0 145-6008/92/1604-0769$3.00/0 ALCOHOLISM: CLINICAL A N D E X P E R l M E N l A L RESEARCH

Vol. 16, No. 4 July/August 1992

Interaction between Ethanol and Calcium Channel Blockers in Humans Mario Perez-Reyes, W. Reid White, and Robert E. Hicks

The purpose of this study was to test the hypothesis that pretreatment with Ca2+channel blockers would antagonize the effects of ethanol intoxication in humans. The Ca2+channel blockers verapamil and nifedipine were chosen because preclinical research has shown them to decrease certain behavioral effects of ethanol in animals. Sixteen healthy, male, paid volunteers, moderate users of ethanol, participated in the study (six in the verapamil and 10 in the nifedipine paradigms). Gelatin capsules containing verapamil (80 mg, 160 mg, or placebo)were administered orally 90 min before ethanol ingestion; whereas, gelatin capsules containing nifedipine (10 mg, 20 mg, or placebo)were administered 30 min before ethanol ingestion. Ethanol (0.85 g/kg or placebo) was administered over a 30-min interval. Subjects were tested in a single-blind, latin-square, cross-over design with each of the following six conditions: placebo ethanolplacebo blocker, placebo ethanol-low dose blocker, placebo ethanolhigh dose blocker, ethanol-placebo blocker, ethanol-low dose blocker, and ethanol-high dose blocker. The variables measured in this study were subjective rating of ethanol intoxication, Addiction Research Center Inventory alcohol scale, heart rate, blood pressure, short-term memory, accuracy and latency of response in the Simulator Evaluation of Drug Impairment task, and blood ethanol concentrations by breath analyzer. Results indicate that pretreatment with either verapamil or nifedipine failed to antagonize the inebriating effects of ethanol including its decremental effects on short-term memory and psychomotor performance. Ethanol, Calcium Channel Blockers, Verapamil, Nifedipine, Drug Interaction.

HE PRECISE NEUROCHEMICAL mechanisms by T which ethanol produces its intoxicating and reinforcing effects are unknown. Recent preclinical studies have shown that Ca2+transport appears to be involved in the mediation of ethanol effects on the brain. Engel et al.,' reported that the intragastric administration to rats of the Ca2+channel blocker nifedipine antagonized the locomotor effects of ethanol and reduced ethanol preference in a free-choice paradigm. Rezvani et a1.' have shown that the Ca2+channel blocker verapamil, when injected intracerebrally to rats, attenuated the behavioral effects of ethanolinduced intoxication and hypothermia. In rats bred to prefer ethanol3 and in ethanol-preferring macaque monFrom the Department of Psychiatry, University of North Cuvolina ,%hoof of Medicine, Chapel Hill, North Carolina. Received for publication November 13, 1991; accepted January 28, I992

This work was supported by Grant 9108 ,from che North Carolina . 4 k ~ h ) l i s mResearch Authority and Grant DA04484 from the National In.sii/ute on Drug Abuse. Reprint rclquests: Mario Perez-Reyes, M.D., CB #7175, Medical Rexwrch Bldg A , Depariment of Psychiatry, University of North Carolina, School of Medicine, Chapd Hill, NC 27599-7I 75. Copvrighr 0 1992 by The Research Society on Alcoholism. .4koIiol Clm Ex11 Res. Vol 16, No 4, 1992: pp 769-175

keys: subcutaneous injection of verapamil has been demonstrated to significantly decrease ethanol consumption. These observations support the hypothesis that Ca2+transport across neuronal membranes at specific receptor sites in the brain mediates the intoxicating and reinforcing effects of ethanol, either directly or by modulating the storage and release of neurotransmitters, i.e., ~erotonin.~ This neurotransmitter appears to be involved in the ethanol preference of selectively bred ethanol-prefemng rats.' Because there are species differences in the mechanisms of drug action, these results cannot be assumed to apply to humans until verified by clinical experimentation. This report contains the results of placebo-controlled, doseresponse studies designed to investigate the possible antagonism of the typical subjective, psychomotor, and recent memory effects of ethanol by the Ca2+channel blockers nifedipine and verapamil in human subjects. MATERIAL AND METHODS SUbJeClS Sixteen healthy, paid, male volunteers who use ethanol in moderate amounts participated in the study (10 in the nifedipine and six in the verapamil paradigms). The health status of subjects was determined by medical and psychiatric history, and complete physical, including electrocardiographic, examination. None had a family history of alcoholism. All were fully informed about the purpose, experimental procedures, and risks of the study. All signed a consent form approved by the Committee on the Protection of the Rights of Human Subjects of the University of North Carolina at Chapel Hill. Demographic characteristics (mean -t SEM) were age 24.9 t 1.4 years, weight 76.7 + 1.8 kg, height 182.9 f 2.9 cm, and ethanol consumption 16.7 + 2.5 g/kg/month. Quantity-frequency estimates of ethanol consumption were derived from a six-item questionnaire adapted from Khavari and Farber.'

Experimental Design Subjects were randomly assigned to treatment-sequence orders in a single-blind, latin-square, cross-over paradigm. The six treatment conditions were placebo ethanol/placebo blocker, placebo ethanol/low dose blocker, placebo ethanol/high dose blocker, ethanol/placebo blocker. ethanol/low dose blocker, and ethanol/high dose blocker. The volunteers were tested at weekly intervals. The subjects contracted to abstain from the use of ethanol and other psychoactive drugs for 3 days before each experimental session. Driig Administration A single dose of ethanol was used in both studies: 2 ml of 100 proof vodka/kg (0.85 g/kg ethanol). The total do.se was diluted to a volume of 300 ml with orange juice and administered in 50-ml doses at 5-min

769

770

intcrvals for 30 min. For the placebo ethanol condition, the subjects were asked to drink, in divided doses, 300 ml of diluted orange juice sprayed with vodka to mimic the aroma of a mixed drink. Gelatin capsules containing either 80 or 160 mg of verapamil (Calan) o r gelatin capsules containing a small amount of metamucil (placebo) were administered orally 90 rnin before the beginning of ethanol ingestion. This interval was selected because peak plasma concentrations of verapamil have been reported to occur between 1 and 2 hr after oral administration of a single dose.* Gelatin capsules containing either 10 or 20 mg of nifedipine (Procardia) or gelatin capsules containinga small amount of metamucil (placebo) were administered orally 30 rnin before the beginning of ethanol ingestion. This interval was selected because peak plasma concentrations of nifedipine have been reported to occur between 30 and 60 min after oral administration of a single dose.’ In order to ensure uniform conditions, drugs were administered after an overnight fast.

I:ji;ucrimental Variables Subjective Ruting of Intoxication. To measure this effect, the subjects were asked to rate their degree of “drunkenness.” A scale of 0 to 100 was used, with 0 representing no ethanol effects, and 100 representing the “drunkest” subjects had ever been. Every time that a rating was to be made, the subjects were given their previous rating to use for comparison. This technique allowed the subjects to rate themselves as experiencing relatively more, less, or the same effects as those rated in the previous time interval. Experiments have shown this method of evaluating subjective drug effects over time to be efficient and informative, and to generate curves that parallel physiologic effects, e g , the cardiovascular effects of marihuana and cocaine.” In addition to this global rating of ethanol intoxication, the subjects were asked to respond to the items of the Addiction Research Center Inventory (ARCI) alcohol scale before and at 60, 105, and 150 rnin after the beginning of ethanol consumption. The ARC1 is a widely used 550-item multipurpose test developed in 1958” and validated for measuring the broad range of physical, emotive, cognitive, and subjective effects of drugs.” Cardiovascular Parameters. The heart rate and blood pressure were measured, at 10-min intervals, 20 rnin before (baseline) and for 6 hr after Ca” channel blocker or placebo administration. For these measurements, an Accutracker (Suntech Medical Instruments) was used. This portable instrument is accurate and allows ambulatory measurement of heart rate and blood pressure at any specified interval (i.e., every 2 rnin or longer intervals). Short-Term Memory. To measure this parameter the Word Recall test was used. Twenty-four words taken from the University of Toronto Categorized Word Pool, divided into six categories of four related words, were presented’ in succession. Individual words were presented on the screen of a desk computer for 2 sec, with a 2-sec interval between words. After all the words were presented, subjects were asked to type every word that they could recall within the next 5 min. This test was administered before and at 30, 60, 120, and 180 rnin after the beginning of

PEREZ-REYES ET AL.

ethanol ingestion. On each occasion, a novel list of words was used. The number of words recalled is a measure of short-term memory. Psychomotor Performance. The test instrument was the Simulator Evaluation of Drug Impairment (SEDI) instrument developed by Mills and Bisgrove,I3 modified by addition of a foot re~ponse.’~ The test apparatus consisted of an IBM 286 desk computer, a 4-button response panel, three 2-cm high 7segment numeric displays mounted in plastic cases (Radio Shack), and a foot pedal. The stimulus displays were arranged on the arc of a circle with a radius of 52.0 cm and were approximately 86.0 cm from the subject’s face. The response panel was positioned midway between the displays and the subject. The four response buttons were 5.0 cm apart and were operated with both hands. The foot response pedal was located under the table supporting the apparatus within easy reach of the subject’s foot. Subjects were required to monitor a display in the central field of vision and the peripheral displays located 34 degrees (or 30 cm) from the central display. The central display was programmed to present 2-digit numbers. If the central display exceeded a critical high value (57), the subjects were instructed to press button 3. lfthese numbers fell below a critical low value (53), then the subject was to press button 2. The peripheral displays were programmed to present 1 of 4 digits (3, 4, 5, and 7) that changed intermittently. If either the left or right peripheral display changed from the steady value of 4 to a 5, the subject was instructed not to respond; if the value changed to a 3, the subject was to press a corresponding left (1) or right (4) response button; or if the value changed to a 7 in either peripheral display, the subject was to press the foot pedal. The results of each test were displayed by the computer for immediate visual inspection, and simultaneously stored. The subjects were allowed up to 2 sec to respond to centrally presented stimuli and 1.5 sec to respond to peripherally presented stimuli. Trials that required a response were always separated by at least two trials that did not require a response. During each 2-min test session, the subjects were presented with an average of 156 stimuli. On the average, subjects received 62 trials that required a response during each 2-min test ( 16 on the central and 19 on each of the peripheral displays requiring a manual response, plus four on each of the peripheral displays requiring a pedal response). Alternative forms of the stimulus sequence were randomly presented on successive tests to minimize the likelihood that stimulus response patterns could be memorized. The following measures were obtained from the SEDI task: (1) the percentage of correct high and low responses for central stimuli with mean response latency, (2) the percentage of correct manual responses for the left and right peripheral displays with mean response latencies, (3) the percentage of correct pedal responses for the peripheral

77 I

ETHANOL AND CALCIUM CHANNEL BLOCKER INTERACTION

displays with mean response latency, and (4) the number of false-positive responses (manual and pedal). All mean response latencies were reported in milliseconds. Optimal and stable performance on this complex task required an extensive period of training prior to testing for the effects of central nervous system-acting drugs, in order to assure that the results obtained are reliable and reproducible. For this purpose, the volunteers who participated in the study were trained until their level of performance was consistently high. Subjects were asked to perform the SEDI task two times, separated by a I-min interval, at 60, 30, and 5 min before drug administration, and at 60,90, 120,240, and 360 min following the beginning of ethanol administration. Multiple repetitions of this complex task over a 7-hr period posed the dual problems of fatigue and lack of motivation. The volunteers were rewarded for their performance (subjects were paid $3.00 for every time point that they executed the SEDI task within 5% of their baseline scores) in an attempt to counteract these problems. Determination of Blood Ethanol Concentration. An Alco-sensor I11 (Intoximeters, Inc) breath analyzer was used to estimate blood ethanol concentration. This instrument was calibrated at appropriate intervals with standards provided by the manufacturer in order to assure accuracy. Breath ethanol readings were taken at baseline, at 45 and 60 min after the beginning of ethanol consumption, and at 30-min intervals thereafter. Statistical Analysis. The data from the nifedipine and verapamil studies were subjected to identical statistical analyses. The analysis of variance model was completely crossed with Subject, Ethanol, Nifedipine (or Verapamil) Dose, and Time of measurement as factors. The effects of Time and the Dose of calcium channel blockers and all their constituent interactions were decomposed using orthogonal polynomial regression analyses.

*-• ETOH/PLACEBO

A-

A

A-A

O’”T

0.00 P 0

ETOH/LOW VERAPAMIL

ETOH/PLACEBO

ETOH/LOW NlFEDlPlNE

I

30

60

90

120

150

180

TIME (Min) Fig. 1. Mean blood ethanol levels as estimated from breath samples. Ethanol was administered in divided doses from 0 to 30 min.

constituent interactions produced no significant effects ( p = 0.6). Nifedipine. In comparison with Placebo Ethanol, Ethanol produced significant mean ratings of intoxication [F(1,9) = 42.69, p = 0.0001], and this effect is time-related [F(11,99) = 34.07, p < 0.00011. Nifedipine and all its constituent interactions are nonsignificant ( p >_ 0.25).

ARCI Alcohol Scale Verapamil Efects. In comparison to Placebo Ethanol, Blood Ethanol Concentrations (Fig. 1) Ethanol produced significantly elevated mean ARCI Verapamil. Mean blood ethanol concentrations varied scores [F(1,5) = 135.94, p = 0.00011, and this ethanol across measurement times [Ethanol X Time; F(8,40) = effect is time-related [F(3,15) = 19.45, p < 0.00011. Ver16 I .28, p < 0.00011. Verapamil did not influence this time apamil’s main effect is nonsignificant as are all of its constituent interactions ( p z 0.6). related effect ( p = 0. I). Nqedipine Effects. In comparison to Placebo Ethanol, Nifedipine. Mean blood ethanol concentrations varied Ethanol produced significantly elevated mean ARCI across measurement times [Ethanol x Time; F(8,72) = scores [F( 1,9) = 58.56, p < 0.00011, and the ethanol effect 108.20, p < 0.00011. Nifedipine did not influence this is time-related [F(3,27) = 7.57, p = 0.00081. No effect time related effect ( p = 0.09). involving nifedipine reached statistical significance ( p 2 0.2). Subjective Rating of Drunkenness (Fig. 2) Cardiovascular Effects. Averaged across measurement Verapamil. In comparison with Placebo Ethanol, times, verapamil and nifedipine per se produced no sigEthanol produced significant mean ratings of intoxication nificant change in heart rate, systolic and diastolic blood [F(1 3 ) = 24.01, p = 0.0051, and this effect is time-related pressure. The interaction of verapamil or nifedipine with [F(I1,55) = 11.82, p < 0.00011. Verapamil and all of its ethanol, on these parameters, is also not significant. RESULTS

PEREZ-REYES ET AL.

772

100-

-0 ETOH/PVICEBO

80--

A-

I-

z W 0

t Y

W

Q

--

60

A

FTOH/LOW M W A M l L

9-9FT0H/H‘GHMWAM’L

:;--/:““,\f%*‘ ‘A=fZ!=

0.

1

performance accuracy (% correct) decreased with ethanol as compared to placebo-ethanol [F(1,5)= 17.33, p = 0.0091.Also averaged across measurement times, verapamil did not significantly influence mean performance accuracy ( p = 0.2),nor did it modify ethanol’s effect ( p = 0.5).The ethanol effect is time-related [Ethanol X Time; F(4,20)= 3.77,p = 0.0191.Verapamil produced no timerelated effect ( p = 0.7),nor did it modify ethanol’s timerelated effect ( p = 0.9). Reaction Time. Averaged across measurement times, mean reaction time (milliseconds) was greater in the

L-

z

W

0

cf W

Q

Nifedipine Effects Accuracy. Averaged over measurement times, mean TIME (Min) Fig. 2. Mean subjective rating of drunkenness following the administration oi ethanol from 0 to 30 min.

performance was less accurate in the ethanol condition than in the placebo-ethanol condition [F(1,9)= 34.42,p = 0.0002].

Also averaged across measurement times, nifedipine did not significantly influence mean performance accuracy ( p Short Term Memory (Word Recall). The results of this = 0.7),nor did it modify ethanol’s effect ( p = 0.13).The task are illustrated in Fig. 3. ethanol effect is time related [Ethanol x Time; F(4,36)= Verapamil. Ethanol significantly decreased the mean 5.45,p = 0.0021.Nifedipine produced no time-related number of words correctly recalled [F(1,5)= 43.53, p = effect ( p = 0.8), nor did it modify ethanol’s time-related 0.0011.This effect is primarily biphasic over time, i.e., a effect ( p = 0.4). Reaction Time. Averaged across measurement times, decrement followed by partial recovery during the time of F( 1,5) = 20.82,p = ethanol resulted in greater mean reaction time than did observation [Ethanol X Timequadratic; 0.0061. Verapamil did not significantly influence the placebo-ethanol [F(1,9)= 9.66,p = 0.0131. Also averaged ethanol effects ( p 2 0. l),nor did verapamil have any main across measurement times, nifedipine did not significantly influence mean reaction time ( p = 0.7), nor did it modify effect ( p L 0.9). Nifedipine. Ethanol significantly decreased the mean ethanol’s effect ( p = 0.5).The ethanol effect is time related number of words correctly recalled [F(1,9)= 53.44,p < [F(4,36)= 7.82,p = 0.00011.Nifedipine produced no 0.00011.This effect is primarily biphasic over time, i.e., a time-related effect ( p = 0.7), nor did it modify ethanol’s decrement followed by partial recovery during the time of time-related effect ( p = 0.3). F(1,9)= 10.91,p = observation [Ethanol X Timequadratic; 0.0091. Nifedipine did not significantly influence the DISCUSSION ethanol effects ( p = 0.15),nor did nifedipine have any main effect ( p = 0.8). This study was designed to test the hypothesis that Psychomotor Performance (SEDI Task). The stimuius- pretreatment with the Ca” channel blockers nifedipine or response ensemble (central, peripheral, and pedal) of the verapamil would antagonize the typical and clinically SEDI task was added as a crossed factor in the statistical important effects of ethanol in humans, i.e., inebriation model. The results of this task are illustrated in Figs. 4 including decrements in short-term memory and psychoand 5. motor performance. Verapamil and nifedipine were chosen because research has shown them to significantly Verapamil Efects decrease the behavioral effects of ethanol in animals.’-4 The experimental design permitted measurement of the Accuracy. Averaged across measurement times, mean effects of placebo, nifedipine, or verapamil per se or in

173

ETHANOL AND CALCIUM CHANNEL BLOCKER INTERACTION

NlFEDlPlNE EFFECTS

VERAPAMIL EFFECTS 20 T

0 W I I :

14

0 In

12

3

,,t

2OT 0-0

PUCEBO/PUCEBO

A-A

LOW NIFEOIPINE/PUCEBO

0-OPUCEEO/PLACEBO

E

6

0

30

90

' I

120

150

0

180

30

120

90

60

150

180

INTERACTIVE EFFECTS

INTERACTIVE EFFECTS 20 0-0

4

PUCEBO/WOL

A-ALOW 0-OHIGH

MwAL(IL/ETwOL

b-0

ETOH/PLACEEO

a-I

ETOH/LOW NIFEDIPINE

MRU'AUIL/ETHANOL

8 --

64 0

I

30

90

60

120

150

I

180

0

60

30

TIME (Min)

120

90

150

180

TIME (Min)

Fig. 3. Short Term Memory (Word Recall Test)-verapamil and nifedipine effects. Neither drug produced significant effects on the number of words recalled or altered the significant decrease produced by ethanol.

NlFEDlPlNE EFFECTS 770 -750

__

0-0

PLACEBO/PIACEBO

0-0

A-A

LOWVERAPAMIL/PLACEBO

A-A

0-0

670 650

630 0

30

790 -

60 90 120 150 INTERACTIVE EFFECTS b-b

180

210

cc2

PLACEBO/PLACEBO LOW NIFEDIPINE/PLACEBO HIGH NIFEDIPiNE/PLACEBO

: 30

120

90

150

180

210

INTERACTIVE EFFECTS

T

PLACEBO/€fH"L

I

60

750 730 710.-

690 - -

0

-

A-A

m--.

0

30

60

90

120

TIME (Min) Fig. 4. Psychomotor Performance (SEDI)-mean

significant decrement produced by ethanol.

150

PLACEBO/ETHANOL LOWNIFEDIPINE/ETHANOL HIGH NIFEDIPINE/ETHANOL

I

180

210

0

30

60

90

120

150

180

I 210

TIME (Min)

effects on accuracy of response. Neither drug significantly altered the mean accuracy of response or changed the

774

PEREZ-REYES ET AL

VERAPAMIL EFFECTS

i W

651 75

N IFEDl PIN E EFFECTS

0-0

PLACEBO/PLACEBO

0-0

PLACEBO/PLACEEO

t--b

LOW VEWAMIL/PIACEBO

&-A

LOW NIFEDIPINE/PLACEBO

0-0

HIGH MWAMIL/PLACEBO

0-0

HIGH NIFEDIPINE/PLACEBO

*-• PLACEEO/ETHANOL

L

T

A-A

LOW VEi"AMIL/ETHANOL

combination with ethanol on the dependent variables measured. In drug interaction studies it is essential that the effects produced by the dose of the reference drug are not of such a magnitude (either too small or too large) as to preclude elucidation of the effects of the companion drug. Therefore, to maximize detection of interactive effects, 0.85 mg/ kg ethanol was selected to produce moderate levels of inebriation with moderate decrements in short-term memory and psychomotor performance. The results indicate that this aim was achieved. Thus, the dose of ethanol used, in comparison to placebo ethanol, altered these parameters consistently but not excessively. For safety reasons, commonly prescribed therapeutic doses of verapamil and nifedipine were used. These drugs were administered in the laboratory at a time when, according to the literature, their maximal plasma concentrations would occur before the period of ethanol ingestion. However, because the plasma concentrations of nifedipine and verapamil were not measured, it is not known whether this aim was achieved. Alternative strategies would have been to administer larger single doses or to dose the subjects with these agents for several days until, theoretically, steady-state drug concentrations were obtained before the administration of ethanol. Although scientifically more rigorous, these strategies would have posed unwarranted risks to the volunteers (i.e., hypotension) and, as a consequence, neither of these strategies were used. This concern was supported by the observation that one of the volunteers who received the high dose of nifedipine, in

A-A

PLACEBO/ETWOL LOW NIFEDIPINE/ETHANOL

the first testing session, developed a severe postural hypotensive episode and was, therefore, excluded from further testing. Ethanol produced blood ethanol concentrations at the legally impaired level in North Carolina (0.1 g/dl), significant ratings of inebriation (both on global and ARC1 scores), minimal cardiovascular effects, and decrements in short-term memory and psychomotor performance. Neither verapamil nor nifedipine per se produced any subjective effects nor alterations of short-term memory and psychomotor performance. In the normotensive subjects tested verapamil and nifedipine produced minimal and clinically unimportant changes in heart rate, systolic, and diastolic blood pressure. Of major importance, neither verapamil nor nifedipine at the doses used, significantly altered blood ethanol concentrations, or the effects of ethanol on subjective ratings of inebriation, short-term memory and psychomotor performance. However, it is possible that larger doses of these drugs may alter the effects of ethanol. For the reasons discussed above, this strategy was not chosen. In summary, pretreatment with the Ca2+channel blockers verapamil or nifedipine, under the experimental conditions used, did not produce significant alteration of the inebriating effects of ethanol or its decremental effects on short-term memory and psychomotor performance. ACKNOWLEDGMENTS We thank Susan McDonald for her technical assistance.

ETHANOL AND CALCIUM CHANNEL BLOCKER INTERACTION

REFERENCES 1. Engel JA, Fahlke C, Hulthe P, et al: Biochemical and behavioral evidence for an interaction between ethanol and calcium channel antagonists. J Neural Transm 74:181-193, 1988 2. Rezvani AH, Mack CM, DeLacy PA, Janowsky DS: Verapamil effects on physiological and behavioral responses to ethanol in rat. Alcohol Alcohol 25:51-58, 1990 3. Rezvani AH, Janowsky DS: Decreased alcohol consumption by verapamil in alcohol preferring rats. Prog Neuro-Psychopharmacol Biol Psychiatry 14:623-63 I , 1990 4. Rezvani AH, Grady DR, Janowsky DS: Effect of calcium-channel blockers on alcohol consumption in alcohol-drinking monkeys. Alcohol Alcohol 26:161-167, 1991 5. Hill NS, Lee SL, Fanburg BL: Effect of calcium channel blockers on serotonin uptake. Proc SOCExp Biol Med 24:326-330, 1990 6. McBride WJ, Murphy JM, Lumeng L, Li TK: Serotonin and ethanol preference, in Galanter M (ed): Recent Developments in Alcoholism. New York, Plenum Press, 1989, pp 187-209 7. Khavari KA, Farber PD: A profile instrument for the quantification and assessment of alcohol consumption. J Stud Alcohol 39: 15251539, 1978

115

8. Schwartz JB, Keefe L, k r s t e n E, et al: Prolongation of verapamil elimination kinetics during chronic oral administration. Am Heart J 104:198-203, 1982 9. Ali SL: Nifedipine, in Florey K (ed): Analytical Profiles of Drug Substances. San Diego, Academic Press, 1987, pp 221-288 10. Perez-Reyes M: Pharmacodynamics of certain drugs of abuse, in Barnett G, Chiang CN (eds): Pharmacokinetics and Pharmacodynamics of Psychoactive Drugs. National Institute on Drug Abuse Research Monograph. Foster City, Biomedical Publications, 1985, pp 287-3 10 1 I . Hill HE, Haertzen CA, Belleville RE: The Addiction Research Center Inventory (ARCI) Test Booklet. Lexington, National Institute of Mental Health, 1958 12. Haertzen CA: An overview of Addiction Research Center Inventory Scales (ARCI): An appendix and manual of scales. Rockville, National Institute of Drug Abuse, 1974 13. Mills KC, Bisgrove EZ: Cognitive impairment and perceived risk from alcohol. J Stud Alcohol 44:26-46, 1983 14. Perez-Reyes M, Hicks RE, Bumbeny J, et al: Interaction between marihuana and ethanol: Effects on psychomotor performance. Alcohol Clin Exp Res 12:268-276, 1988