As several countries in Europe have taken policies to decrease the penalties for cannabis possession, many people especially young persons have interpreted this move as giving support to consider cannabis as a benign drug 1.

However as stated by several reports cannabis is not a harmless substance and requires urgent attention considering public health issues such as car driving for example 2. The relationship between Cannabis and acute psychosis is another important issue. In Pakistan and also India, Bhang, a beverage made from an infusion of cannabis leaves, and flowering tops combined with milk and nuts is reported to frequently induce psychotic manifestations among consumers 3. Presenting symptoms include grandiosity, excitement, hostility, uncooperativeness, disorientation, hallucinatory behaviour and unusual thought content 3.

Recently five large-scale longitudinal studies and a systematic review have shown that cannabis use in adolescence is associated with a two-to threefold increase in the relative risk of later developing schizophrenia 4. Furthermore short-term psychotic reactions, particularly in naive users have also been reported. Thomas 5 describes that one in seven people reported psychotic-like symptoms. Such reactions are usually acute, transient, self-limited however very unpleasant ("hearing voices, becoming convinced that someone is trying to harm you or that you are persecuted") 6. But cannabinoids are considered able to trigger long-lasting psychotic decompensations in predisposed individuals, which may in part account for the epidemiological association described between cannabis consumption and psychotic disorders 78.

The therapeutic aspects of cannabis represent additional issues, as they are in constant development since several years. Synthetic THC (dronabinol) is available for restricted medical use in the USA since 1985. Nabilone, a synthetic THC analogue, is licensed in UK for the treatment of nausea and vomiting caused by chemotherapy unresponsive to usual anti-emetics. Clinical applications actually include nausea and vomiting, muscle spasticity in demyelinating diseases, loss of appetite in cancer and AIDS, pain, insomnia, asthma as well as other applications 9.

As these oral medications are becoming increasingly available, we think it is useful to report two cases of severe psychological sides effects, especially considering the lack of data in the literature on psychotic symptoms associated with oral synthetic or natural THC.

We report two cases out of 8 healthy male volunteers who were included in a double blind crossover clinical study, approved by the ethics committee of the Department of Internal Medicine of the University of Lausanne. All subjects had to be occasional but regular cannabis users. Their urines were controlled to be negative for any drug of abuse (cannabis, opiates, amphetamines, cocaine, benzodiazepines) before each study period. The presence of ethanol was checked using a breathalyzer. All of them provided their written informed consent. This study was carried out to assess the effects of delta-9-tetrahydrocannabinol (THC) on psychomotor function and driving performance. It compared a medication containing 20 mg dronabinol (Marinol^R), and 2 hemp milk decoction containing either a medium (15.8 mg average dose determination) or a high dose of THC (45.7 mg) with matched placebos. The hemp plant fragments containing 1.5 % THC and 4.4 % THC-A were provided by Hiscia institute in Arlesheim, Switzerland. After administration, blood was sampled at regular intervals for cannabinoids determination by gas chromatography coupled with mass spectrometry (GC-MS-NCI). Clinical observations and 2 psychometric tests (roadsign recognition speed and accuracy on a tracking task) were also carried out. Furthermore, the subjects were asked to report their willingness to drive and the subjective effects on a VAS scale extending from 0 to 10 cm. The effects were assessed against placebo.

These 2 cases were withdrawn from the study because of adverse events. We consider them worth reporting for the following reasons: in contrast to most other case reports where circumstances and blood concentrations are not known, our two cases reported here happened under defined clinical setting: the ingested dose, the time-events of effects on behavior and performance as well as the cannabinoid blood levels are fully documented. In addition, the consumption of other psychotropic major drugs could be ruled out.

A temporary form of drug-induced psychotic reaction after administration of oral cannabis has occurred in these two cases. Cannabis psychosis is the term proposed in the literature 10. In 1958, Ames 11 reported in an experimental design with 10 subjects psychological symptoms such as severe anxiety, panic attacks, paranoid delusions and depersonalization. Talbott 12 in 1969 described 12 soldiers in Vietnam who had disorientation and hallucinations after their first use of cannabis. In Germany, 19 cases of toxic psychosis were reported after hashish use 13 and in Calcutta, Chopra 14 described retrospectively 200 patients hospitalized after the ingestion of large dose of cannabis between 1963 and 1968. Other reports in different countries showed similar features after bhang ingestion 315. It usually results from taking large amount of the drug, generally in food or drink. The symptoms have some similarity with paranoid schizophrenia, which could raise the hypothesis that " symptoms of schizophrenic illness might be caused by an abnormal over-activity of endogenous cannabinoid mechanism in the brain" (Iversen 10 citing Emrich 16). However because of the poor quality of information on previous cannabis experience, cannabis dose intake, other drug consumption and previous psychiatric comorbidity, some commentators have criticized these case series 121718. Case-control studies have been conducted comparing people with cannabis psychosis with persons suffering from schizophrenia 192021. However the results were inconsistent due in part to the small sample size of these studies.

The originality of our two cases is that they were observed in an experimental setting, and therefore adds more evidence for the ability of oral cannabis to produce psychotic symptoms. In both our subjects, the effects appeared 1 hour to 1.5 hours after oral drug intake and lasted for 3 to 4 hours. Dronabinol (synthetic THC) is reported to have an onset of action at approximately 0.5 to 1 hour and peak effects between 2 and 4 hours. Psychoactive effects last 4 to 6 hours but the appetite stimulant effect may continue for 24 hours 22.

The issue of THC dose level is very important in terms of public health. A traditional cigarette of herbal cannabis in the 1960s and 1970s contained 1–3% THC: for a joint made of 750 mg of cannabis plant, the corresponding THC amount was 7 to 20 mg. However, the actual amount of cannabis taken up (i.e. the percent delivery to the respiratory tree) strongly depends on the smoking technique; it has been reported to reach approximately 50% 23. Modern cigarettes (joint) based on intensive cannabis selection and improvement in plant cultivation contain 6 to 30% THC. Therefore, an average joint would correspond to 75 mg to 225 mg of THC! 22.

Through smoking, a 3.5% marijuana cigarette with about 900 mg plant materials can achieve plasma concentration in the range of 50 to 100 ng/ml. The maximum psychotropic effect or "high" occurs faster after smoking than by the oral route. Smoking is therefore the preferred route of cannabis administration for young users. Psychomotor function is considered to be obviously impaired above 10 ng/ml plasma THC for smoking cannabis. However in our two cases, the oral administration of cannabis produced circulating THC concentrations much lower than 10 ng/ml. We suggest several explanations for these differences. Firstly, the oral administration produces more active metabolite (11-OH-THC), which could more efficiently reach the effect site than THC. Secondly, as suggested by Chaudry 3, consuming oral cannabis may produce more potent, yet unknown psychotomimetic metabolites of THC. Thirdly, the slow absorption kinetics produces sustained plateau levels in the blood, which could influence the body and brain distribution. In a cocaine fatality, Giroud 24 found that THC and OH-THC were in higher concentration in brain than in blood.

Finally, Leweke 25 in a study including 17 healthy volunteers found also one case that suffered a two-hour episode of paranoid psychotic state following the administration of dronabinol with a lower dose than our study. Furthermore D'Souza 26 administrating intravenous THC to 22 healthy subjects in a double blind randomised clinical trial found a range of transient symptoms resembling those seen in endogenous psychosis. At last, it is important to differentiate these transient psychotic states with spontaneous resolution from the type of psychosis that persist beyond the persistence of drug in the brain, therefore probably indicating a worsening of an underlying pathologic problem.

In conclusion, doctors and users should be aware of the increasing availability of oral cannabis in "special" drinks or food as well as in medications under development. While the oral route of administration achieves only limited blood concentrations, significant psychotic reactions may occur. An increased incidence of psychotic episodes might be induced by this new trend and requires attention regarding this phenomenon in a public health perspective.

The author(s) declare that they have no competing interests.