Hypnotic & Antiepileptic Effects of CBD I

HYPNOTIC AND ANTIEPILEPTIC EFFECTS OF CANNABIDIOL

J Clin Pharmacol. 1981; 21: 417S-427S

Elisaldo A. Carlini, M.D., and Jomar M. Cunha, M.D. Department of
Psychobiology, Escola Paulista de Medicina, Sao Paulo, Brazil.

Abstract: Clinical trials with cannabidiol (CBD) in healthy volunteers,
insomniacs, and epileptic patients conducted in the authors' laboratory
from 1972 up to the present are reviewed. Acute doses of cannabidiol
ranging from 10 to 600 mg and chronic administration of 10 mg. for 20 days
or 3 mg/kg/day for 30 days did not induce psychologic or physical symptoms
suggestive of psychotropic or toxic effects; however, several volunteers
complained of somnolence. Complementary laboratory testsl (EKG, blood
pressure, and blood and urine analysis) revealed no sign of toxicity.
Doses of 40, 80s, and 160 mg cannabidiol were compared to placebo and 5 mg
nitrazepam in 15 insomniac volunteers. Subjects receiving 160 mg
cannabidiol reported having slept significantly more than those receiving
placebo; the volunteers also reported significantly less dream recall with
the three doses of cannabidiol than with placebo. Fifteen patients
suffering from secondary generalized epilepsy refractory to known
antiepileptic drugs received either 200 to 300 mg cannabidiol daily or
placebo for as long as 4.5 months. Seven out of the eight epileptics
receiving cannabidiol had improvement of their disease state, whereas only
one placebo patient improved.

An advertisement published in 1905 by a Brazilian newspaper advised
the use of "Cigarros Indios" prepared from Cannabis indica to treat
insomnia, among other diseases. Anecdotal reports also claim that Queen
Carlota Joaquina of Portugal, while living in Brazil at the beginning of
the last century, used to drink a tea of "Maconha" (Brazilian name for
marihuana) to calm herself. These uses of marihuana in the past were by no
means restricted to Brazil. Its use for sedative, hypnotic, anesthetic,
and antiepileptic effects goes back to ancient times and all continents.
(1-3)
In our times, the unraveling of the chemistry of Cannabis sativa
brought renewed interest in old reports concerning the therapeutic
properties of marihuana, and much new research has been carried out. Quite
expectedly, the attention has focused on delta-9-tetrahydrocannabinol
(THC), considered the main active principle of marihuana, due to its
remarkable psychotropic properties. Its hypnotic and anticonvulsive
effects in animals and in man have been reported by several laboratories,
(4-11) but eventual medical use of THC would be hindered by its
hallucinogenic effects. Furthermore, THC also possesses convulsant
activity. (12-14)
Cannabidiol (CBD), on the other hand, lacks hallucinogenic
properties and is therefore a better candidate for therapeutic use. In
fact, work carried out with laboratory animals has shown that cannabidiol
has depressant effects on the central nervous system in such tests as
potentiation of barbiturate sleeping time and decrease of spontaneous motor
activity. (15, 16) It has been further demonstrated that cannabidiol
possesses a rather potent protective effect against certain types of
experimental convulsions in animals (17-20) and an hypnotic-like effect in
the rat. (21)s Therefore, trial of cannabidiol in humans was in order.
This report reviews data obtained in our laboratory from 1972 up to
the present dealing with the effects of acute and chronic administration of
cannabidiol to healthy volunteers and to insomniacs and epileptic patients.
More detailed information of these experiments can be found elsewhere.
(22-25)
ACUTE ADMINISTRATION OF CANNABIDIOL TO HEALTHY VOLUNTEERS (Phase I
Studies)
Three experiments were performed on different occasions.
Experiment 1 (performed November 1972). Four physicians, three
psychologists, and three biologists, all staff members of the Department of
Psychobiology, varying in age from 23 to 42 years (four females and six
males) volunteered. On the first day of the experiment, they underwent a
complete medical checkup, including clinical and neurologic examinations,
EKG and EEG, blood tests (hematocrit, hemoglobin, leukocyte and erythrocyte
counts, bilirubin, oxaloacetic and puruvic transaminases, and creatinine)
and urine tests (osmolarity, pH, albumin, leukocyyte and erythrocyte
counts, cylinders, and crystals). All volunteers were in good health,
showing neither clinical nor laboratory evidence of cardiovascular, renal,
hepatic, or other impairment,
On the morning of the seventh day, two volunteers were randomly
assigned, in a blind procedure, to receive identical gelatin capsules
containing either glucose or 10, 40, 80, or 160 mg cannabidiol. The sample
of crystalline cannabidiol was kindly supplied by the National Institutes
of Health, United States. At 30, 90, 240, and 480 minutes, EKGs were again
performed. EEGs were also obtained 240 and 480 minutes after the dose, and
so were blood and urine samples for laboratory procedures. These
examinations were repeated in the morning of the following day.
All laboratory results obtained on days 7 and 8 were within normal
limits and similar to the values obtained on the first day, without drug.
Physical and neurologic examinations revealed no abnormalities, and the
volunteers did not complain of untoward effects.

Experiment 2 (January 1976). Two male physicians, aged 37 and 47
years, from the staff of the Department of Psychobiology received 600 mg
cannabidiol for two days, 300 mg in the morning and 300 mg in the evening.
The drug was furnished by R. Mechoulam, Department of Natural Products,
University of Jerusalem, Israel. The experimental protocol was identical
to that of experiment 1, with the exception that EEG and blood and urine
tests were not performed. With the exception that one volunteer reported
having slept more heavily on the nights after taking the drug, no abnormal
finding was detected.

Experiment 3 (February and June 1978). The overall design of this
experiment included a study of the interaction between cannabidiol and
alcohol on psychomotor and psychologic functions of healthy volunteers. In
the present review only data concerning the effects of cannabidiol and
placebo will be compared. Details concerning cannabidiol-alcohol
interactions can be found elsewhere. (23)
Ten healthy postgraduate students volunteered (sic males and four
females), 21 to 33 years of age and weighing 52 to 85 kg. Crystalline
cannabidiol (supplied by R. Mechoulam, Israel) and glucose, 200 mg each,
were formulated separately in opaque gelatin capsules. The experiment
comprised four different drug administrations, at weekly intervals, one
being with glucose and another with cannabidiol. The other two were
alcohol and alcohol plus cannabidiol administration and will not be further
referred to here. Inasmuch as drug administration was given randomly in a
double-blind procedure, glucose and cannabidiol ingestions were from one to
three weeks apart. The experiments began at 11:00 A.M. and continued until
4:00 P.M. The subjects were submitted to objective and subjective tests
1/2, 1, 2, and 4 hours after cannabidiol or glucose ingestion. The
following battery of tests, taking about 25 minutes per session, was
applied: cancellation test, differential aptitude test (Form A, the
Psychological Corporation, New York, N.Y.), time production task, finger
tap test, and a subjective drug reaction scale. Details of these tests are
given elsewhere. (23)
Table I presents the results of the first four tests in a condensed
form. For each test the data obtained at times 30, 60, 120, and 240
minutes after drug ingestion were similar and were therefore pooled. It is
seen that in the cancellation test the volunteers reacher very close to the
total of 40 possible correct responses, regardless of whether they were
under placebo or cannabidiol. Also with both treatments they have a
near-perfect performance in the differential aptitude test and reached
above 300 presses per minute in the finger tap test. Finally, Table I
shows that treatment did not influence the capacity of the volunteers to
produce ddd1-minute time. The results obtained with the subjective drug
reaction scale also confirmed the lack of effects of cannabidiol on the
subjective functioning of the volunteers, as they felt equally sober,
non-drugged, alert, and nondizzy under either cannabidiol or glucose.

Chronic Administration of Cannabidiol to Healthy Volunteers (Phase
I Studies)
Two experiments were performed on different occasions.

Experiment 4 (December 1972 and January 1973). Due to the small
amount of cannabidiol available (remaining NIH material from experiment 1
and a sample supplied by Prof. R. Mechoulam, Israel), only four male
volunteers from the Department of Psychobiology staff were selected--two
physicians, one biochemist, and one psychologist, age range 23 to 30 years.
The experiment was run double-blind, and on day 1 complete medical
checkups and complementary laboratory tests were done as in experiment 1.
Two volunteers were randomly assigned to receive 10 mg cannabidiol daily
for 20 days (two gelatin capsules containing 5 mg each, one to be ingested
at 9:00 A.M. and the other at 5:00 P.M. every day). The third volunteer
received two identical capsules with glucose for 20 days, and the fourth
ingested cannabidiol during the first 10 days and was crossed over to
placebo for the last 10 days. Drug ingestion began on day 7, extending to
day 26, and medical and laboratory examinations were repeated in the
morning of days 8, 15, 22, and 27 (corresponding to days 1, 8, 15, and 21
after the beginning drug ingestion).
The two volunteers receiving 10 mg cannabidiol for 20 days and the
one who received it for the fist 10 days did not complain of any physical
discomfort or psychologic disturbnance. However, two volunteers complained
of somnolence with cannabidiol, one after day 3 and another after day 4;
the symptom subsided by day 15s. Physical and neurologic examinations,
EEG, EKG, and blood and urine tests were all within normal limits and did
not differ in the same volunteer regardless of being performed before or at
different times after the beginning of drug ingestion. Further details of
these results can be found elsewhere. (25)
Experiment 5 (between February and April 1976). Sixteen adult
volunteers (11 men and five women), aged 22 to 35 years, with an average
weight of 65 kg. were chosen from the staff of Escola Paulista de Medicina.
They were all in apparent good health, which was confirmed on day 1 when
they were subjected to complete medical and laboratory examinations as
described in experiment 1.
On day 7 (having given informed consent), they were randomly
divided into two groups of eight. Each group received capsules containing
either glucose or cannabidiol. Crystalline cannabidiol was supplied by R.
Mechoulam, Israel. The experiment was performed on a double-blind basis,
and the subjects were instructed to ingest the assigned capsules, one in
the morning and one in the afternoon. Each capsule contained an amount of
cannabidiol equivalent to 1.5 mg/kg, i.e., a daily dosage of 3.0 mg/kg. On
days 3, 7, 31 and 37 after the beginning of drug ingestion, the volunteers
returned to undergo the medical and laboratory examinations as described
above.
The results confirmed the absence of toxic effects of cannabidiol
Thus, of the eight volunteers receiving placebo, one withdrew on the 21st
day for personal reasons; a second placebo subject complained of sudoresis
and "palpitations" from days 7 to 10 in the veins of the feet. Clinical
and laboratory examinations were normal, and the symptoms subsided after
day 11 without intervention being required on the part of the
investigators. On the other hand, none of the eight cannabidiol volunteers
complained of any symptom suggestive of psychotropic effect during the
entire period of the experiment. However, two of them reported
somnolence--one during the first week and the other throughout the 30 days
of drug ingestion. A third subject, with a history of mild insomnia,
reported better sleep during the first week of medication.
For the 16 subjects (placebo and cannabidiol groups), neurologic
and clinical examinations, EEG and EKG tracings, and blood and urine
analysis were within normal limits before, during and after the experiment.
Detailed accounts of these results have been published recently. (24)

Clinical Trial of Cannabidiol as an Hypnotic Drug

The lack of toxicity of cannabidiol observed in our Phase I studies
(experiments 1, 2, 4, and 5) and the complaints of somnolence by several
volunteers receiving the drug (experiments 2, 4, and 5), together with the
reported hypnotic-like effect of cannabidiol in rats, (21) encouraged us
to give a first trial of cannabidiol as an hypnotic agent in humans. This
experiment was carried out during June--November 1977. A detailed account
of the findings was previously published. (22)
A new sample crystalline cannabidiol, kindly supplied by NIH, was
incorporated into gelatin capsules in amounts of 40, 80, and 160 mg. Pills
containing 5 mg nitrazepam (Mogadon kindly supplied by Roche Laboratories)
were crushed and encapsulated. As placebo, 200 mg glucose was encapsulated
in identical gelatin capsules.
After physical examination and psychiatric interview, 15 relatives
of staff members of Escola Paulista de Medicina were selected. They were
chosen on the basis of (a) complaints of difficulty in falling asleep (at
least 1 hour) and poor sleep throughout the night; (b) history of previous
good physical and mental health; (c) no previous history of nonmedical use
of drugs; and (d) no prescribed medicines for at least 15 days before.
In a double-blind procedure and in random order, once a week, the
volunteers were instructed to ingest one assigned capsule. At dinner of
the chosen days, they were requested to refrain from alcohol and coffee,
and to ingest the capsule 30 minutes before going to bed, between 10:00 and
10:30 P.M. All volunteers received all treatments, that is, placebo, 5 mg
nitrazepam, and the three dosages of cannabidiol Therefore, the experiment
lasted five weeks. In the mornings of days after drug administration, the
volunteers answered a ten-point questionnaire, adapted from Bloomfield et
al., (26) and were briefly interviewed. The questions sought information
on induction of sleep (questions 1 and 2), sleep quality (questions 3, 4,
and 5), dream recall (questions 6 and 7), and reawakening (questions 8, 9,
and 10). Answers to questions 1, 2, 3, 4, and 8 had five possibilities,
scored from 0 to 4, score 4 indicting the best hypnotic effect. For
example, the five possible answers to question 1, "How much time elapsed
between going to bed and onset of sleep?" were: less than 15 minutes
(grade 4), between 15 and 30 minutes (grade 3), between 30 and 45 minutes
(grade 2), between 45 and 60 minutes (grade 1) and more than 60 minutes
(grade 0). Answers to questions 5, 6, 7, 9, and 10 were not transformed
into scores, as they had only two options, yes and no.
The results are condensed in Table II. The first noticeable
finding was the placebo effect, which is rather common in this kind of
research. Thus, before starting drug administration all patients took at
least 1 hour to fall asleep (score 0). However, under placebo, six
patients scored grades 3 and 4--five reported having slept within 15 to 30
minutes after going to bed (grade 3) and one in less than 15 minutes (grade
4). Cannabidiol, and also 5 mg. nitrazepam, did not produce a positive
sleep induction effect in the volunteers. Thus, although eight subjects
reported grades 3 and 4 for 160 mg cannabidiol and 5 mg nitrazepam, this
number fell short of statistical significance when compared to the large
value also found for placebo. The quality of sleep induction, assessed by
the answers to question 2, "What was the effect o the medication before you
fell asleep?" also was not influenced by the treatments. Thus, the
majority of the volunteers answered "nil" to that question, and only a few
volunteers (see second row of Table II) answered "pleasant" (grade 3). It
is interesting that Bloomfield et al (26) also reported that secobarbital
failed to influence sleep induction.
Sleep quality was significantly influenced by 160 mg cannabidiol,
as two thirds of the subjects slept more than 7 hours, scoring 3 and 4 in
answer to question 3. Furthermore, most subjects had few interruptions of
sleep (question 4) and reported having a good night's sleep (question 5).
The three doses of cannabidiol were effective in significantly
reducing dream recall of the volunteers. This could have occurred because
cannabidiol may decrease the capacity to dream in itself, that is, the
volunteers dreamed less. Conversely, cannabidiol may also have reduced
dream recall, that is, the volunteers dreamed as usual but could not
remember. The latter possibility would indirectly indicate that
cannabidiol may have decreased the frequency of small awakening periods
during the night sleep which facilitates remembering the oneiric events.
Finally, the last four questions of Table II indicate that
cannabidiol did not induce nightmares or symptoms that could indicate
hangover the following morning.

Antiepileptic Effects of Cannabidiol

This experiment was done between May 1976 and March 1978, and the
results were recently published. (24)
Fifteen epileptic patients (11 women and four men), aged 14 to 49
years (average 24 years) with a documented history of frequent convulsions
for at least one year were selected. These patients were not responding
satisfactorily to their prescribed antiepileptic drugs. The patients were
diagnosed as cases of secondary generalized epilepsy; EEG tracings revealed
irritative activity with temporal projections. They had at least one
generalized convulsive crisis weekly. In the two weeks before cannabidiol
or placebo administration, the number of focal and generalized convulsive
crises was recorded and considered as the baseline to evaluate treatment.
On the first day of the experiment, the patients were submitted to the
examinations described in experiment 1. They were randomly divided into
one group of eight (control) and another group of seven )cannabidiol). One
week later, each group began to receive placebo or cannabidiol capsules in
a double-blind procedure in addition to the antiepileptic drugs they were
already receiving. The patients were instructed to take two or three
capsules daily (containing 100 mg cannabidiol or glucose) and to return to
the hospital every week for clinical and/or laboratory examinations.
Clinical evaluation of drug treatment was made weekly using a score
scale of 0 to 3, which took into consideration absence of convulsive crises
or absence of generalization and self-reported subjective improvement. On
this scale, a score of 3 meant no improvement, that is, no reduction in
crises and no self-reported subjective improvement; a score of 2, small
improvement or self-reported subjective improvement; a score of 1, partial
improvement or absence of generalization of crisis and self-reported
improvement; finally, a score o 0, complete improvement or total absence of
convulsive crisis and self-reported subjective improvement. According to
these criteria, all patients had a score of 3 during the pre drug phase
(baseline).
The results are seen in Table III. At the end of placebo
treatment, seven patients had a median score of 3 (i.e., no improvement),
whereas one patient (patient 7) showed complete improvement (median score
0). Placebo patients 2 and 3, with no improvement, received the capsules
for the fourth week of treatment but did not return. Three other placebo
patients (1, 4, and 5) remained under treatment for 12, 12, and eight
weeks, respectively, after which they were withdrawn from the experiment
and underwent a change in the prescribed antiepileptic drugs in an attempt
to improve their condition. Finally, patient 8 remained on placebo for
four weeks with no improvement (median score 3) and wanted to give up,
being transferred without her knowledge to cannabidiol with a small
improvement (median score 2).
Of the eight patients receiving cannabidiol, four (patients 10, 11,
12, and 15) showed considerable improvement (median score 0); however, in
one case (patient 15) this was achieved by increasing the dosage to 300 mg
daily. Patient 11, who showed much improvement from the first week, moved
to another city after completing six weeks of treatment. Patient 9
improved partially (median score 1) although he attained score 0 ( no
convulsive crisis and subjective improvement) in seven out of the 16 weeks
of treatment. Finally, two of the three remaining patients had a median
score of 2 (only self-reported subjective improvement), whereas the last
patient, 14, did not improve at all in spite of increasing cannabidiol to
300 mg daily for the last two weeks of treatment.
Clinical examination proved normal for all cannabidiol patients,
and the pulse, cardiac, and respiratory rates remained constant during the
course of the experiment. Also cannabidiol did not alter the creatinine,
bilirubin, or transaminase values in the eight patients. However, four
patients complained of somnolence during the experiment.

General Comments

The first relevant finding of these studies concerns the rather
remarkable lack of toxicity from acute and chronic administration of
cannabidiol Thus, in five experiments with healthy volunteers there was
no subjective or physical symptom suggestive of toxic effect. The same
conclusion was also reached after the laboratory examinations. Although in
some experiments the number of subjects was low due to the limited
availability of cannabidiol, taken all together the data strongly support
our contention, regarding the excellent toleration of cannabidiol
Furthermore, the acute administration of the drug to the insomniacs and the
3 to 4.5 months of treatment of the epileptics confirm the lack of
toxicity.
Concerning the eventual hypnotic effect of cannabidiol, our
preliminary data with 15 subjects tend to confirm that it has such
potential use. In fact, 160 mg cannabidiol significantly increased the
number of hours the subjects slept, and all three doses decreased dream
recall, which could be the consequence of less sleep interruptions during
the night's sleep. It is significant in this respect that several healthy
volunteers complained of somnolence when under cannabidiol, a fact also
reported by four of the eight epileptic patients. It is interesting that 5
mg nitrazepam (the amount commercially available in sleeping pills in
Brazil), did not differ in its effect from placebo. Other authors have
reported clear hypnotic effects with 10 mg nitrazepam. (27,28)
A note is in order about the method used to measure hypnotic effect
of cannabidiol The use of self-reported questionnaires to evaluate
hypnotic effects is considered a subjective method, in opposition to the
objective methods such as the inspection of the volunteers in
inpatient-settings throughout the night or the obtention of continuous EEG
recordings. Both methods have enthusiastic followers. (29,30)
The clinical trial of cannabidiol as an eventual antiepileptic
agent was conducted with eight cases of severely ill patients, refractory
to all known antiepileptic drugs. The results indicate that cannabidiol
has a beneficial effect in patients suffering from secondary generalized
epilepsy with temporal focus who do not benefit from known antiepileptic
drugs. The mechanism by which cannabidiol benefited our epileptic patients
is not known. One possibility is that cannabidiol potentiates the action
of the other anticonvulsants since enhancement by cannabidiol of the
anticonvulsant activity of phenobarbital and phenytoin in animals has been
demonstrated. (18, 31,32) In man however, 50 to 500 mcg/kg cannabidiol
given in cigarette form is not able to alter plasma concentrations of
secobarbital. (33) The possibility that cannabidiol acts per se should
also be taken into consideration, as shown by several reports describing
its direct anticonvulsant effects in animals. In this respect it is
important to note that Tamir et al. (34) have reported that cannabidiol
and phenytoin both possess similar stereochemical requirements for
anticonvulsant drug action.

NEXT INDEX HOME