Content uploaded by Arno Hazekamp
Author content
All content in this area was uploaded by Arno Hazekamp on Aug 12, 2019
Content may be subject to copyright.
Author Proof
A
Evaluation of a Vaporizing Device (Volcano
1
) for the
Pulmonary Administration of Tetrahydrocannabinol
ARNO HAZEKAMP,
1
RENEE RUHAAK,
1
LINEKE ZUURMAN,
2
JOOP VAN GERVEN,
2
ROB VERPOORTE
3
1
Division of Pharmacognosy, Institute of Biology, Leiden University, Leiden, The Netherlands
2
Centre for Human Drug Research, Leiden, The Netherlands
3
Division of Pharmacognosy, Section Metabolomics, Institute of Biology, Leiden University, Leiden, The Netherlands
Received 1 January 2005; revised 15 April 2005; accepted 25 October 2005
Published online ? ???inWiley InterScience (www.interscience.wiley.com). DOI 10.1002/jps.20574
ABSTRACT: Xxx
Q2
xx. ß2005 Wiley-Liss, Inc. and the American Pharmacists Association J
Pharm Sci 9999:1– 10, 2005
Keywords: delta-9-tetrahydrocannabinol; cannabis; vaporizer; aerosol; pulmonary
drug delivery; formulation vehicle; controlled delivery
INTRODUCTION
Cannabis (Cannabis sativa L.) has a long history
as a recreational drug and as part of traditional
medicine in many cultures of the world. Nowa-
days, cannabis is used medically by patients
suffering from diseases varying from cancer and
HIV/AIDS to multiple sclerosis, frequently
in the form of unprescribed self-medication.
1,2
Marinol
1
, an oral form of the main psychoactive
constituent of cannabis, delta-9-tetrahydro-
cannabinol (THC), has been developed for some
indications. However, oral THC is notoriously
unreliable in its effects.
3
Drawbacks of Marinol
1
include its slow onset of action, large varia-
bility in bioavailability, and extensive first
pass metabolism. Moreover, there is the incon-
venience of taking oral medication in case of
nausea or vomiting. Therefore, for many patients
the demand for more effective cannabinoid-based
medications persists. For this group of pati-
ents cannabis smoking is a more convenient
method of administration, allowing self-titration
of the desired effects. However, inhalation of toxic
compounds during cannabis smoking poses a
serious hazard. This risk is not thought to be due
to cannabinoids, but rather to noxious pyrolytic
byproducts.
4,5
Consequently, the shortcomings of
smoked cannabis have been widely viewed as a
major obstacle for approval of crude cannabis as a
medicine by public health authorities.
6
Cannabis ‘‘vaporization’’ or ‘‘volatilization’’ is a
technique aimed at suppressing irritating respira-
tory toxins by heating cannabis to a temperature
where active cannabinoid vapors are formed, but
below the point of combustion where pyrolytic
toxic compounds are made. Vaporization offers
patients who use medicinal cannabis the advan-
tages of the pulmonary routes of administration,
that is: rapid delivery into the bloodstream, ease of
self-titration, and concomitant minimizing the
risk of over- and under-dosing, while avoiding
the respiratory disadvantages of smoking.
In a series of studies the vaporizing of can-
nabis samples was systematically tested to show
its advantage over smoking. When a variety of
smoking devices (including water pipes) were
compared, specifically examining THC and solid
smoke tars, it was found that only vaporizers were
capable of achieving reductions in tar relative to
THC when compared to direct smoking of canna-
bis.
7,8
A follow-up study tested a vaporizer that
was found to deliver THC while completely elimi-
JPS/05-082(20574)
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 00, XXXXX 2006 1
Correspondence to: Arno Hazekamp (Telephone: 31-71-527-
4519; Fax: 31-71-527-4511;
E-mail: ahazekamp@rocketmail.com)
Journal of Pharmaceutical Sciences, Vol. 9999, 1 –10 (2005)
ß2005 Wiley-Liss, Inc. and the American Pharmacists Association
Author Proof
A
nating three specific toxins (naphthalene, benzene,
and toluene) in the solid phase of the vapor.
9
The study also detected a 56% reduction in tars
and a qualitative reduction in carbon monoxide,
but did not test for any other chemicals.
10
In a more
recent study,
11
GC-mass-spectrometry was used to
analyze the gas phase of vaporized cannabis for
a wide range of toxins, particularly concentrating
on the highly carcinogenic polynuclear aromatic
hydrocarbons (PAHs). The vaporizer that was
used was the Volcano
1
.
12
It consists of a heater,
a ventilator, a filling chamber, a valve, and a
balloon. During operation the balloon is inflated
with hot air and cannabinoid vapors. Using
cannabis plant material as the sample, vapors
were found to consist overwhelmingly of cannabi-
noids, while the combusted control contained over
one hundred additional chemicals, including sev-
eral known PAHs.
Although a large variety of vaporizing devices is
available on the market, the Volcano is one of the
few devices that have been tested scientifically to
some extent. It is a herbal vaporizer, intended for
the vaporization of whole cannabis plant materials
(i.e., flowertops), but numerous unexplored vari-
ables could affect the efficiency and output of
vaporization. These parameters are variations in
temperature; differences in specimen density,
weight, content of water and essential oils, and
consistency of material in the filling chamber;
differences in the variety and potency of cannabis
used; and use of different preparations such as
crude flowertops, hashish, hash oil, etc. Because of
the paucity of data it has so far been difficult to
show that the Volcano vaporizer can be used as a
reliable tool for the reproducible administration of
THC or other cannabinoids. A solution to this
would be in the use of pure cannabinoid prepara-
tions of known concentration to guarantee an
exact and reproducible loading of cannabinoids.
In this study the Volcano vaporizer was eval-
uated as a novel method for the administration of
THC. Pure cannabinoid preparations were used in
order to obtain quantitative results in terms of
efficiency and reproducibility of THC delivery into
the balloon of the Volcano. By changing para-
meters such as temperature setting, type of
evaporation sample, and balloon volume, the
vaporization of THC was systematically improved
to its maximum yield, while preventing the
formation of degradation products. Factors that
resulted in loss of THC by condensation, that is,
storage time of the balloon and use of the filling
chamber, were evaluated. The inter-device repro-
ducibility of THC vaporization under optimized
conditions was determined. Finally, the results of
this study were used for the clinical administra-
tion of THC by vaporizing. The amount of exhaled
THC was determined and compared to the dose,
which was inhaled through the Volcano.
Our results indicate that the Volcano is a
convenient device for the administration of THC
by inhalation.
MATERIALS AND METHODS
Materials
All organic solvents were HPLC or analytical
grade, and were purchased from J.T. Baker
(Deventer, The Netherlands). Anthracene (min.
99% purity) was purchased from Aldrich (St.
Louis, MO). Deuteriated chloroform (CDCl
3
) was
from Eurisotop, Gif-sur-Yvette, France. Glass
fiber filters (Cambridge type, borosilicate glass,
92 mm diameter) and tightly fitting filter holders
for vapor extraction were obtained from Borg-
waldt Technik GmbH (Hamburg, Germany).
Cannabis plant material (female flowertops)
was medical grade and obtained from Bedro-
can
Q3
BV (The Netherlands). It had a water
content of about 8%, a THCA content of about
12%, and virtually no free THC.
Purified THC and THCA (purity 98%) were
produced and quantified as reported earlier.
13,14
THC was of pharmaceutical grade. The cannabi-
noids were stored as ethanolic solutions at 208C
at a concentration of 50 mg/mL.
The Volcano Device
The Volcano
1
was obtained from Storz & Bickel
GmbH & Co. (Tuttlingen, Germany) and was
used according to the manual as provided by the
manufacturer. It is a vaporizer or evaporator
that can evaporate the active substances or
aromas from plant material by using a hot air
flow (Fig. 1). Depending on the type of filling
chamber used, whole plant material or liquid
samples (e.g., aromatic oil, extract, or pure
compounds in solution) can be used. Evaporated
compounds are collected in a detachable plastic
balloon, which can be removed and fitted with
a mouthpiece for inhalation. Volume of the
balloon can be varied. Unless otherwise stated, a
balloon length of 55 cm (around 8 L) was used, as
recommended by the manufacturer. The tempera-
2HAZEKAMP ET AL.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 00, XXXXX 2006
Author Proof
A
ture control ranges from setting 1 –9, correspond-
ing to temperatures of 130–2268C (see Tab. 1).
Before each new set of experiments the whole
device was thoroughly cleaned with ethanol.
At the start of each evaporation the Volcano
was preheated until the indicator light showed
that the target temperature was reached. The
balloon, connected to the filling chamber, was
then immediately placed onto the Volcano and the
ventilation was started. When the balloon was
completely inflated, ventilation was stopped and
the content of the balloon was processed for
analysis within 5 min, unless stated otherwise.
All laboratory experiments were carried out in a
standard laboratory fume hood under constant
ventilation with an ambient room temperature of
about 228C and a humidity of 40 –60%. The air was
not conditioned (e.g., by HEPA filters).
Use of the Liquid Pad
The pure cannabinoids THC or THCA were used
as ethanolic solutions. For these liquid samples an
adapted filling chamber was used, containing a
removable disc made of tightly packed stainless
steel wire mesh (liquid pad), obtained from the
manufacturer of the Volcano. For each experi-
ment the appropriate amount of the cannabinoid
was dissolved in a final volume of 200 ml of ethanol
for application onto the liquid pad and ethanol
was allowed to evaporate for 10 min under
ambient conditions. A new liquid pad was used
for each experiment.
Extraction of THC from the Vapor and the Liquid Pad
Cannabinoids were recovered from the vapor
phase inside the balloon by condensation onto
glass fiber filters, designed to capture particles
>0.1 microns. Vapor was slowly aspired through
the glass-fiber filter which was then extracted
twice with 15 mL of methanol/chloroform (9:1, v/v)
under ultrasonication. After evaporating the
extraction solvent, samples were reconstituted
in 5 mL of ethanol for analysis by HPLC or NMR.
These ethanolic samples will be further referred
to as vapor extracts.
Residual THC on the liquid pad was recovered
by extracting the liquid pad twice using methanol/
chloroform (9:1, v/v) under ultrasonication. Extra-
cts were further handled as described above for the
vapor extracts. Recovery was determined by spik-
ing filters or liquid pads with THC (2 mg) and
performing the described extraction procedure.
To assess the efficiency of condensation of
cannabinoids onto the glass fiber filter, a wash
bottle filled with ethanol was placed after the filter.
The escaping gases were bubbled through this
liquid which was thereafter analyzed by HPLC to
measure untrapped cannabinoids.
Quantitative
1
H-Nuclear Magnetic
Resonance Spectroscopy (NMR)
Quantification of THC in the extracts was done by
quantitative
1
H-NMR using a Bruker 300 MHz
NMR apparatus as described by Hazekamp et al.
14
Table 1. Temperature (8C) of the Heating Unit of the
Volcano Corresponding to the Different Temperature
Settings
Temperature Setting Temperature in 8C
1 130
3 154
5 178
7 202
9 226
Figure 1. The Volcano vaporizer.
EVALUATION
Q1
OF A VAPORIZING DEVICE 3
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 00, XXXXX 2006
Author Proof
A
In short, an exact volume of the sample was mixed
with 1.0 mg of anthracene as internal standard
for quantification. The sample was then evapo-
rated to dryness under vacuum and reconstituted
in chloroform (deuteriated) for
1
H-NMR analysis.
High Pressure Liquid Chromatography (HPLC)
HPLC was used for both qualitative and quanti-
tative analysis of the obtained extracts. The
HPLC profiles were acquired on a Waters (Mil-
ford, MA) HPLC system consisting of a 626 pump,
a 717 plus autosampler, and a 2996 diode array
detector (DAD), controlled by Waters Millennium
3.2 software. Full spectra were recorded in the
range of 200–400 nm. The analytical column was
a Vydac (Hesperia, CA) C
18
, type 218MS54
(4.6 250 mm, 5 mm), with a Waters Bondapak
C
18
(2 20 mm, 50 mm) guard column. The mobile
phase consisted of a mixture of methanol-water
containing 25 mM of formic acid in gradient mode;
methanol:water in ratios from 65:35 to 100:0 over
25 min, then isocratic to 28 min. The column was
reequilibrated under initial conditions for 4 min.
Flow-rate was 1.5 mL/min and total runtime was
32 min. All determinations were carried out
at ambient temperature. The main neutral and
acidic cannabinoids were well separated with
this method.
15
Analyzed concentrations were
well above the limit of quantification of the used
method.
Evaluation of Temperature Control
Temperature control was evaluated at setting 1,
3, 5, 7, and 9 (see Tab. 1). Time needed to reach
target temperature, and accuracy and stability of
target temperature were determined using an
electronic thermometer (response time; 250 ms).
Temperature was measured in the middle of the
filling chamber, on top of the liquid pad, and each
measurement was started directly after the
indicator light of the heater had switched off.
Inter-device variability for the same parameters
was tested for four different Volcano devices. All
experiments were repeated three times.
Optimization of Vaporizing Parameters
(a) Temperature: Cannabis plant material, and
pure cannabinoids THCA and THC were
vaporized at temperature settings 1, 3, 5, 7,
and 9 in order to determine the delivery
into the balloon as well as the formation of
degradation products. Vapor extracts were
qualitatively analyzed by HPLC for detec-
tion of degradation products, while quanti-
tative analysis by NMR was used for
determination of delivery.
(b) Heating time: In order to determine the
minimal time that is needed to reach
maximal evaporation of THC, the following
experiment was performed: THC (2 mg)
was applied onto the liquid pad and the
ventilation was activated for a duration
ranging from 10 to 300 s, without balloon
attached to the device so THC could
evaporate freely. Subsequently, residual
THC was extracted from the liquid pads
and extracts were quantitatively analyzed
by NMR.
Relationship Between Loaded Dose and Delivery
The relationship between quantity of THC loaded
onto the filling chamber and delivery into the
balloon was determined in the range of 2–8mgof
THC. Vapor extracts were analyzed by NMR and
HPLC, and each experiment was performed
threefold.
Inter-Device Variability
Using the optimized parameters as determined in
this study, four Volcano devices were finally
evaluated for inter-device variability in THC
delivery. Samples of 4 mg of THC were used for
vaporizing and each Volcano was tested on five
occasions. Vapor extracts were analyzed by NMR.
Condensation of THC onto the
Balloon and Filling Chamber
The effect of storage time of the balloons on
condensation of THC was determined by storage
of the balloon at room temperature for a duration
of up to 180 min after vaporizing (2 mg THC). The
vapor extract was then collected for analysis.
Each experiment was performed threefold.
Throughout this study balloons were always
processed within 5 min after vaporizing. There-
fore, it was determined more exactly how much
THC was lost due to condensation onto the walls
of the balloon after 5 min of storage by carefully
4HAZEKAMP ET AL.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 00, XXXXX 2006
Author Proof
A
cutting the balloon (n¼5) into pieces and extract-
ing twice with ethanol under ultrasonication.
In order to determine the amount of THC that
condensated onto the filling chamber (excluding
liquid pad) and valve, after some experiments
these parts were extracted twice with ethanol
under ultrasonication. Finally, extracts were con-
centrated and THC was quantified by NMR.
Clinical Application of the Volcano
At the Centre for Human Drug Research (CHDR,
Leiden, The Netherlands) a methodology study
was performed to study the effects of THC
administration using the Volcano vaporizer. The
study was approved by the Medical Ethical
Committee of Leiden University, The Nether-
lands. Preliminary results of this study were
published recently,
16
and full results will be
published in the near future. In short, during
two separate occasions subjects received a rising
dose of 2, 4, 6, and 8 mg THC (loading dose in
filling chamber) or placebo (ethanol only) admi-
nistered via the Volcano, using the optimized
parameters as determined in this study. Admin-
istrations were given with 1.5 h intervals. The
balloon (8 L) had to be inhaled through the mouth
within 3 min and breath was held for 10 s after
each inhalation. Following each inhalation, sub-
jects were asked to exhale through a filter of the
same type as used for vapor extraction. Filters
were subsequently extracted as mentioned before
and the quantity of exhaled THC was determined
by NMR. Because of time restraints, no further
evaluation of lung function (e.g., FEV1) could be
performed.
RESULTS
Trapping and Recovery of THC for Analysis
Since no trace of THC could be found in the ethanol
fraction of the wash bottle inserted after the filter,
it was concluded that THC was completely trapped
onto the used type of filter. Recovery of THC was
found to be 99.3 (1.1)% from the filter and 83.0
(2.5)% from the liquid pad. All measurements
were corrected for these values.
Accuracy of the Temperature Setting
At all tested temperature settings it was found
that temperature reached a first plateau after
about 30 s. After that temperatures remained
relatively stable for some time, but kept below
accepted limits (target temperature 48C, as
claimed by the manufacturer) for all tested
settings. Results can be seen in Figure 2a. How-
ever, after about 45–60 s, depending on the
setting, the heating element was activated again
by the temperature sensor, and about 20 s later
temperatures increased by a few degrees bringing
the temperature within specified limits. It must
be concluded that the liquid pad and the filling
chamber need some time to heat up to the target
temperature.
Reproducibility of the Vaporizer
When four different Volcano devices were eval-
uated under equal conditions to evaluate inter-
device variability (Fig. 2b), some small differences
Figure 2. (a): Temperature profile over time of the
Volcano at different settings. Dotted lines indicate
target temperatures at settings 1, 3, 5, and 7. (b):
Comparison of temperature profile of four different
Volcano devices at setting 9. Dotted lines indicate
allowed target temperature range (48C). Data is shown
as mean values of three experiments, and errorbars
indicate standard deviation.
EVALUATION
Q1
OF A VAPORIZING DEVICE 5
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 00, XXXXX 2006
Author Proof
A
in heating profile were found. Only temperature
setting 9 was evaluated here after it had been
shown to be the optimal temperature for THC
delivery. Although two devices reached target
temperature (accepted variation 48C) already
after 30 s, the two others needed 60 s or more to do
so. For two devices the temperature increased
above the maximum limit of target temperature
in the 90 s duration of our experiment. In
conclusion each individual Volcano device shows
little variability during sequential uses (intra-
device variability), although small differences do
exist between different devices (inter-device
variability).
Optimizing of Vaporizing Parameters
with Different Substrates
THCA: Under the influence of heat THCA can be
converted into THC by decarboxylation. Indeed,
when THCA was used it was observed that this
conversion increased with temperature and max-
imum delivery of THC was about 33% at the
highest temperature setting (Fig. 3). However,
conversion was not complete and THCA was
present in the vapor extracts at a level of about
5.5 (1.3)% relative to THC.
Crude flower tops: The use of plant material
(200 mg at 12% THCA) resulted in a maximum
THC delivery of only 29% (Fig. 3). In fresh
cannabis plant materials THC is present in the
form of its acidic precursor THCA and the use of
plant material resulted in an incomplete decar-
boxylation with about 3.8% residual THCA pre-
sent in the vapor. Besides THC, several other
cannabinoids as well as a range of other plant
components were detected. Therefore, the use of
cannabis plant material in the Volcano should not
be recommended for the administration and study
of THC alone.
Pure THC: Evaporation of THC was shown to
increase with temperature with a maximal deliv-
ery of about 53% at setting 9 (Fig. 3), while no
degradation products (delta-8-THC (D
8
-THC),
cannabinol (CBN), or other unknown peaks in
the HPLC-chromatogram) were observed at any
setting. Therefore, using the Volcano device, it was
concluded that the highest delivery yield was
achieved with an ethanolic of pure THC. When
liquid pads were extracted after vaporizing it
showed a very low amount of residual THC,
indicating a very high yield of evaporation, at the
highest temperature setting. This strongly sug-
gests that nondelivered THC does not remain on
the liquid pad, but is probably lost by condensation
after initial evaporation.
Minimum time was determined for the maximal
evaporation of THC from the liquid pad by
measuring residual THC after vaporizing.
Figure 4 shows that the amount of residual THC
rapidly decreases between 20 and 40 s after
starting of the vaporizing. This corresponds with
the observation that in the same time-period the
(near) target temperature of the Volcano is
reached (Fig. 2a and b). After 45 s most of the
Figure 3. Amount of delta-9-tetrahydrocannabinol
(THC) delivered into the balloon after using different
sample types in the filling chamber: THC (~, 8 mg);
THCA (&, 9 mg); plant material (^, 200 mg) at different
temperature settings. Delivery is expressed as %
relative to the loaded dose of THC. For THCA and plant
material, the theoretical loaded dose of THC was
calculated based on a 100% conversion of THCA into
THC. Data is shown as mean values. Error bars indicate
standard deviation.
Figure 4. Residual THC on liquid pad after varying
vaporizing time at setting 9. Datais shown as mean values
of three experiments, and error bars indicate standard
deviation. Values were corrected for the maximum
recovery of 83% for extraction of the liquid pads.
6HAZEKAMP ET AL.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 00, XXXXX 2006
Author Proof
A
THC is evaporated and just a small fraction of THC
can be found in the liquid pad extract, indicating
that vaporizing time should be at least 45 s. When
using a temperature setting of 9 with a balloon
volume of 4 L (filling time around 30 s), a low THC
delivery (only 30% for 8 mg of THC) with a high
dose variability (relative SD 22%) was observed
indicating that the maximum delivery yield was
not yet reached.
It was observed that the maximal evaporation of
THC is reached after 120 s, (Fig. 4). Since the
Volcano is blowing air at a constant rate of about 9
L per min, this corresponds to a balloon volume of
about 18 L. However, by empirical testing in our
laboratory (data not shown) it was found that a
maximum volume of about 8 L could be inhaled
within 3 min when following the protocol of the
clinical trial. Therefore, a balloon volume of 8 L
(filling time of about 55 s) was selected for further
study.
Relationship between Loaded Dose and
Delivery under Optimal Conditions
With a Volcano operating under the aforemen-
tioned optimized conditions (temperature setting
9, balloon volume 8 L) the delivery was deter-
mined with an increasing amount of THC ranging
from 2 to 8 mg. It is shown in Figure 5 that the
delivery was proportional to the loaded dose of
THC; a linear curve was obtained with a regres-
sion coefficient (R
2
-value) of 0.99. From the slope
of the line, a mean delivery yield of 57.8 (6.9)%
could be calculated.
Four available devices were then tested under
conditions as mentioned above using a sample of
4 mg of THC. Differences in delivery between the
Volcano devices were relatively small. Average
delivery of all four Volcanos was 53.9 (8.1)%, and
this value was taken as the average delivery for
further considerations.
Condensation onto Balloon and Filling Chambers
Loss of THC during experiments could partially
be accounted for by incomplete evaporation and
condensation onto parts of the Volcano vaporizer.
Prolonged storage of the balloon at room tem-
perature after vaporizing led to a steadily increas-
ing loss of THC by condensation up to the point
that after 180 min almost no THC could be
detected anymore in vapor extracts (Fig. 6).
However, if the balloon was extracted within
5 min after vaporizing, less than 2% of the total
dose was recovered from the inner surface of the
balloon. Condensation of THC onto the other
parts of the Volcano setup was found to be of
significant importance. Visual inspection of the
filling chamber shows the presence of a conden-
sate mostly on the inside of the filling chamber
just above the liquid pad. Extraction of the filling
chamber together with the valve, but excluding
the liquid pad, showed that an average of 23.6
(14.1)% of the loaded THC had condensated onto
Figure 5. Relationship between delivery of THC
under optimized setting conditions with four different
THC loading doses ranging from 2 to 8 mg. Data are
shown as mean values of three experiments and error
bars indicate standard deviation. Linearity (r
2
-value)
was 0.99, as determined by linear regression.
Figure 6. Relationship between storage time and
percentage of initial THC that could be recovered from
the balloon. Data are shown as mean values of three
experiments, expressed as % of initially recovered THC.
Error bars indicate standard deviation. During this
study all balloons were processed within 5 min after
evaporation, which is indicated by the dotted line.
EVALUATION
Q1
OF A VAPORIZING DEVICE 7
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 00, XXXXX 2006
Author Proof
A
these parts of the Volcano, and could therefore
account for a large part of the nondelivered THC.
Clinical Study and Loss by Exhalation
The clinical trial was finished without any serious
complaints by the test subjects. Some mild
complaints included irritation of the throat and
lungs, and coughing. However, these effects were
also observed during inhalation of placebo and
therefore could be an effect of residual ethanol.
The development of significant physiologic chan-
ges after inhalation of vaporized THC indicates
that THC can be effectively administered by this
route.
Interestingly, it was shown that a large propor-
tion of inhaled THC was not absorbed by the lungs.
The total amount of THC used for evaporation was
20 mg of THC for each subject (Rising dose of 2, 4, 6,
and 8 mg resulting in a total dose of 20 mg). Taking
into account the average delivery yield of 53.9% as
found in this study, only an average of 10.8 mg of
THC was totally inhaled from the balloon. The
amount of THC recovered from exhaled breath
ranged from 2.5 to 4.4 mg, which means that up to
30%–40% of inhaled THC was not absorbed by the
lungs. The variability of THC in exhaled breath
(relative SD 5.4%) is comparable to the varia-
bility in delivery of THC by the Volcano. Taking
this into account it could be concluded that
absorption of THC by the lungs is probably very
similar between different subjects.
DISCUSSION AND CONCLUSION
The Volcano
1
vaporizer was validated for the
efficient and reproducible delivery of delta-9-THC,
and was found to be able to deliver an average
amount of about 54% of applied dose of THC into
the balloon for inhalation. THC recoveries from
smoke was found to range from 34% to 69% in a
variety of studies using different types of smoking
procedures.
19–21
Because the plant material is not
burned in the Volcano, no significant harmful
cancer causing combustion products are expected
and the noxious intake, when compared to smok-
ing, is greatly reduced.
10,11
Using the Volcano
device for pulmonary administration of THC, a
delivery is reached that is comparable to smoking,
but without the presence of degradation products
or harmful byproducts in significant amounts.
Loading the Volcano with Cannabis plant
material or pure THCA resulted in a residual
amount of THCA in the vapor in the order of 5%
relative to THC. Not much is known about
biological effects or metabolism of THCA, and
therefore the use of THCA as sample for intended
clinical administration of pure THC should be
avoided. Older studies at least indicate that
THCA is not psychoactive in monkeys.
22
Altho-
ugh in our study cannabis plant material was used
only for comparative reasons, it is clear that a
variety of cannabinoids and other compounds such
as terpenoids are present in the vapor.
With pure THC as the loading sample, tem-
perature setting and balloon volume were opti-
mized for a maximal, reproducible delivery of THC
without formation of detectable amounts of degra-
dation products. Using the highest temperature
setting together with a balloon volume of 8 L was
found to yield optimal results. Balloon volumes
over 8 L were not tested because of restraints in the
clinical trial protocol. The target temperature of
the Volcano was found to be not completely
accurate and stable. Possibly this is a contributing
factor to the relative variability in the delivery of
THC, which was about 15% at setting 9. However,
this is reasonable when compared to the varia-
bility that has been previously found in smoking
studies of cannabis plant material.
19
Accuracy of
temperature control therefore does not seem to be
of crucial importance under these conditions,
although a more accurate temperature control
might result in an even lower variability in THC
delivery.
In the range of 2–8 mg of THC, the delivery was
found to be linear with the amount of THC used.
Prolonged storage of the balloon before inhalation
resulted in an increasing loss of THC by condensa-
tion inside the balloon and after 3 h almost no THC
could be recovered from the vapor in the balloon.
However, if the content was extracted within 5 min
after vaporization not more than 2% of THC
present was lost. Vaporized THC was visible inside
the balloon as a thin gray mist which was absent in
placebo balloons, so during the clinical trial
balloons had to be blinded with a black plastic
cover.
During the clinical administration, it was found
that about 35% of total THC was exhaled directly
after inhalation and was therefore not absorbed by
the lungs. When the efficiency of delivery during
vaporizing and incomplete absorption by the lungs
is considered, the final administered dose equaled
about 6–8 mg of THC of the total amount of 20 mg
loaded. The subjective effect upon the subjects
seemed to be in accordance with such a dose as
8HAZEKAMP ET AL.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 00, XXXXX 2006
Author Proof
A
described in other papers.
17,18
So it seems that a
final uptake of 30–40% was reached (relative to
loaded amount of THC), which is comparable to the
efficiency reached by smoking of cannabis.
It has been shown that the administration of
THC by aerosol is capable of producing the full
constellation of cannabinoid effects in mice. These
effects were CB1-receptor mediated, as shown by
the use of selective CB1 antagonists,
23
which
confirms that the pulmonary administration
of cannabinoids certainly has a clinical potential.
Several studies have been performed using an
aerosol for the administration of THC.
23–26
But
because cannabinoids are almost completely inso-
luble in water this requires the use of solubilizers
that are to be inhaled together with THC, which
frequently results in irritation of the lungs and
coughing. Moreover, part of an administered
aerosol can be swallowed and thereby adminis-
tered orally, complicating the effect, kinetics, and
metabolism of the administered compound. This
has already been shown for aerosol administration
of radiolabeled isoproterenol.
27
Using the Volcano vaporizer seems to eliminate
at least part of the problems associated with the
use of an aerosol for the delivery of THC or other
cannabinoids. It is likely that the Volcano also
produces an aerosol, that is, droplets of various
sizes in a gas phase made up of vapor and air.
However, in an artificial lung model the majority of
vaporized THC could reach the deepest compart-
ment (personal communication with Volcano
manufacturer) indicating that the exhaust blown
from the Volcano consists for a large part of very
fine droplets and vapor. Nonetheless, the composi-
tion of an aerosol is partially dependent on the
ambient conditions such as humidity and presence
of nuclei for condensation. So although our results
were found to be reproducible with a relatively low
variability, these factors must be taken into
consideration for further development of the
Volcano.
What is currently needed for optimal use of
medicinal cannabinoids is a feasible, nonsmoked,
rapid-onset delivery system. With the Volcano a
safe and effective cannabinoid delivery system
seems to be available to patients. Although our
current study has concentrated on the delivery of
THC, it should be noted that other cannabinoids
might also have a role to play for some indications.
In several medical studies, the effect of THC or
dronabinol alone could not match the effect of a
total cannabis preparation, indicating there might
be other active cannabinoids needed for a full
range of effects.
28
As an example, a combination of
THC with CBD is now under clinical investigation
for the treatment of chronic pain conditions.
29
The
next step in the evaluation of the Volcano vapor-
izer should therefore include the study of mixtures
of pure cannabinoids.
ACKNOWLEDGMENTS
The authors thank the manufacturer of the
Volcano vaporizer, Storz & Bickel GmbH & Co.,
for providing the department of Pharmacognosy
with the Volcano devices for our study. Bedrocan
BV (The Netherlands) is acknowledged for pro-
viding us with medical grade cannabis plant
materials. Farmalyse BV (Zaandam, The Nether-
lands) was involved in the development of the
procedure to produce clinical grade cannabinoid
samples of THC and THCA.
REFERENCES
1. Page SA, Verhoef MJ, Stebbins RA, Metz LM, Levy
JC. 2003. Cannabis use as described by people with
multiple sclerosis. Can J Neurol Sci 30:201–205.
2. Furler MD, Einarson TR, Millson M, Walmsley S,
Bendayan R. 2004. Medicinal and recreational
marijuana use by patients infected with HIV. AIDS
Patient Care STDS 18:215–228.
3. Grinspoon L. 1997. Marijuana, the forbidden
medicine. Revised edition. New Haven, CT: Yale
University Press.
4. Hiller FC, Wilson FJJ, Mazumder MK, Wilson JD,
Bone RC. 1984. Concentration and particle size
distribution in smoke from marijuana cigarettes
with different D9-tetrahydrocannabinol content.
Fundam Appl Toxicol 4:451–454.
5. Matthias P, Tashkin DP, Marques-Magallanes JA,
Wilkins JN, Simmons MS. 1997. Effects of varying
marijuana potency on deposition of tar and delta9-
THC in the lung during smoking. Pharmacol
Biochem Behav 58:1145–1150.
6. Institute of Medicine. 1999. Marijuana and medi-
cine: Assessing the scientific base. Washington DC,
MD: National Academy Press.
7. Gieringer D. 1996. Marijuana research: Waterpipe
study. MAPS (Multidisciplinary Association for
Psychedelic Studies) Bull 6:59–66.
8. McPartland JM, Pruitt PL. 1997. Medical mar-
ijuana and its use by the immunocompromised.
Altern Ther Health Med 3:39–45.
9. Chemic Laboratories. 2000. Proof of concept:
Release of chemical constituents in cannabis sativa
at 170–1858C versus combustion. Unpublished
report to California NORML and MAPS, Nov
17th, 2000.
EVALUATION
Q1
OF A VAPORIZING DEVICE 9
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 00, XXXXX 2006
Author Proof
A
10. Gieringer D. 2001. Cannabis vaporization: A
promising strategy for smoke harm reduction.
J Cannabis Ther 1:153–170.
11. Gieringer D, StLaurent J, Goodrich S. 2004.
Cannabis vaporizer combines efficient delivery of
THC with effective suppression of pyrolytic com-
pounds. J Cannabis Ther 4:7–27.
12. http://www.storz-bickel.de, website visited 10 January
2005.
13. Hazekamp A, Simons R, Peltenburg-Looman A,
Sengers M, van Zweden R, Verpoorte R. 2004a.
Preparative isolation of cannabinoids from Canna-
bis sativa by centrifugal partition chromatography.
J Liq Chrom Rel Technol 27:2421–2439.
14. Hazekamp A, Choi YH, Verpoorte R. 2004b.
Quantitative analysis of cannabinoids from Canna-
bis sative using
1
H-NMR. Chem Pharm Bull 52:
718–721.
15. Hazekamp A, Giroud C, Peltenburg A, Verpoorte R.
2005. Spectroscopic and chromatographic data of
cannabinoids from Cannabis sativa. J Liq Chrom
Rel Technol 28:2361–2382.
16. Zuurman L, Roy C, Hazekamp A, Schoemaker R,
den Hartigh J, Bender JCME, Pinquier JL, Cohen
AF, van Gerven JMA. 2005. Effect of THC admin-
istration in humans: Methodology study for further
pharmacodynamic studies with cannabinoid
agonist or antagonist. Br J Clin Pharmacol 59:
625.
17. Leweke FM. 2002. Acute effects of cannabis and the
cannabinoids. In: Grotenhermen F, Russo E,
editors. Cannabis and cannabinoids. New York,
NY: Haworth Press. pp 249–256.
18. Abood ME, Martin BR. 1992. Neurobiology of
marijuana abuse. Trends Pharmacol Sci 13:201–
206.
19. Fehr KO, Kalant H. 1972. Analysis of cannabis
smoke obtained under different combustion condi-
tions. Can J Physiol Pharmacol 50:761–767.
20. Manno JE, Kiplinger GF, Haine SE, Bennett IF,
Forney RB. 1970. Comparative effects of smoking
marihuana or placebo on human motor perfor-
mance. Clin Pharmacol Ther 11:808–815.
21. Davis KH. 1984. Some smoking characteristics of
marijuana cigarettes. In: Agurell S, Dewey WL,
Willette RE, editors. The cannabinoids: Chemical,
pharmacologic and therapeutic aspects. New York,
NY: Academic Press.
22. Edery H, Grunfeld Y, Porath G, Ben-Zvi Z, Shani A,
Mechoulam R. 1972. Structure-activity relation-
ships in the tetrahydrocannabinol series. Modifica-
tions on the aromatic ring and it the side-chain.
Arzneimittelforschung 22:1995–2003.
23. Wilson DM, Peart J, Martin BR, Bridgen DT, Byron
PR, Lichtman AH. 2002. Physiochemical and
pharmacological characterization of a D
9
-THC
aerosol generated by a metered dose inhaler. Drug
Alcohol Depend 67:259–267.
24. Naef M, Russmann S, Petersen-Felix S, Brenneisen
R. 2004. Development and pharmacokinetic char-
acterization of pulmonal and intravenous delta-9-
tetrahydrocannabinol (THC) in humans. J Pharm
Sci 93:1176–1184.
25. Lichtman AH, Peart J, Poklis JL, Bridgen DT,
Razdan RK, Wilson DM, Poklis A, Meng Y, Byron
PR, Martin BR. 2000. Pharmacological evaluation
of aerosolized cannabinoids in mice. Eur J Phar-
macol 399:141–149.
26. Hartley JP, Nogrady SG, Seaton A. 1978. Bronch-
odilator effect of delta1-tetrahydrocannabinol. Br J
Clin Pharmacol 5:523–525.
27. Lyons HA, Ayres SM, Dworetzky M, Failliers CS,
Harris MC, Dollery CT, Gandevia B. 1973. Sympo-
sium on isoproterenol therapy in asthma. Ann
Allergy 31:1–44.
28. Williamson EM, Evans FJ. 2000. Cannabinoids in
clinical practice. Drugs 60:1303–1314.
29. Notcutt W, Price M, Miller R, Newport S, Phillips
C, Simmons S, Sansom C. 2004. Initial experiences
with medicinal extracts of cannabis for chronic
pain: Results from 34 ‘Nof1’studies. Anaesthesia
59:440–452.
Q1: Please check suggested short title.
Q2: Please provide abstract.
Q3: Please provide complete location of manufacturer.
10 HAZEKAMP ET AL.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 00, XXXXX 2006
