Characterization of the absorption phase .. of marijuana smolung /-
1
Rapid blood collection, a paced smoking protocol and timely collection of physiologic and behavioral measures were used to characterize the absorption phase of marijuana smoking. Six healthy males smoked a single marijuana cigarette (placebo, 1.75%, or 3.55% A-9-tetrahydrocannabinol) in a doubleblind, randomized, Latin square study design. Rapid blood sampling with a continuous withdrawal pump allowed simultaneous collection with concurrent physiologic and behavioral measures. Mean plasma levels of 7.0 and 18.1 nglml A-9-tetrahydrocannabinol were observed after the first inhalation of a 1.75% and 3.55% A-9-tetrahydrocannabinol cigarette, respectively. Blood levels increased rapidly and peaked at 9 minutes, before initiation of the last puff sequence at 9.8 minutes. Three of six subjects reported increases in drug "liking" scores after the first puff, and all subjects responded by the second puff of a high dose cigarette. Significant increases in heart rate and diastolic blood pressure occurred shortly after peak blood levels. Previous studies have indicated that there is a substantial time delay between peak plasma levels of A-9-tetrahydrocannabinol and drug-induced effects. This study showed that behavioral and physiologic effects appear concurrently or within minutes after the rapid appearance of A-9tetrahydrocannabinol in blood during marijuana smoking. (CLINPHARMACOL THERl992;52:3 1-41.)
Marilyn A. Huestis, MS, Angela H. Sampson, RN, Barbara J. Holicky, RN, Jack E. Henningfield, PhD, and Edward J. Cone, PhD Baltimore, Md. More than one third of the population of the United States have used marijuana one or more times in their lifetimes; 10% have used it within the past year.' Although use of marijuana among American youth continues to decline, annual use in 1990 was greater than .~ is the preferred 25% among young a d u ~ t sInhalation route of exposure, although oral use is not uncommon. Drug delivery during marijuana smoking is characterized by rapid absorption of A-9-tetrahydrocannabinol, the major psychoactive component of marijuana, with peak levels paralleling those found afIntense pleasurable ter intravenous admini~tration.~.~ and strongly reinforcing effects may be produced by way of both routes because of almost immediate drug exposure to the central nervous system.' Nicotine and other inhaled drugs have been reported to distribute quickly from the lungs to the brain.6 Estimates of 11 seconds or less were made from the start of a cigarette puff to the delivery of nicotine to the central nervous
From the Addiction Research Center, National Institute on Drug Abuse. Received for publication Dec. 20, 1991; accepted March 24, 1992. Reprint requests: Edward J. Cone, PhD, Addiction Research Center, NIDA, P.O. Box 5180, Baltimore, MD 21224. 13/1/38106
system. Less than 19 seconds was required for complete drug transit through the brain. Rapid absorption of A-9-tetrahydrocannabinol during marijuana smoking also has been documented.' Quantitation of possible behavioral and physiologic effects during the early stages of marijuana smoking have been limited by the rapidity of drug absorption by way of this route of administration. A second impediment to the investigation of the absorption of A-9-tetrahydrocannabinol is the inability to deliver a precise dose. Only 20% to 37% of active drug is delivered in the mainstream smoke.' Sidestream smoke losses of 40% to 50% and pyrolytic destruction of 23% to 30% of the active drug have been documented. Intrasubject and intersubject variability in smoking dynamics contribute to the uncertainty in dose delivery. The number, duration and spacing of puffs, hold time, and inhalation volume greatly influ. ~ alternative apence the degree of drug e x p o ~ u r e An proach to accurate dosing is the determination of blood levels of A-9-tetrahydrocannabinol. Measurement during the absorption phase is a difficult task because of the need for fast sequential blood sampling. Also, sampling methods should not conflict with the collection of physiologic, subjective, or performance measures. These constraints have led other investiga-
31
CLIN PHAKMACOL 1'HF.K JULY 1992
32 Huestis et al. Table I. Subject characteristics Subject
Age (yr)
B 36 C 33 E 29 F 28 G 33 H 29 Mean -+ SD 31.3 t 3.1
Age at jrst Height (inches) Weight (kg) marijuana use (yr) Average use* Drug use previous 30 daysf 68.0 71.3 72.5 64.8 67.0 68.8 68.7 5 2.8
70.9 80.9 86.4 68.6 65.0 93.6 77.6 ? 11.2
16 14 14 15 15 13 14.5 i 1.0
0.4 7.9 0.8 1 .O 0.5 3.0 2.3
5
NAH,M N,E, M N,E,C,M N,E,H,C, M N,E,H, C N,E,H,C,B,M 2.9
*Average number of mar~juanacigarettes smoked each week. t N , Nicotine; E, ethanol; H , hcroin: M , marijuana; C, cocaine; B , barbiturates
tors to collect blood samples and behavioral data in separate study sessions or to eliminate drug measurements entirely. This study was designed to characterize the absorption of A-9-tetrahydrocannabinol during marijuana smoking and to define the onset, peak, and duration of the pharmacodynamic effects of marijuana. A continuous withdrawal pump with adjustable speeds was used for blood sampling. The use of the fast sequential sampling technique did not compromise subjects' attention during completion of experimental tests. Subjective, performance, and physiologic measures were collected concurrently with blood levels.
'".' '
METHODS Subjects. Six healthy male volunteers with histories of marijuana use were recruited from the local population through advertisements and personal communication. All subjects resided on a closed research ward of the Addiction Research Center, National Institute on Drug Abuse, for 4 to 6 weeks. Subjects provided informed consent and were paid for their participation. Medical histories, physical examinations, and electrocardiograms were reviewed to determine if active infectious disease or underlying cardiac, seizure, or renal disorder precluded the inclusion of any subject in the study. Any ongoing psychiatric illness that distorted the subjects' perception of reality or that interfered with their ability to function on a daily basis was also a basis for study exclusion. Subjects were not accepted for the study if they were currently receiving any prescribed medication. Additional evaluation criteria included blood chemistry and hematology screens, Shipley Institute Living Scale and SCL-90-R Mood Scale evaluations (Hopkins Symptom Checklist 90), and descriptions of current and previous drug use. Subjects were required to have a raw score on the vocabulary section of the Shipley Institute Living
Scale of at least 18 to be accepted in the study. In addition, the general severity index score (T score) on the SCL-90-R had to be less than 63. Light to moderate previous marijuana use was required, but urine drug screens for cannabinoid metabolites had to be negative before admission. Urine drug testing for amphetamines, barbiturates, benzodiazepines, cocaine metabolite, methadone, opiates, and phencyclidine was performed before admission and during residence to ensure lack of exposure to unauthorized drugs during the study. Characteristics of the subjects are detailed in Table I. Dose and route of administration. Marijuana cigarettes were obtained from the Research Technology Branch, National Institute on Drug Abuse. Subjects smoked a single marijuana cigarette (placebo or 1.75% or 3.55% A-9-tetrahydrocannabinol) according to a computer-controlled paced smoking procedure on each of 3 test days. The smoking protocol consisted of a 2-second inhalation, a 10-second hold period, and a 72-second exhalation and rest period. A total of eight puffs were inhaled in 11.2 minutes. Doses were administered according to a randomized, double-blind Latin square design at weekly intervals. Collection, preparation, and analysis of plasma samples. The Dakmed ambulatory withdrawal pump model ML6-5S3R (Dakmed, Inc., Buffalo, N.Y .) was used for rapid blood sampling and integrated blood collections from 1 hour before until 3 hours after drug administration. Blood was collected through 4 feet of Dakmed silanized tubing by way of a 19-gauge butterfly needle inserted in a nondominant forearm vein. Subjects received 5000 units of sodium heparin before drug administration to prevent blood clotting in the tubing. The clotting times of all subjects had returned to normal within 4 hours after heparin administration. The pump was calibrated for collection of 5 ml blood samples in 1, 5, or 25 minutes. The pump was
VOLUME 52 NUMBER 1
Absovption ofsmoked marijuana 3 3
Table 11. Marijuana study time line Time
Blood samples (5 ml)
-60 to 0 minutes
Angiocatheter and butterfly placement Heparin administration Continuous withdrawal, 0.2 mlimin (X2) Continuous withdrawal, 5 mlimin (X 10)
0 to I1 minutes
11 to 20 minutes
Intermittent sampling, 5 mlimin (X3)
20 to 50 minutes
Integrated sampling, 1 mlimin ( X 3 )
50 to 180 minutes
Integrated blood sampling 0.2 mlirnin (X 5)
180 minutes to 7 days
Disconnect withdrawal Pump Sampling through angiocatheter ( X 12)
Data collection Baseline measures of all parameters (X 2)
Computer directed smoking of one marijuana cigarette (075, 1.75%, or 3.55% THC) Continuous ECG; Pulse, blood pressure, and skin temperature every 3 minutes (X4) Alternating critical flicker fusion and pupil measures with each puff (X4) Three visual analog scales during each puff (X8) Continuous ECG Pulse, blood pressure, and skin temperature every 3 rninutes (X3) Alternating critical flicker fusion and pupil measures (X3) Three visual analog scales ( x 4 ) Continuous ECG Pulse, blood pressure, and skin temperature every 5 rninutes ( x 6 ) Alternating critical flicker fusion and pupil measures ( x 3) Three visual analog scales ( x 4 ) First administrations of performance battery (6 tasks, X2) First administration of ARCI and SDQ ( X I) Pulse, blood pressure, and skin temperature every 15 minutes (X 1 I) Alternating critical flicker fusion and pupil measures (X9) Three visual analog scales (X 10) Performance battery (6 tasks, X 5) ARCI and SDQ (X5) Decreased frequency of data collection, every 1 to 12 hours Pulse, blood pressure and skin temperature (X 12) Alternating critical flicker fusion and pupil measures ( X 9) Three visual analog scales (X 12) Performance battery (6 tasks, X 12) ARCI and SDQ (X 12)
THC, A-9-Tetrahydrocannab~nol;ECG, electrocardiogram; ARCI, Addiction Research Center Inventory; SDQ, Slngle Dose Questionna~re.
operated according to the time line described in Table 11. Samples were obtained in the hour before smoking with the pump operated in the slow mode (0.2 mllmin). During and immediately after smoking (from 0 to 20 minutes), the pump was operated in the fast mode (5 mllmin) for rapid blood sampling. From 20 to 60 minutes, the medium pump speed provided integrated 5 ml samples over 5-minute collection periods. Samples from 1 to 3 hours after smoking were collected in the slow mode (0.2 mllmin) in 25-minute collection periods. Additional samples were collected over the next 3 days through an angiocatheter in the forearm vein in the dominant arm. Blood samples were collected on ice in 7 ml sodium
EDTA vacutainer tubes. Samples were centrifuged and the plasma removed for storage in silanized glass tubes at -30" C until the time of analysis. A-9-Tetrahydrocannabinol levels were determined by negative chemical ionization gas chromatography mass spectrometry by the method reported by Foltz et a].'* The limit of detection for A-9-tetrahydrocannabinol in plasma was 0.5 ng/ml with a 0.5 ml sample. Physiologic measures. A continuous electrocardiogram (Physio-control Lifepak 6 s cardiac monitor, Redmond, Wash.) was obtained from 1 hour before drug administration to 3 hours after smoking. An IVAC Vital Check model 4000 (IVAC Corp., San Diego, Calif.) was operated in the manual mode to col-
34 Huestis et al. lect pulse, mean arterial pressure, and diastolic and systolic blood pressures throughout the 4-hour study session. Skin and room temperatures were monitored with a Bailey Instruments model BAT8 (Bailey Instruments, Saddlebrook, N.J.) with a range of O" to 140" C. Critical flicker fusion threshold (CFF) was determined in the ascending direction (fusion response) with a Lafayette Instrument Co. model 12023 (Lafayette Instrument Co., Lafayette, Ind.) equipped with a model 12033 viewing chamber and operated in the autofrequency mode. The critical flicker fusion threshold or pupil diameter of the subjects was assessed alternately after each puff. The subject was instructed to peer into the viewing box and press a signal button when the light appeared to stop flickering and fuse into a solid beam. Pupil diameter was obtained after alternate marijuana puffs. A Polaroid (Polaroid Corp., Cambridge, Mass.) camera picture of the left eye of the subject was taken, and the diameter of the pupil was measured from the calibrated photograph. Behavioral and performance measures. All subjective measures and performance tasks were computerized for standardization of data collection and to facilitate adherence to the complex time line. An IBM personal computer AT model 5170 with an IBM color monitor model 5153 (IBM Instruments, Inc., Danbury, Conn.) was programmed to present questionnaires and performance tasks at appropriate times for data collection. Three visual analog scales, "feel drug," "like drug," and "dislike drug," were administered before, during, and after smoking. The scales consisted of analog lines representing the range of effects from none to very strong. The lines were presented in series with the questions "How much of a drug effect do you feel?," "How much do you like the drug?," or "How much do you dislike the drug?'An arrow located at the center of the line moved as subjects pressed computer keys to position the arrow in the correct spot along the 0 to 100 incremented analog scale. T o monitor the onset of subjective effects, the visual analog scales were administered after each marijuana cigarette inhalation. These measures were continued after the completion of smoking. The Single Dose Questionnaire, a standardized questionnaire that compares the addictive properties of a wide range of drugs, was included to evaluate subjects' classification, description, and liking of drug effects. The Single Dose Questionnaire was developed to quantitate transient changes in mood and subjective state induced by the administration or withdrawal of drugs. In
CLIN FHARMACOI. THER JULY 1992
addition, the Addiction Research Center Inventory short form was administered for evaluation of the effects of marijuana. l 3 This form consisted of the following subscales: pentobarbital-chlorpromazine-alcohol group (PCAG), morphine-benzedrine group (MBG), and the lysergic acid diethylamide (LSD) group. These subscales were designed to be measures of drug-induced sedation (PCAG), euphoria (MBG), and dysphoria (LSD). Five performance tasks from the Walter Reed Performance Assessment Battery and the Digit Symbol Substitution Test (DSST) were adapted by shortening the time allotted for each task.I4,l5 Each task lasted 60 seconds except for the serial addition and subtraction task, which was limited to 45 seconds. Subjects were trained for 1 week to reach an acceptable baseline level of performance before drug administration. Cognitive function was evaluated with the logical reasoning, manikin, and serial addition and subtraction tasks. Short-term memory and sustained attention were tested with the matrix and serial addition and subtraction tasks. Abstract reasoning was an important component of the manikin test, and time estimation was evaluated with the time wall task. The DSST psychomotor task tested both cognitive and motor skills. A single performance session consisted of a sequence of the six tasks. The computer automatically calculated the percentage of correct responses, throughput, and speed for each task and each session. Throughput was a measure of the number of correct responses per unit of time, and speed was the time from response opportunity to actual response. Subjects were provided feedback on their performance during training but not during the placebo or drug exposures. Procedure. Subjects were acclimated to the research environment and underwent training during the first week of residence. On each of 3 test days, subjects remained in the research laboratory from 7:45 AM to 12: 15 P M . Baseline biochemical, physiologic, and behavioral measures were collected from 8 AM to 9 A M . At 9 AM subjects smoked a single marijuana cigarette, and measurements were collected frequently over the course of the day and less frequently over the next 6 days according to the time line presented in Table 11. Subjects relaxed, but did not sleep, or read between measurements, After 12:15 PM, subjects returned to the ward and conducted routine activities, except for scheduled data collections in the research laboratory. Statistical analysis. Data were analyzed by a twoway repeated-measures ANOVA with Tukey's post hoc analysis to test for variations between conditions,
VOLUME 52 NUMBER I
Absorption of smoked marijuana
between times, and across condition and time. HuynhFeldt p values were used to correct for violations of the sphericity assumption.
RESULTS A-9-Tetrahydrocannabinol levels. The primary psychoactive constituent of marijuana, A-9-tetrahydrocannabinol, was rapidly absorbed during smoking. Blood levels of A-9-tetrahydrocannabinol during and immediately after smoking of a single 1.75% or 3.55% A-9-tetrahydrocannabinol cigarette are shown in Fig. 1. Active drug was found in plasma samples collected at the time of the first marijuana puff. Mean A-9-tetrahydrocannabinol concentrations of 7.0 + 8.1 and 18.1 k 12.0 nglml were observed after the first inhalation of a low- or high-dose cigarette, respectively. Levels continued to increase rapidly, and the peak mean levels occurred at 9 minutes, before initiation of the last puff sequence at 9.8 minutes. Mean peak levels were 79.0 2 25.2 and 152.0 k 86.3 nglml for the 1.75% and 3.55% A-9-tetrahydrocannabinol cigarettes, respectively. Differences in plasma drug concentrations were significant across conditions [F,,, = 7.55 @ < 0.040)]. A-9-Tetrahydrocannabinol decreased rapidly after the peak. Concentrations at 15 minutes after smoking of the low and high doses were 52.3 t 18.5 and 94.8 + 53.1 nglml, respectively. Levels of 17.3 7.3 and 29.7 + 15.2 ng/ml were noted after approximately 30 minutes. Within 2 hours, levels were at or below 5 nglml but remained detectable for up to 12 hours after smoking. Individual variability in A-9tetrahydrocannabinol levels after the smoking of a 3.55% A-9-tetrahydrocannabinol cigarette are illustrated in Fig. 2, a. Wide interindividual differences in A-9-tetrahydrocannabinol levels were noted despite use of the paced smoking protocol for controlled dose administration.
35
1.75% THC 3.55%THC
-2
2
6
10
14
18
22
Minutes Fig. 1. Mean A-9-tetrahydrocannabinol (THC) levels during
smoking of a single marijuana cigarette. Data represent the mean responses of six subjects. Drug was not detectable after smoking of placebo cigarettes. Arrows indicate the start of each inhalation. Each puff consisted of a 2-second inhalation, a 10-second hold period, and a 72-second exhale and rest period. Blood was collected by way of a continuous withdrawal pump and analyzed by negative chemical ionization gas chromatography- mass spectrometry. Note that THC levels peaked just before the last inhalation.
*
Physiologic effects Heart rate. Significant increases in heart rate after marijuana smoking are illustrated in Fig. 3, a [F2,,, = 24.65 @ < 0.0001)]. Mean peak increases of 46.0 + 18.6 and 55.8 + 22.2 beatslmin over baseline levels were noted after the smoking of a single 1.75% or 3.55% A-9-tetrahydrocannabinol cigarette. Mean peak heart rate increases occurred at 17.4 4.8 and 13.8 + 4.2 minutes after the low- or high-dose cigarette, respectively. Individual heart rate increases are illustrated in Fig. 2, b. Subject E showed the greatest increase in pulse, although he attained the lowest A-9-tetrahydrocannabinol blood levels. Heart rate re-
*
mained elevated for 3 hours after the high-dose exposure but had returned to baseline levels at this time point after the low-dose exposure. Blood pressure. Significant increases [F,,,, = 4.65 @ < 0.05)l in diastolic blood pressure were noted after the smoking of a single 1.75% or 3.55% A-9-tetrahydrocannabinol cigarette (Fig. 3, b). Peak mean diastolic blood pressure occurred 18 minutes after the high-dose cigarette (15.8 ? 10.8 mm Hg over baseline) and 30 minutes after the low-dose cigarette (12.7 + 11.3 mm Hg over baseline). Diastolic blood pressure returned to baseline levels within 3 to 4 hours. Mean arterial pressure and systolic blood pressure also increased after marijuana smoking, but effects did not reach statistical significance. Mean arterial pressure increased over baseline levels by 16.5 + 20.9 mm Hg at 40 minutes and 23.0 + 26.7 mm Hg at 18 minutes after the 1.75% and 3.55% doses, respectively. Mean systolic blood pressure increased 17.3 t 28.1 mm Hg above baseline at 35 minutes and 15.8 + 13.8 mm Hg above baseline at 21 minutes after the low- and high-dose cigarettes, respectively. Duration of effect for both measures was 1 to 3 hours.
CLIN PHARMACOL THER JULY 1992
36 Huestis et al. a) THC
1201
300 1
b) Pulse
d) Logical Reasoning
c) Feel Drug 60
40
!f cn8
20
0 0
1
2
0
3
1
3
2
Hours Fig. 2. Individual cannabinoid levels and pharmacodynamic effects after smoking of a single 3.55% A-9-tetrahydrocannabinol (THC) cigarette. a, Illustrates THC levels during and after the paced smoking protocol. b, Illustrates changes from baseline in pulse rate (in beats per minute). c, Illustrates changes from baseline in "feel drug" scores from the visual analog scale. d, Illustrates changes from baseline in accuracy (percentage correct) on the logical reasoning performance task.
Skin temperature. Skin temperature decreased after marijuana smoking, with mean peak effects observed at 45.0 and 55.2 minutes after the low- and high-dose cigarettes, respectively (Fig. 3, c). Significant differences in skin temperature were noted between placebo and both active doses between 30 to 80 minutes [F48,240= 6.28 (p < 0.001)]. Maximum decreases of 2.7 -C 5.8" C and 3.1 -C 3.9" C were observed after the 1.75% and 3.55% A-9-tetrahydrocannabinol cigarettes, respectively. Skin temperature had returned to baseline temperatures within 1 Y2 to 2Y2 hours. Critical Picker fusion. No significant effects were noted in critical flicker fusion threshold after marijuana smoking (Fig. 3, 4. Placebo and low-dose effects were unremarkable, but a decreased threshold was apparent with the 3.55% A-9-tetrahydrocannabino1 dose. Mean peak effect was observed 31 minutes
*
5.5 Hz decrease in after dosing with a mean 6.6 critical flicker fusion threshold. Pupils. No significant effects were noted on pupil diameter after marijuana smoking.
Subjective effects Visual analog scales. Visual analog scales for "feel drug" and "like drug" were sensitive indicators of marijuana's effects. Responses to the question, "How much drug effect do you feel?'showed significant increases [F2,,,, = 12.66 @ < 0.002)] after marijuana smoking (Fig. 4, a). Mean peak difference scores of 41.2 k 16.3 and 40.2 5 20.1 were noted after 16 and 10 minutes after the low- and high-dose exposures, respectively. Both the 1.75% and 3.55% A-Ptetrahydrocannabinol cigarettes produced similar subjective responses; a dose-response relationship was not shown
VOLUME 52 NUMBER 1
Absorption of smoked marijuana 37 b) Diastolic Blood Pressure 0
-20-1
1
1
-I"
c) Skin Temperature
.
01
d) Critical Flicker Fusion 0
-2
24
4
-6
1.75% THC
Hours Fig. 3. Changes in pulse, diastolic blood pressure, skin temperature, and critical flicker fusion threshold after smoking of a single marijuana cigarette. Physiologic data represent mean (N = 6)
differences from baseline.
between the two active doses but was significant between placebo and both active doses. Effects persisted for more than 12 hours. Large interindividual variations in "feel drug" responses were observed (Fig. 2, c). Subject E showed the greatest increase in "feel drug" scores, although he attained the lowest A-9-tetrahydrocannabinol blood levels. A less robust response was noted to the question "How much do you like the drug?'Mean peak effects were observed at 8.4 and 10.2 minutes during and immediately after smoking of a 1.75% or 3.55% A-9-tetrahydrocannabino1 cigarette, respectively. Little difference in response was apparent between the high and low doses, but measures reached significance between placebo and active doses. Effects were demonstrable for 6 to 12 hours. The visual analog scale "How much do you dislike the drug?" was not sensitive to the effects of smoked marijuana; there was no distinction in responses of the subjects between the placebo and active drug conditions.
Single Dose Questionnaire. The "feel drug" and "like drug" components of the Single Dose Questionnaire were sensitive indicators of the effects of marijuana; peak responses were noted at the first measurements at 31 minutes for both low and high doses. The score for subject-reported drug "liking" is illustrated in Fig. 4, b. Effects approached baseline levels at 12 hours after drug exposure. A small placebo effect was evident, but both "feel drug" and "like drug" responses were rated significantly greater than placebo ("feel drug" [F,,,, = 6.75 0, < 0.014)]; "like drug" [F,,,, = 5.82 @ < 0.035)]. Addiction Research Center Inventory. Significant increases on the PCAG subscale were noted between the placebo and 3.55% A-9-tetrahydrocannabinol conditions [F,,,, = 4.38 @ < 0.043)]. Mean peak sedation was reported at 3 hours (Fig. 4, c) after the lowdose cigarettes and 6 hours (data not presented) after the high-dose cigarettes, with effects diminishing beyond 12 hours. The LSD subscale illustrated mean
CLlN PHARMACOL THER TULY 1992
38 Huestis e t al.
1
a) Feel Drug Placebo
c) PCAG
r4
Hours Fig. 4. Mean changes (N = 6) in subjective effects after smoking of a single marijuana cigarette. Subject-reported "feel drug" (a) was obtained from the visual analog scale (VAS) for "How much drug effect do you feel?'Subject-reported "liking" (b) was determined from the Single Dose Questionnaire. The pentobarbital-chlorpromazine-alcohol (PCAG) group (c) and the lysergic acid diethylamide (LSD) group (d) scores were obtained from the Addiction Research Center Inventory.
peak dysphoric effects at 11/2 hours after smoking (Fig. 4, 4, although effects were not significantly different than placebo. The MBG subscale was not a sensitive indicator of the effects of marijuana; no significant differences were noted after smoking of placebo, 1.75% or 3.55% A-9-tetrahydrocannabinol cigarettes.
Performance Walter Reed performance battery. Three different parameters were assessed in each performance task; percentage of correct responses, throughput, and speed. None of the five tasks showed significant effects on throughput and speed after marijuana smoking. After smoking one 1.75% or 3.55% A-9-tetrahydrocannabinol cigarette, the percentage of correct
scores on the logical reasoning task were significantly lower [F,,,, = 5.00 (p < 0.031)] than after placebo (Fig. 5, a). The mean peak decrease in accuracy for the logical reasoning task was observed at the time of first measurement at 22 minutes. Mean peak decreases of 14.1% -+ 10.8% and 12.5% + 18.0% were found after smoking one 1.75% or 3.55% A-9-tetrahydrocannabinol cigarette. Significant decreases in accuracy were observed between 22 and 105 minutes. The percent accuracy on the logical reasoning task had returned to baseline levels within 3 hours for all subjects (Fig. 2, 4. No significant differences in accuracy after marijuana smoking were noted in the matrix, serial addition and subtraction, manikin, and time wall performance tasks. In the time wall task, subjects had to estimate the completion of a 10-second time span. As
VOLUME 52 NUMBER 1
Absorption of s m k e d marijuana 39
loo
- a) Logical reasoning
loo- b) DSST
90 -
80-
+ 1.75% THC
10.8 7
,d) Manikin
c) Timewall
10.6 10.4 10.2
I @-
-
a
50-
10.0 -
9.8 -1
1
0
1
2
3
40, -1
1
0
1
2
3
Hours Fig. 5. Effects on mean performance measures (N = 6) after smoking of a single marijuana cigarette. Tasks included the logical reasoning (a), time wall (c), and manikin (d) tasks adapted from
the Walter Reed Performance Assessment Battery and the Digit Symbol Substitution Test (DSST) (b). The percentage of correct responses is illustrated for the logical reasoning and DSST tasks. The ability of subjects to estimate a 10-second time span is illustrated in the time wall task, and response speed is shown for the manikin task.
seen in Fig. 5, c, no significant differences in estimated time were observed after marijuana smoking. Response speed also was evaluated for all tasks and was not affected by marijuana administration. An example of the response speed for the manikin task is illustrated in Fig. 5, d. DSST. No significant differences in accuracy or response speed were noted in the DSST after smoking placebo or low- or high-dose marijuana cigarettes. Fig. 5, b, illustrates the percentage of correct scores for the DSST task after smoking a single marijuana cigarette.
DISCUSSION The first inhalation of marijuana smoke produced detectable blood levels of A-9-tetrahydrocannabinol, with peak concentrations occurring shortly before the cigarette had been completely smoked. Integrated
blood samples collected over a 60-second period after the first puff of a low- or high-dose marijuana cigarette (1.75% or 3.55% A-9-tetrahydrocannabinol) contained 7.0 8.1 and 18.0 12.0 nglml of A-9tetrahydrocannabinol, respectively. Concentrations increased rapidly with each additional puff. Halfway through smoking, after four puffs, levels of 41.0 10.4 and 102.7 + 39.1 nglml were attained after the low and high doses. Mean peak drug levels of 79.0 ? 25.2 and 152.0 ? 86.3 nglml occurred at 9 minutes, just before the last inhalation. Because of the rapidity of changes occurring during smoking, few other studies have attempted blood collection during the smoking process. Perez-Reyes et al.7.9 were the first to note that peak A-9-tetrahydrocannabinol concentrations developed before the end of smoking. In their studies, subjects determined their own rate and speed of smoking, inhaling many puffs (28 to 30) over 15 to 19 min-
*
*
*
40 Huestis et al. utes. One proposed explanation for the occurrence of peak levels before cessation of smoking was that the peak levels could be an artifact of the sampling protocoil6; that is, random sampling over a large number of individual peaks and valleys after each puff could have produced artificial peak responses. In the present study, continuous blood sampling eliminated the possibility of a sampling artifact and documented clearly that peak A-9-tetrahydrocannabinol levels occurred during marijuana smoking. It is apparent, however, that the time to peak blood levels is greatly affected by the individual smoking protocol; the number of puffs, time between puffs, and volume and length of inhalations can have substantial impact on drug delivery. Two additional explanations have been proposed for the occurrence of peak levels of A-9-tetrahydrocannabinol before the end of ~ r n o k i n gHot . ~ air extraction of drug could possibly deliver higher levels during the first few puffs. As the subject inhales and draws hot smoke through the marijuana cigarette, the A-9-tetrahydrocannabinol could be extracted differentially out of the plant material. However, preliminary studies did not support this explanation. The amount of active drug in the terminal portion of unsmoked and smoked marijuana cigarettes was measured and the amounts were found to be similar. Although this suggests that smoking does not deplete A-9-tetrahydrocannabinol content, a more critical experiment would be quantitative analysis of A-9-tetrahydrocannabinol in the mainstream smoke of a succession of individual puffs. Another explanation for the early peaking of A-9tetrahydrocannabinol levels during smoking could be that subjects do not uniformly smoke cigarettes from beginning to end. The smoking dynamics of subjects have been evaluated throughout test sessions and no significant differences in the number of puffs, hold time, or time between puffs were noted.7 A critical factor that was not reported was the inhalation volume of each puff. If the subject breathed deeper during the early puffs, more drug presumably would be delivered. These early differences in smoking dynamics also could account for peak drug levels during marijuana smoking. This would be consistent with findings of tobacco cigarette smoking, which confirm the tendencies of smokers to self-administer decreasing amounts of tobacco smoke across successive puffs on a cigarette. l 7 An additional finding from the current data was the similarity between time-to-peak drug levels and timeto-peak effects for some measurements. Previous studies have emphasized time delays between peak drug
C I J N PHARMACOL THEK JULY 1992
levels and peak physiologic and behavioral effects. One of the most sensitive indicators of marijuana exposure, subjective "high," has been reported to peak 30 and 60 minutes after ~ m o k i n g . ~ The . ' ~ present study design incorporated measurements of subjective effects after each marijuana inhalation to monitor the rapid onset of effects. Three of six subjects reported increases in drug "liking" scores after the first puff, 15 seconds after smoking began, and all subjects reported increases by the second puff of a high-dose cigarette. Maximum drug "liking" occurred at 8.4 minutes (six puffs) and 10.2 minutes (eight puffs) during smoking of a low- or high-dose cigarette, respectively. A similar pattern of effects was observed with the "feel drug" visual analog scale. Subjects responded rapidly with three of six subjects recording increases at 15 seconds. Four inhalations were necessary for all subjects to report increased "feel" drug scores with peak effects noted after 13.2 and 16.2 minutes for the low and high doses, respectively. Previous reports of subjective "high" have indicated that peak subjective effects appeared between 5 and 20 minutes after the end of Marijuana exposure also produced rapid changes in some physiologic effects. Continuous electrocardiogram monitoring documented increased heart rate in all subjects soon after the start of smoking. Maximum heart rates were recorded within 4 minutes of the last inhalation after the high-dose cigarette. Previous reports also showed peak heart rate increases between 5 and 10 minutes after smoking c e s ~ a t i o n . In ~ , addi~~ tion, increased diastolic blood pressure was recorded for four of six subjects after two puffs of the highdose cigarette. Peak responses were recorded within 18 minutes after the start of smoking. The understanding of the pharmacology of marijuana requires careful observation of the onset, peak, and duration of significant physiologic, behavioral, and biochemical effects. Capture of rapid, fleeting changes or more delayed, persistent effects is highly dependent on the choice of experimental methods and the frequency of testing. In the present study, performance data was not collected until 22 minutes into the complex time line. The time required for administration of the six performance tasks prevented earlier scheduled testing. Maximal decreases in percent correct responses for the logical reasoning task already had been achieved at the first measurement. Consequently, the onset and time of peak effects could not be determined. Early decrements in performance on a variety of tasks 5 to 15 minutes after the end of smoking have been previously reported.'0321Obviously,
VOLUME 52 NUIMRER 1
drug effects should be monitored at the time of drug delivery to active sites, that is, during drug absorption, because of the fast delivery of drug and rapid onset of effects during marijuana smoking. Studies that combine rapid blood collection, paced smoking protocols, and timely collection of physiologic and behavioral measures are essential for the complete characterization of the absorption phase of marijuana smoking. We gratefully acknowledge the support of Dakmed Corporation, Buffalo, N.Y., for use of the ambulatory withdrawal pump, the Center for Human Toxicology for performance of blood analyses for A-9-tetrahydrocannabinol, and Dr. Stephen Heishman for technical assistance with the Walter Reed Performance Assessment Battery.
References National Institute on Drug Abuse. National household survey on drug abuse: Highlights 1990. Washington: US Government Printing Office; 1991; DHHS pub no (ADM)91-1789. Johnstone LD, O'Malley PM, and Bachman JG. Drug use among American high school seniors, college students and young adults, 1975-1990. Washington: US Government Printing Office; 1991; DHHS pub no (ADM)91-1813. Lindgren JE, Ohlsson A, Agurell S , Hollister L, Gillespie H. Clinical effects and plasma levels of THC in heavy and light users. Psychopharmacology 1981; 74:2O8- 12. Ohlsson A, Lindgren JE, Wahlen A , Agurell S, Hollister LE, Gillespie HK. Plasma delta-9-tetrahydrocannabin01 concentrations and clinical effects after oral and intravenous administration and smoking. CLINPHARMACOL THERl980;28:409- 16. Hollister LE, Gillespie HK, Ohlsson A, Lindgren JE, Wahlen A, Agurell S. Do plasma concentrations of delta-9-tetrahydrocannabinol reflect the degree of intoxication? J Clin Pharmacol I98 1;2 1: 17 1S-7s. Benowitz N. Clinical pharmacology of inhaled drugs of abuse: implications in understanding nicotine dependence. In: Chiang CN, Hawks RL, eds. Research findings on smoking of abused substances. Washington: NIDA Res Monogr 99, 1990. Perez-Reyes M, Di Guiseppi S, Davis KH, Schindler V, Cook CE. Comparison of effects of marihuana cigarettes of three different potencies. CLINPHARMACOL THER 1982;31:617-24. Davis KH, McDaniel IA, Caddel LW, Moody PL. Some smoking characteristics of marijuana cigarettes. In: Agurell S, Dewey WL, Willette RE, eds. The cannabinoids: chemical, pharmacologic, and therapeutic aspects. New York: Academic Press, 1984:97-109.
Absorption of smoked mavijuana 4 1 9. Perez-Reyes M, Owens SM, Di Guiseppi S. The clinical pharmacology and dynamics of marihuana cigarette smoking. J Clin Pharmacol 1981;21:201S-7s. 10. Barnett G, Licko V, Thompson T. Behavioral pharmacokinetics of marijuana. Psychopharmacology 1985; 8551-6. 11. Yesavage JA, Leirer VO, Denari M, Hollister LE. Carry-over effects of marijuana intoxication on aircraft pilot performance: a preliminary report. Am J Psychiatry 1985;142: 1325-9. 12. Foltz RL, McGinnis KM, Chinn DM. Quantitative measurement of A-9-tetrahydrocannabinol and two major metabolites in physiological specimens using capillary column gas chromatography negative ion chemical ionization mass spectrometry. Biomed Mass Spectrometry 1983;10:316-23. 13. Haertzen CA, Hickey JE. Addiction Research Center Inventory (ARCI): measurement of euphoria and other drug effects. In: Bozarth MA, ed. Methods of assessing the reinforcing properties of abused drugs. Heidelberg: Springer-Verlag, 1987:489-524. 14. Thorne DR, Censer SG, Sing HC, Hegge FW. The Walter Reed Performance Assessment Battery. Neurobehav Toxic01 Teratol 1985;7:4 15-8. 15. McLeod DR, Griffiths RR, Bigelow GE, Yingling J. Computer technology: an automated version of the digit symbol substitution test (DSST). Behav Res Methods Instrumentation 1982;14:463-6. 16. Barnett G, Chiang CN, Perez-Reyes M, Owens SM. Kinetic study of smoking marijuana. J Pharmacokinet Biopharm 1982;10:495-506. 17. Griffiths RR, Henningfield JE. Experimental analysis of human cigarette smoking behavior. Fed Proc 1982; 41 :234-40. 18. Lemberger L, Weiss JL, Watanabe AM, Galanter IM, Wyatt RJ, Cardon PV. Delta-9-tetrahydrocannabinol: temporal correlation of the psychologic effects and blood levels after various routes of administration. N Engl J Med 1972;286:685-8. 19. Cocchetto DM, Owens SM, Perez-Reyes M, DiGuiseppi S, Miller LL. Relationship between plasma delta9-tetrahydrocannabinol concentration and pharmacologic effects in man. Psychopharmacology 1981 ;75: 158-64. 20. Agurell S, Levander S, Binder M, et al. Pharrnacokinetics of delta-8-tetrahydrocannabinol (delta-6-tetrahydrocannabinol) in man after smoking: relations to physiological and psychological effects. In: Braude MC. Szara S, eds. The pharmacology of marijuana. New York: Raven Press, 1976:49-6 1 . 21. Heishman SJ, Stitzer M, Yingling JE. Effects of tetrahydrocannabinol content on marijuana smoking behavior, subjective reports, and performance. Pharmacol Biochem Behav 1989;34:173-9.
1
Rapid blood collection, a paced smoking protocol and timely collection of physiologic and behavioral measures were used to characterize the absorption phase of marijuana smoking. Six healthy males smoked a single marijuana cigarette (placebo, 1.75%, or 3.55% A-9-tetrahydrocannabinol) in a doubleblind, randomized, Latin square study design. Rapid blood sampling with a continuous withdrawal pump allowed simultaneous collection with concurrent physiologic and behavioral measures. Mean plasma levels of 7.0 and 18.1 nglml A-9-tetrahydrocannabinol were observed after the first inhalation of a 1.75% and 3.55% A-9-tetrahydrocannabinol cigarette, respectively. Blood levels increased rapidly and peaked at 9 minutes, before initiation of the last puff sequence at 9.8 minutes. Three of six subjects reported increases in drug "liking" scores after the first puff, and all subjects responded by the second puff of a high dose cigarette. Significant increases in heart rate and diastolic blood pressure occurred shortly after peak blood levels. Previous studies have indicated that there is a substantial time delay between peak plasma levels of A-9-tetrahydrocannabinol and drug-induced effects. This study showed that behavioral and physiologic effects appear concurrently or within minutes after the rapid appearance of A-9tetrahydrocannabinol in blood during marijuana smoking. (CLINPHARMACOL THERl992;52:3 1-41.)
Marilyn A. Huestis, MS, Angela H. Sampson, RN, Barbara J. Holicky, RN, Jack E. Henningfield, PhD, and Edward J. Cone, PhD Baltimore, Md. More than one third of the population of the United States have used marijuana one or more times in their lifetimes; 10% have used it within the past year.' Although use of marijuana among American youth continues to decline, annual use in 1990 was greater than .~ is the preferred 25% among young a d u ~ t sInhalation route of exposure, although oral use is not uncommon. Drug delivery during marijuana smoking is characterized by rapid absorption of A-9-tetrahydrocannabinol, the major psychoactive component of marijuana, with peak levels paralleling those found afIntense pleasurable ter intravenous admini~tration.~.~ and strongly reinforcing effects may be produced by way of both routes because of almost immediate drug exposure to the central nervous system.' Nicotine and other inhaled drugs have been reported to distribute quickly from the lungs to the brain.6 Estimates of 11 seconds or less were made from the start of a cigarette puff to the delivery of nicotine to the central nervous
From the Addiction Research Center, National Institute on Drug Abuse. Received for publication Dec. 20, 1991; accepted March 24, 1992. Reprint requests: Edward J. Cone, PhD, Addiction Research Center, NIDA, P.O. Box 5180, Baltimore, MD 21224. 13/1/38106
system. Less than 19 seconds was required for complete drug transit through the brain. Rapid absorption of A-9-tetrahydrocannabinol during marijuana smoking also has been documented.' Quantitation of possible behavioral and physiologic effects during the early stages of marijuana smoking have been limited by the rapidity of drug absorption by way of this route of administration. A second impediment to the investigation of the absorption of A-9-tetrahydrocannabinol is the inability to deliver a precise dose. Only 20% to 37% of active drug is delivered in the mainstream smoke.' Sidestream smoke losses of 40% to 50% and pyrolytic destruction of 23% to 30% of the active drug have been documented. Intrasubject and intersubject variability in smoking dynamics contribute to the uncertainty in dose delivery. The number, duration and spacing of puffs, hold time, and inhalation volume greatly influ. ~ alternative apence the degree of drug e x p o ~ u r e An proach to accurate dosing is the determination of blood levels of A-9-tetrahydrocannabinol. Measurement during the absorption phase is a difficult task because of the need for fast sequential blood sampling. Also, sampling methods should not conflict with the collection of physiologic, subjective, or performance measures. These constraints have led other investiga-
31
CLIN PHAKMACOL 1'HF.K JULY 1992
32 Huestis et al. Table I. Subject characteristics Subject
Age (yr)
B 36 C 33 E 29 F 28 G 33 H 29 Mean -+ SD 31.3 t 3.1
Age at jrst Height (inches) Weight (kg) marijuana use (yr) Average use* Drug use previous 30 daysf 68.0 71.3 72.5 64.8 67.0 68.8 68.7 5 2.8
70.9 80.9 86.4 68.6 65.0 93.6 77.6 ? 11.2
16 14 14 15 15 13 14.5 i 1.0
0.4 7.9 0.8 1 .O 0.5 3.0 2.3
5
NAH,M N,E, M N,E,C,M N,E,H,C, M N,E,H, C N,E,H,C,B,M 2.9
*Average number of mar~juanacigarettes smoked each week. t N , Nicotine; E, ethanol; H , hcroin: M , marijuana; C, cocaine; B , barbiturates
tors to collect blood samples and behavioral data in separate study sessions or to eliminate drug measurements entirely. This study was designed to characterize the absorption of A-9-tetrahydrocannabinol during marijuana smoking and to define the onset, peak, and duration of the pharmacodynamic effects of marijuana. A continuous withdrawal pump with adjustable speeds was used for blood sampling. The use of the fast sequential sampling technique did not compromise subjects' attention during completion of experimental tests. Subjective, performance, and physiologic measures were collected concurrently with blood levels.
'".' '
METHODS Subjects. Six healthy male volunteers with histories of marijuana use were recruited from the local population through advertisements and personal communication. All subjects resided on a closed research ward of the Addiction Research Center, National Institute on Drug Abuse, for 4 to 6 weeks. Subjects provided informed consent and were paid for their participation. Medical histories, physical examinations, and electrocardiograms were reviewed to determine if active infectious disease or underlying cardiac, seizure, or renal disorder precluded the inclusion of any subject in the study. Any ongoing psychiatric illness that distorted the subjects' perception of reality or that interfered with their ability to function on a daily basis was also a basis for study exclusion. Subjects were not accepted for the study if they were currently receiving any prescribed medication. Additional evaluation criteria included blood chemistry and hematology screens, Shipley Institute Living Scale and SCL-90-R Mood Scale evaluations (Hopkins Symptom Checklist 90), and descriptions of current and previous drug use. Subjects were required to have a raw score on the vocabulary section of the Shipley Institute Living
Scale of at least 18 to be accepted in the study. In addition, the general severity index score (T score) on the SCL-90-R had to be less than 63. Light to moderate previous marijuana use was required, but urine drug screens for cannabinoid metabolites had to be negative before admission. Urine drug testing for amphetamines, barbiturates, benzodiazepines, cocaine metabolite, methadone, opiates, and phencyclidine was performed before admission and during residence to ensure lack of exposure to unauthorized drugs during the study. Characteristics of the subjects are detailed in Table I. Dose and route of administration. Marijuana cigarettes were obtained from the Research Technology Branch, National Institute on Drug Abuse. Subjects smoked a single marijuana cigarette (placebo or 1.75% or 3.55% A-9-tetrahydrocannabinol) according to a computer-controlled paced smoking procedure on each of 3 test days. The smoking protocol consisted of a 2-second inhalation, a 10-second hold period, and a 72-second exhalation and rest period. A total of eight puffs were inhaled in 11.2 minutes. Doses were administered according to a randomized, double-blind Latin square design at weekly intervals. Collection, preparation, and analysis of plasma samples. The Dakmed ambulatory withdrawal pump model ML6-5S3R (Dakmed, Inc., Buffalo, N.Y .) was used for rapid blood sampling and integrated blood collections from 1 hour before until 3 hours after drug administration. Blood was collected through 4 feet of Dakmed silanized tubing by way of a 19-gauge butterfly needle inserted in a nondominant forearm vein. Subjects received 5000 units of sodium heparin before drug administration to prevent blood clotting in the tubing. The clotting times of all subjects had returned to normal within 4 hours after heparin administration. The pump was calibrated for collection of 5 ml blood samples in 1, 5, or 25 minutes. The pump was
VOLUME 52 NUMBER 1
Absovption ofsmoked marijuana 3 3
Table 11. Marijuana study time line Time
Blood samples (5 ml)
-60 to 0 minutes
Angiocatheter and butterfly placement Heparin administration Continuous withdrawal, 0.2 mlimin (X2) Continuous withdrawal, 5 mlimin (X 10)
0 to I1 minutes
11 to 20 minutes
Intermittent sampling, 5 mlimin (X3)
20 to 50 minutes
Integrated sampling, 1 mlimin ( X 3 )
50 to 180 minutes
Integrated blood sampling 0.2 mlirnin (X 5)
180 minutes to 7 days
Disconnect withdrawal Pump Sampling through angiocatheter ( X 12)
Data collection Baseline measures of all parameters (X 2)
Computer directed smoking of one marijuana cigarette (075, 1.75%, or 3.55% THC) Continuous ECG; Pulse, blood pressure, and skin temperature every 3 minutes (X4) Alternating critical flicker fusion and pupil measures with each puff (X4) Three visual analog scales during each puff (X8) Continuous ECG Pulse, blood pressure, and skin temperature every 3 rninutes (X3) Alternating critical flicker fusion and pupil measures (X3) Three visual analog scales ( x 4 ) Continuous ECG Pulse, blood pressure, and skin temperature every 5 rninutes ( x 6 ) Alternating critical flicker fusion and pupil measures ( x 3) Three visual analog scales ( x 4 ) First administrations of performance battery (6 tasks, X2) First administration of ARCI and SDQ ( X I) Pulse, blood pressure, and skin temperature every 15 minutes (X 1 I) Alternating critical flicker fusion and pupil measures (X9) Three visual analog scales (X 10) Performance battery (6 tasks, X 5) ARCI and SDQ (X5) Decreased frequency of data collection, every 1 to 12 hours Pulse, blood pressure and skin temperature (X 12) Alternating critical flicker fusion and pupil measures ( X 9) Three visual analog scales (X 12) Performance battery (6 tasks, X 12) ARCI and SDQ (X 12)
THC, A-9-Tetrahydrocannab~nol;ECG, electrocardiogram; ARCI, Addiction Research Center Inventory; SDQ, Slngle Dose Questionna~re.
operated according to the time line described in Table 11. Samples were obtained in the hour before smoking with the pump operated in the slow mode (0.2 mllmin). During and immediately after smoking (from 0 to 20 minutes), the pump was operated in the fast mode (5 mllmin) for rapid blood sampling. From 20 to 60 minutes, the medium pump speed provided integrated 5 ml samples over 5-minute collection periods. Samples from 1 to 3 hours after smoking were collected in the slow mode (0.2 mllmin) in 25-minute collection periods. Additional samples were collected over the next 3 days through an angiocatheter in the forearm vein in the dominant arm. Blood samples were collected on ice in 7 ml sodium
EDTA vacutainer tubes. Samples were centrifuged and the plasma removed for storage in silanized glass tubes at -30" C until the time of analysis. A-9-Tetrahydrocannabinol levels were determined by negative chemical ionization gas chromatography mass spectrometry by the method reported by Foltz et a].'* The limit of detection for A-9-tetrahydrocannabinol in plasma was 0.5 ng/ml with a 0.5 ml sample. Physiologic measures. A continuous electrocardiogram (Physio-control Lifepak 6 s cardiac monitor, Redmond, Wash.) was obtained from 1 hour before drug administration to 3 hours after smoking. An IVAC Vital Check model 4000 (IVAC Corp., San Diego, Calif.) was operated in the manual mode to col-
34 Huestis et al. lect pulse, mean arterial pressure, and diastolic and systolic blood pressures throughout the 4-hour study session. Skin and room temperatures were monitored with a Bailey Instruments model BAT8 (Bailey Instruments, Saddlebrook, N.J.) with a range of O" to 140" C. Critical flicker fusion threshold (CFF) was determined in the ascending direction (fusion response) with a Lafayette Instrument Co. model 12023 (Lafayette Instrument Co., Lafayette, Ind.) equipped with a model 12033 viewing chamber and operated in the autofrequency mode. The critical flicker fusion threshold or pupil diameter of the subjects was assessed alternately after each puff. The subject was instructed to peer into the viewing box and press a signal button when the light appeared to stop flickering and fuse into a solid beam. Pupil diameter was obtained after alternate marijuana puffs. A Polaroid (Polaroid Corp., Cambridge, Mass.) camera picture of the left eye of the subject was taken, and the diameter of the pupil was measured from the calibrated photograph. Behavioral and performance measures. All subjective measures and performance tasks were computerized for standardization of data collection and to facilitate adherence to the complex time line. An IBM personal computer AT model 5170 with an IBM color monitor model 5153 (IBM Instruments, Inc., Danbury, Conn.) was programmed to present questionnaires and performance tasks at appropriate times for data collection. Three visual analog scales, "feel drug," "like drug," and "dislike drug," were administered before, during, and after smoking. The scales consisted of analog lines representing the range of effects from none to very strong. The lines were presented in series with the questions "How much of a drug effect do you feel?," "How much do you like the drug?," or "How much do you dislike the drug?'An arrow located at the center of the line moved as subjects pressed computer keys to position the arrow in the correct spot along the 0 to 100 incremented analog scale. T o monitor the onset of subjective effects, the visual analog scales were administered after each marijuana cigarette inhalation. These measures were continued after the completion of smoking. The Single Dose Questionnaire, a standardized questionnaire that compares the addictive properties of a wide range of drugs, was included to evaluate subjects' classification, description, and liking of drug effects. The Single Dose Questionnaire was developed to quantitate transient changes in mood and subjective state induced by the administration or withdrawal of drugs. In
CLIN FHARMACOI. THER JULY 1992
addition, the Addiction Research Center Inventory short form was administered for evaluation of the effects of marijuana. l 3 This form consisted of the following subscales: pentobarbital-chlorpromazine-alcohol group (PCAG), morphine-benzedrine group (MBG), and the lysergic acid diethylamide (LSD) group. These subscales were designed to be measures of drug-induced sedation (PCAG), euphoria (MBG), and dysphoria (LSD). Five performance tasks from the Walter Reed Performance Assessment Battery and the Digit Symbol Substitution Test (DSST) were adapted by shortening the time allotted for each task.I4,l5 Each task lasted 60 seconds except for the serial addition and subtraction task, which was limited to 45 seconds. Subjects were trained for 1 week to reach an acceptable baseline level of performance before drug administration. Cognitive function was evaluated with the logical reasoning, manikin, and serial addition and subtraction tasks. Short-term memory and sustained attention were tested with the matrix and serial addition and subtraction tasks. Abstract reasoning was an important component of the manikin test, and time estimation was evaluated with the time wall task. The DSST psychomotor task tested both cognitive and motor skills. A single performance session consisted of a sequence of the six tasks. The computer automatically calculated the percentage of correct responses, throughput, and speed for each task and each session. Throughput was a measure of the number of correct responses per unit of time, and speed was the time from response opportunity to actual response. Subjects were provided feedback on their performance during training but not during the placebo or drug exposures. Procedure. Subjects were acclimated to the research environment and underwent training during the first week of residence. On each of 3 test days, subjects remained in the research laboratory from 7:45 AM to 12: 15 P M . Baseline biochemical, physiologic, and behavioral measures were collected from 8 AM to 9 A M . At 9 AM subjects smoked a single marijuana cigarette, and measurements were collected frequently over the course of the day and less frequently over the next 6 days according to the time line presented in Table 11. Subjects relaxed, but did not sleep, or read between measurements, After 12:15 PM, subjects returned to the ward and conducted routine activities, except for scheduled data collections in the research laboratory. Statistical analysis. Data were analyzed by a twoway repeated-measures ANOVA with Tukey's post hoc analysis to test for variations between conditions,
VOLUME 52 NUMBER I
Absorption of smoked marijuana
between times, and across condition and time. HuynhFeldt p values were used to correct for violations of the sphericity assumption.
RESULTS A-9-Tetrahydrocannabinol levels. The primary psychoactive constituent of marijuana, A-9-tetrahydrocannabinol, was rapidly absorbed during smoking. Blood levels of A-9-tetrahydrocannabinol during and immediately after smoking of a single 1.75% or 3.55% A-9-tetrahydrocannabinol cigarette are shown in Fig. 1. Active drug was found in plasma samples collected at the time of the first marijuana puff. Mean A-9-tetrahydrocannabinol concentrations of 7.0 + 8.1 and 18.1 k 12.0 nglml were observed after the first inhalation of a low- or high-dose cigarette, respectively. Levels continued to increase rapidly, and the peak mean levels occurred at 9 minutes, before initiation of the last puff sequence at 9.8 minutes. Mean peak levels were 79.0 2 25.2 and 152.0 k 86.3 nglml for the 1.75% and 3.55% A-9-tetrahydrocannabinol cigarettes, respectively. Differences in plasma drug concentrations were significant across conditions [F,,, = 7.55 @ < 0.040)]. A-9-Tetrahydrocannabinol decreased rapidly after the peak. Concentrations at 15 minutes after smoking of the low and high doses were 52.3 t 18.5 and 94.8 + 53.1 nglml, respectively. Levels of 17.3 7.3 and 29.7 + 15.2 ng/ml were noted after approximately 30 minutes. Within 2 hours, levels were at or below 5 nglml but remained detectable for up to 12 hours after smoking. Individual variability in A-9tetrahydrocannabinol levels after the smoking of a 3.55% A-9-tetrahydrocannabinol cigarette are illustrated in Fig. 2, a. Wide interindividual differences in A-9-tetrahydrocannabinol levels were noted despite use of the paced smoking protocol for controlled dose administration.
35
1.75% THC 3.55%THC
-2
2
6
10
14
18
22
Minutes Fig. 1. Mean A-9-tetrahydrocannabinol (THC) levels during
smoking of a single marijuana cigarette. Data represent the mean responses of six subjects. Drug was not detectable after smoking of placebo cigarettes. Arrows indicate the start of each inhalation. Each puff consisted of a 2-second inhalation, a 10-second hold period, and a 72-second exhale and rest period. Blood was collected by way of a continuous withdrawal pump and analyzed by negative chemical ionization gas chromatography- mass spectrometry. Note that THC levels peaked just before the last inhalation.
*
Physiologic effects Heart rate. Significant increases in heart rate after marijuana smoking are illustrated in Fig. 3, a [F2,,, = 24.65 @ < 0.0001)]. Mean peak increases of 46.0 + 18.6 and 55.8 + 22.2 beatslmin over baseline levels were noted after the smoking of a single 1.75% or 3.55% A-9-tetrahydrocannabinol cigarette. Mean peak heart rate increases occurred at 17.4 4.8 and 13.8 + 4.2 minutes after the low- or high-dose cigarette, respectively. Individual heart rate increases are illustrated in Fig. 2, b. Subject E showed the greatest increase in pulse, although he attained the lowest A-9-tetrahydrocannabinol blood levels. Heart rate re-
*
mained elevated for 3 hours after the high-dose exposure but had returned to baseline levels at this time point after the low-dose exposure. Blood pressure. Significant increases [F,,,, = 4.65 @ < 0.05)l in diastolic blood pressure were noted after the smoking of a single 1.75% or 3.55% A-9-tetrahydrocannabinol cigarette (Fig. 3, b). Peak mean diastolic blood pressure occurred 18 minutes after the high-dose cigarette (15.8 ? 10.8 mm Hg over baseline) and 30 minutes after the low-dose cigarette (12.7 + 11.3 mm Hg over baseline). Diastolic blood pressure returned to baseline levels within 3 to 4 hours. Mean arterial pressure and systolic blood pressure also increased after marijuana smoking, but effects did not reach statistical significance. Mean arterial pressure increased over baseline levels by 16.5 + 20.9 mm Hg at 40 minutes and 23.0 + 26.7 mm Hg at 18 minutes after the 1.75% and 3.55% doses, respectively. Mean systolic blood pressure increased 17.3 t 28.1 mm Hg above baseline at 35 minutes and 15.8 + 13.8 mm Hg above baseline at 21 minutes after the low- and high-dose cigarettes, respectively. Duration of effect for both measures was 1 to 3 hours.
CLIN PHARMACOL THER JULY 1992
36 Huestis et al. a) THC
1201
300 1
b) Pulse
d) Logical Reasoning
c) Feel Drug 60
40
!f cn8
20
0 0
1
2
0
3
1
3
2
Hours Fig. 2. Individual cannabinoid levels and pharmacodynamic effects after smoking of a single 3.55% A-9-tetrahydrocannabinol (THC) cigarette. a, Illustrates THC levels during and after the paced smoking protocol. b, Illustrates changes from baseline in pulse rate (in beats per minute). c, Illustrates changes from baseline in "feel drug" scores from the visual analog scale. d, Illustrates changes from baseline in accuracy (percentage correct) on the logical reasoning performance task.
Skin temperature. Skin temperature decreased after marijuana smoking, with mean peak effects observed at 45.0 and 55.2 minutes after the low- and high-dose cigarettes, respectively (Fig. 3, c). Significant differences in skin temperature were noted between placebo and both active doses between 30 to 80 minutes [F48,240= 6.28 (p < 0.001)]. Maximum decreases of 2.7 -C 5.8" C and 3.1 -C 3.9" C were observed after the 1.75% and 3.55% A-9-tetrahydrocannabinol cigarettes, respectively. Skin temperature had returned to baseline temperatures within 1 Y2 to 2Y2 hours. Critical Picker fusion. No significant effects were noted in critical flicker fusion threshold after marijuana smoking (Fig. 3, 4. Placebo and low-dose effects were unremarkable, but a decreased threshold was apparent with the 3.55% A-9-tetrahydrocannabino1 dose. Mean peak effect was observed 31 minutes
*
5.5 Hz decrease in after dosing with a mean 6.6 critical flicker fusion threshold. Pupils. No significant effects were noted on pupil diameter after marijuana smoking.
Subjective effects Visual analog scales. Visual analog scales for "feel drug" and "like drug" were sensitive indicators of marijuana's effects. Responses to the question, "How much drug effect do you feel?'showed significant increases [F2,,,, = 12.66 @ < 0.002)] after marijuana smoking (Fig. 4, a). Mean peak difference scores of 41.2 k 16.3 and 40.2 5 20.1 were noted after 16 and 10 minutes after the low- and high-dose exposures, respectively. Both the 1.75% and 3.55% A-Ptetrahydrocannabinol cigarettes produced similar subjective responses; a dose-response relationship was not shown
VOLUME 52 NUMBER 1
Absorption of smoked marijuana 37 b) Diastolic Blood Pressure 0
-20-1
1
1
-I"
c) Skin Temperature
.
01
d) Critical Flicker Fusion 0
-2
24
4
-6
1.75% THC
Hours Fig. 3. Changes in pulse, diastolic blood pressure, skin temperature, and critical flicker fusion threshold after smoking of a single marijuana cigarette. Physiologic data represent mean (N = 6)
differences from baseline.
between the two active doses but was significant between placebo and both active doses. Effects persisted for more than 12 hours. Large interindividual variations in "feel drug" responses were observed (Fig. 2, c). Subject E showed the greatest increase in "feel drug" scores, although he attained the lowest A-9-tetrahydrocannabinol blood levels. A less robust response was noted to the question "How much do you like the drug?'Mean peak effects were observed at 8.4 and 10.2 minutes during and immediately after smoking of a 1.75% or 3.55% A-9-tetrahydrocannabino1 cigarette, respectively. Little difference in response was apparent between the high and low doses, but measures reached significance between placebo and active doses. Effects were demonstrable for 6 to 12 hours. The visual analog scale "How much do you dislike the drug?" was not sensitive to the effects of smoked marijuana; there was no distinction in responses of the subjects between the placebo and active drug conditions.
Single Dose Questionnaire. The "feel drug" and "like drug" components of the Single Dose Questionnaire were sensitive indicators of the effects of marijuana; peak responses were noted at the first measurements at 31 minutes for both low and high doses. The score for subject-reported drug "liking" is illustrated in Fig. 4, b. Effects approached baseline levels at 12 hours after drug exposure. A small placebo effect was evident, but both "feel drug" and "like drug" responses were rated significantly greater than placebo ("feel drug" [F,,,, = 6.75 0, < 0.014)]; "like drug" [F,,,, = 5.82 @ < 0.035)]. Addiction Research Center Inventory. Significant increases on the PCAG subscale were noted between the placebo and 3.55% A-9-tetrahydrocannabinol conditions [F,,,, = 4.38 @ < 0.043)]. Mean peak sedation was reported at 3 hours (Fig. 4, c) after the lowdose cigarettes and 6 hours (data not presented) after the high-dose cigarettes, with effects diminishing beyond 12 hours. The LSD subscale illustrated mean
CLlN PHARMACOL THER TULY 1992
38 Huestis e t al.
1
a) Feel Drug Placebo
c) PCAG
r4
Hours Fig. 4. Mean changes (N = 6) in subjective effects after smoking of a single marijuana cigarette. Subject-reported "feel drug" (a) was obtained from the visual analog scale (VAS) for "How much drug effect do you feel?'Subject-reported "liking" (b) was determined from the Single Dose Questionnaire. The pentobarbital-chlorpromazine-alcohol (PCAG) group (c) and the lysergic acid diethylamide (LSD) group (d) scores were obtained from the Addiction Research Center Inventory.
peak dysphoric effects at 11/2 hours after smoking (Fig. 4, 4, although effects were not significantly different than placebo. The MBG subscale was not a sensitive indicator of the effects of marijuana; no significant differences were noted after smoking of placebo, 1.75% or 3.55% A-9-tetrahydrocannabinol cigarettes.
Performance Walter Reed performance battery. Three different parameters were assessed in each performance task; percentage of correct responses, throughput, and speed. None of the five tasks showed significant effects on throughput and speed after marijuana smoking. After smoking one 1.75% or 3.55% A-9-tetrahydrocannabinol cigarette, the percentage of correct
scores on the logical reasoning task were significantly lower [F,,,, = 5.00 (p < 0.031)] than after placebo (Fig. 5, a). The mean peak decrease in accuracy for the logical reasoning task was observed at the time of first measurement at 22 minutes. Mean peak decreases of 14.1% -+ 10.8% and 12.5% + 18.0% were found after smoking one 1.75% or 3.55% A-9-tetrahydrocannabinol cigarette. Significant decreases in accuracy were observed between 22 and 105 minutes. The percent accuracy on the logical reasoning task had returned to baseline levels within 3 hours for all subjects (Fig. 2, 4. No significant differences in accuracy after marijuana smoking were noted in the matrix, serial addition and subtraction, manikin, and time wall performance tasks. In the time wall task, subjects had to estimate the completion of a 10-second time span. As
VOLUME 52 NUMBER 1
Absorption of s m k e d marijuana 39
loo
- a) Logical reasoning
loo- b) DSST
90 -
80-
+ 1.75% THC
10.8 7
,d) Manikin
c) Timewall
10.6 10.4 10.2
I @-
-
a
50-
10.0 -
9.8 -1
1
0
1
2
3
40, -1
1
0
1
2
3
Hours Fig. 5. Effects on mean performance measures (N = 6) after smoking of a single marijuana cigarette. Tasks included the logical reasoning (a), time wall (c), and manikin (d) tasks adapted from
the Walter Reed Performance Assessment Battery and the Digit Symbol Substitution Test (DSST) (b). The percentage of correct responses is illustrated for the logical reasoning and DSST tasks. The ability of subjects to estimate a 10-second time span is illustrated in the time wall task, and response speed is shown for the manikin task.
seen in Fig. 5, c, no significant differences in estimated time were observed after marijuana smoking. Response speed also was evaluated for all tasks and was not affected by marijuana administration. An example of the response speed for the manikin task is illustrated in Fig. 5, d. DSST. No significant differences in accuracy or response speed were noted in the DSST after smoking placebo or low- or high-dose marijuana cigarettes. Fig. 5, b, illustrates the percentage of correct scores for the DSST task after smoking a single marijuana cigarette.
DISCUSSION The first inhalation of marijuana smoke produced detectable blood levels of A-9-tetrahydrocannabinol, with peak concentrations occurring shortly before the cigarette had been completely smoked. Integrated
blood samples collected over a 60-second period after the first puff of a low- or high-dose marijuana cigarette (1.75% or 3.55% A-9-tetrahydrocannabinol) contained 7.0 8.1 and 18.0 12.0 nglml of A-9tetrahydrocannabinol, respectively. Concentrations increased rapidly with each additional puff. Halfway through smoking, after four puffs, levels of 41.0 10.4 and 102.7 + 39.1 nglml were attained after the low and high doses. Mean peak drug levels of 79.0 ? 25.2 and 152.0 ? 86.3 nglml occurred at 9 minutes, just before the last inhalation. Because of the rapidity of changes occurring during smoking, few other studies have attempted blood collection during the smoking process. Perez-Reyes et al.7.9 were the first to note that peak A-9-tetrahydrocannabinol concentrations developed before the end of smoking. In their studies, subjects determined their own rate and speed of smoking, inhaling many puffs (28 to 30) over 15 to 19 min-
*
*
*
40 Huestis et al. utes. One proposed explanation for the occurrence of peak levels before cessation of smoking was that the peak levels could be an artifact of the sampling protocoil6; that is, random sampling over a large number of individual peaks and valleys after each puff could have produced artificial peak responses. In the present study, continuous blood sampling eliminated the possibility of a sampling artifact and documented clearly that peak A-9-tetrahydrocannabinol levels occurred during marijuana smoking. It is apparent, however, that the time to peak blood levels is greatly affected by the individual smoking protocol; the number of puffs, time between puffs, and volume and length of inhalations can have substantial impact on drug delivery. Two additional explanations have been proposed for the occurrence of peak levels of A-9-tetrahydrocannabinol before the end of ~ r n o k i n gHot . ~ air extraction of drug could possibly deliver higher levels during the first few puffs. As the subject inhales and draws hot smoke through the marijuana cigarette, the A-9-tetrahydrocannabinol could be extracted differentially out of the plant material. However, preliminary studies did not support this explanation. The amount of active drug in the terminal portion of unsmoked and smoked marijuana cigarettes was measured and the amounts were found to be similar. Although this suggests that smoking does not deplete A-9-tetrahydrocannabinol content, a more critical experiment would be quantitative analysis of A-9-tetrahydrocannabinol in the mainstream smoke of a succession of individual puffs. Another explanation for the early peaking of A-9tetrahydrocannabinol levels during smoking could be that subjects do not uniformly smoke cigarettes from beginning to end. The smoking dynamics of subjects have been evaluated throughout test sessions and no significant differences in the number of puffs, hold time, or time between puffs were noted.7 A critical factor that was not reported was the inhalation volume of each puff. If the subject breathed deeper during the early puffs, more drug presumably would be delivered. These early differences in smoking dynamics also could account for peak drug levels during marijuana smoking. This would be consistent with findings of tobacco cigarette smoking, which confirm the tendencies of smokers to self-administer decreasing amounts of tobacco smoke across successive puffs on a cigarette. l 7 An additional finding from the current data was the similarity between time-to-peak drug levels and timeto-peak effects for some measurements. Previous studies have emphasized time delays between peak drug
C I J N PHARMACOL THEK JULY 1992
levels and peak physiologic and behavioral effects. One of the most sensitive indicators of marijuana exposure, subjective "high," has been reported to peak 30 and 60 minutes after ~ m o k i n g . ~ The . ' ~ present study design incorporated measurements of subjective effects after each marijuana inhalation to monitor the rapid onset of effects. Three of six subjects reported increases in drug "liking" scores after the first puff, 15 seconds after smoking began, and all subjects reported increases by the second puff of a high-dose cigarette. Maximum drug "liking" occurred at 8.4 minutes (six puffs) and 10.2 minutes (eight puffs) during smoking of a low- or high-dose cigarette, respectively. A similar pattern of effects was observed with the "feel drug" visual analog scale. Subjects responded rapidly with three of six subjects recording increases at 15 seconds. Four inhalations were necessary for all subjects to report increased "feel" drug scores with peak effects noted after 13.2 and 16.2 minutes for the low and high doses, respectively. Previous reports of subjective "high" have indicated that peak subjective effects appeared between 5 and 20 minutes after the end of Marijuana exposure also produced rapid changes in some physiologic effects. Continuous electrocardiogram monitoring documented increased heart rate in all subjects soon after the start of smoking. Maximum heart rates were recorded within 4 minutes of the last inhalation after the high-dose cigarette. Previous reports also showed peak heart rate increases between 5 and 10 minutes after smoking c e s ~ a t i o n . In ~ , addi~~ tion, increased diastolic blood pressure was recorded for four of six subjects after two puffs of the highdose cigarette. Peak responses were recorded within 18 minutes after the start of smoking. The understanding of the pharmacology of marijuana requires careful observation of the onset, peak, and duration of significant physiologic, behavioral, and biochemical effects. Capture of rapid, fleeting changes or more delayed, persistent effects is highly dependent on the choice of experimental methods and the frequency of testing. In the present study, performance data was not collected until 22 minutes into the complex time line. The time required for administration of the six performance tasks prevented earlier scheduled testing. Maximal decreases in percent correct responses for the logical reasoning task already had been achieved at the first measurement. Consequently, the onset and time of peak effects could not be determined. Early decrements in performance on a variety of tasks 5 to 15 minutes after the end of smoking have been previously reported.'0321Obviously,
VOLUME 52 NUIMRER 1
drug effects should be monitored at the time of drug delivery to active sites, that is, during drug absorption, because of the fast delivery of drug and rapid onset of effects during marijuana smoking. Studies that combine rapid blood collection, paced smoking protocols, and timely collection of physiologic and behavioral measures are essential for the complete characterization of the absorption phase of marijuana smoking. We gratefully acknowledge the support of Dakmed Corporation, Buffalo, N.Y., for use of the ambulatory withdrawal pump, the Center for Human Toxicology for performance of blood analyses for A-9-tetrahydrocannabinol, and Dr. Stephen Heishman for technical assistance with the Walter Reed Performance Assessment Battery.
References National Institute on Drug Abuse. National household survey on drug abuse: Highlights 1990. Washington: US Government Printing Office; 1991; DHHS pub no (ADM)91-1789. Johnstone LD, O'Malley PM, and Bachman JG. Drug use among American high school seniors, college students and young adults, 1975-1990. Washington: US Government Printing Office; 1991; DHHS pub no (ADM)91-1813. Lindgren JE, Ohlsson A, Agurell S , Hollister L, Gillespie H. Clinical effects and plasma levels of THC in heavy and light users. Psychopharmacology 1981; 74:2O8- 12. Ohlsson A, Lindgren JE, Wahlen A , Agurell S, Hollister LE, Gillespie HK. Plasma delta-9-tetrahydrocannabin01 concentrations and clinical effects after oral and intravenous administration and smoking. CLINPHARMACOL THERl980;28:409- 16. Hollister LE, Gillespie HK, Ohlsson A, Lindgren JE, Wahlen A, Agurell S. Do plasma concentrations of delta-9-tetrahydrocannabinol reflect the degree of intoxication? J Clin Pharmacol I98 1;2 1: 17 1S-7s. Benowitz N. Clinical pharmacology of inhaled drugs of abuse: implications in understanding nicotine dependence. In: Chiang CN, Hawks RL, eds. Research findings on smoking of abused substances. Washington: NIDA Res Monogr 99, 1990. Perez-Reyes M, Di Guiseppi S, Davis KH, Schindler V, Cook CE. Comparison of effects of marihuana cigarettes of three different potencies. CLINPHARMACOL THER 1982;31:617-24. Davis KH, McDaniel IA, Caddel LW, Moody PL. Some smoking characteristics of marijuana cigarettes. In: Agurell S, Dewey WL, Willette RE, eds. The cannabinoids: chemical, pharmacologic, and therapeutic aspects. New York: Academic Press, 1984:97-109.
Absorption of smoked mavijuana 4 1 9. Perez-Reyes M, Owens SM, Di Guiseppi S. The clinical pharmacology and dynamics of marihuana cigarette smoking. J Clin Pharmacol 1981;21:201S-7s. 10. Barnett G, Licko V, Thompson T. Behavioral pharmacokinetics of marijuana. Psychopharmacology 1985; 8551-6. 11. Yesavage JA, Leirer VO, Denari M, Hollister LE. Carry-over effects of marijuana intoxication on aircraft pilot performance: a preliminary report. Am J Psychiatry 1985;142: 1325-9. 12. Foltz RL, McGinnis KM, Chinn DM. Quantitative measurement of A-9-tetrahydrocannabinol and two major metabolites in physiological specimens using capillary column gas chromatography negative ion chemical ionization mass spectrometry. Biomed Mass Spectrometry 1983;10:316-23. 13. Haertzen CA, Hickey JE. Addiction Research Center Inventory (ARCI): measurement of euphoria and other drug effects. In: Bozarth MA, ed. Methods of assessing the reinforcing properties of abused drugs. Heidelberg: Springer-Verlag, 1987:489-524. 14. Thorne DR, Censer SG, Sing HC, Hegge FW. The Walter Reed Performance Assessment Battery. Neurobehav Toxic01 Teratol 1985;7:4 15-8. 15. McLeod DR, Griffiths RR, Bigelow GE, Yingling J. Computer technology: an automated version of the digit symbol substitution test (DSST). Behav Res Methods Instrumentation 1982;14:463-6. 16. Barnett G, Chiang CN, Perez-Reyes M, Owens SM. Kinetic study of smoking marijuana. J Pharmacokinet Biopharm 1982;10:495-506. 17. Griffiths RR, Henningfield JE. Experimental analysis of human cigarette smoking behavior. Fed Proc 1982; 41 :234-40. 18. Lemberger L, Weiss JL, Watanabe AM, Galanter IM, Wyatt RJ, Cardon PV. Delta-9-tetrahydrocannabinol: temporal correlation of the psychologic effects and blood levels after various routes of administration. N Engl J Med 1972;286:685-8. 19. Cocchetto DM, Owens SM, Perez-Reyes M, DiGuiseppi S, Miller LL. Relationship between plasma delta9-tetrahydrocannabinol concentration and pharmacologic effects in man. Psychopharmacology 1981 ;75: 158-64. 20. Agurell S, Levander S, Binder M, et al. Pharrnacokinetics of delta-8-tetrahydrocannabinol (delta-6-tetrahydrocannabinol) in man after smoking: relations to physiological and psychological effects. In: Braude MC. Szara S, eds. The pharmacology of marijuana. New York: Raven Press, 1976:49-6 1 . 21. Heishman SJ, Stitzer M, Yingling JE. Effects of tetrahydrocannabinol content on marijuana smoking behavior, subjective reports, and performance. Pharmacol Biochem Behav 1989;34:173-9.
