Henningfield, J.E.; et al. Characterization of the absorption phase of marijuana smoking. Clin Pharmacol Ther 52 (1):31-41, July 1992b.

Institute of Medicine. Division of Health Sciences Policy. Marijuana and Health: Report of a Study by a Committee of the Institute of Medicine, Division of Health Sciences Policy. Washington, DC: National Academy Press, 1982.

Johansson, E.; Halldin, M.M.; Agurell, S.; Hollister, L.E.; and Gillespie, H.K. Terminal elimination plasma half-life of delta 1-tetrahydrocannabinol (delta 1-THC) in heavy users of marijuana. Eur J Clin Pharmacol 37(3):273-277, 1989.

Jones, R.T. Drug of abuse profile: Cannabis. Clin Chem 33 (11 Suppl):72B-81B, October 1987.

Jones, R.T. Cardiovascular effects of cannabinoids. In: Harvey, D.J., ed.¬†Marihuana, ’84: Proceedings of the Oxford Symposium on Cannabis. Oxford: IRL Press, 1985. pp. 325-334.

Jones, R.T.; Benowitz, N.L.; and Herning, R.I. Clinical relevance of cannabis tolerance and dependence. J Clin Pharmacol 21:143S-152S, 1981.

Kelly, P., and Jones, R.T. Metabolism of tetrahydrocannabinol in frequent and infrequent marijuana users. J Anal Toxicol 16:228-235, 1992.

Mechoulam, R., ed. Marijuana: Chemistry, Pharmacology, Metabolism and Clinical Effects. New York: Academic Press, 1973.

Mechoulam, R.; Devane, W.A.; Breuer, A.; and Zahalka, J. A random walk through a cannabis field. Special Issue: Pharmacological, chemical, biochemical and behavioral research on cannabis and the cannabinoids. Pharmacol Biochem Behav 40(3):461-464, November 1991.

Mendelson, J.H., and Mello, N.K. Effects of marijuana on neuroendocrine hormones in human males and females. In: Braude, M.C., and Ludford, J.P., eds.Marijuana Effects on the Endocrine and Reproductive Systems. National Institute on Drug Abuse Research Monograph 44. DHHS Pub. No. (ADM)84-1278. Washington, DC: Supt. of Docs., U.S. Govt. Print. Off., 1984. pp. 97-114.

Mendelson, J.H.; Mello, N.K.; Cristofaro, P.; Ellingboe, J.; and Benedikt, R. Acute effects of marijuana on pituitary and gonadal hormones during the periovulatory phase of the menstrual cycle. In: Harris, L.S., ed. Problems of Drug Dependence, 1984: Proceedings of the 46th Annual Scientific Meeting, The Committee on Problems of Drug Dependence, Inc. National Institute on Drug Abuse Research Monograph 55. DHHS Pub. No. (ADM)85-1393. Washington, DC: Supt. of Docs., U.S. Govt. Print. Off., 1984a. pp. 24-31.

Mendelson, J.H.; Mello, N.K.; Lex, B.W.; and Bavli, S. Marijuana withdrawal syndrome in a woman. Am J Psychiatry 141(10):1289-1290, October 1984b.

Polen, M.R.; Sidney, S.; Tekawa, I.S.; Sadler, M.; and Friedman, G.D. Health care use by frequent marijuana smokers who do not smoke tobacco. West J Med158(6):596-601, June 1993.

Pope, H.G., Jr., and Yurgelun-Todd, D. The residual cognitive effects of heavy marijuana use in college students. JAMA 275(7):521-527, February 21, 1996.

Pope, H.G.; Gruber, A.J.; and Yurgelun-Todd, D. The residual neuropsychological effects of cannabis: The current status of research. Drug Alcohol Depend38(1):25-34, April 1995.

Tart, C.T. On Being Stoned: A Psychological Study of Marijuana Intoxication. Palo Alto, CA: Science and Behavior Books, 1971.

Thomas, B.F.; Adams, I.B.; Mascarella, S.W.; Martin, B.R.; and Razdan, R.K. Structure-activity analysis of anandamide analogs: Relationship to a cannabinoid pharmacophore. J Med Chem 39(2):471-497, January 19, 1996.

Yesavage, J.A.; Leirer, V.O.; Denari, M.; and Hollister, L.E. Carry-over effects of marijuana intoxication on aircraft pilot performance: A preliminary report. Am J Psychiatry 142(11):1325-1329, November 1985.