Dose-Related Ethanol-Like Effects of the NMDA Antagonist, Ketamine, in Recently Detoxified Alcoholics

ORIGINAL ARTICLE Dose-Related Ethanol-like Effects of the NMDA Antagonist, Ketamine, in Recently Detoxified Alcoholics

John H. Krystal, MD; Ismene L. Petrakis, MD; Elizabeth Webb; Ned L. Cooney, PhD; Laurence P. Karper, MD; Sheila Namanworth; Philip Stetson, PhD; Louis A. Trevisan, MD; Dennis S. Charney, MD

Background: This study evaluated the dose-related etha- Results: Ketamine produced dose-related ethanol-like ef- nol-like subjective effects of the N-methyl-D-aspartate fects on each scale measuring its similarity to ethanol. Its (NMDA) glutamate receptor antagonist ketamine hydro- effects were more similar to the sedative or descending limb chloride in recently detoxified alcoholics. effects of ethanol than to the stimulant or ascending limb effects. Ketamine effects also were more like ethanol than Methods: Twenty male inpatients meeting DSM-III-R cri- marijuana or cocaine. Ethanol-like effects were more promi- teria for alcohol dependence and who had not consumed nent at the higher ketamine dose, a dose rated as similar alcohol for 10 to 27 days prior to the study completed 3 test to greater levels of ethanol intoxication. However, ket- days that involved the intravenous infusion of ketamine hy- amine did not increase craving for ethanol. drochloride (0.1 mg/kg or 0.5 mg/kg) or saline solution under randomized double-blind conditions. Ethanol-like sub- Conclusion: The production of ethanol-like subjective jective effects were assessed using the Sensation Scale; the effects by ketamine supports the potential clinical im- Biphasic Alcohol Effects Scale; visual analog scales to mea- portance of NMDA receptor antagonism among the sure “high” and degree of similarity to ethanol, cocaine, and mechanisms underlying the subjective effects of etha- marijuana; a scale assessing the number of standard alco- nol in humans. hol drinks producing similar subjective effects; and visual analog scales measuring ethanol craving. Arch Gen Psychiatry. 1998;55:354-360

GROWING body of research paradigms.14-17 In these studies, the capac- indicates that the capacity ity of NMDA antagonists to substitute for of ethanol to block gluta- ethanol was greater with increasing refer- mateeffectsattheN-methyl- ence doses of ethanol. This finding sug- D-aspartate (NMDA) recep- gested that NMDA receptor blockade con- tor contributes to its acute behavioral effects tributed more prominently to the subjective A 15 and to the natural history and neuropa- effects of higher ethanol doses. thology of alcoholism.1 Ethanol reduces Our study evaluated whether ket- NMDA-stimulated ion currents in a non- amine produced ethanol-like subjective ef- competitive and concentration-depen- fects in recently detoxified alcoholic pa- dent fashion across the range of ethanol tients. To our knowledge, there are no concentrations (5-100 mmol/L) associ- previous clinical studies evaluating the ated with human ethanol intoxication.2-7 contributions of NMDA receptors to the Long-term ethanol administration in- behavioral effects of ethanol in humans. From the Department of creases the levels of NMDA receptor sub- Psychiatry, Yale University units, up-regulates NMDA receptor- School of Medicine and the related binding, and produces cross- RESULTS Veterans Affairs–Yale tolerance with other noncompetitive University Alcoholism Research NMDA antagonists.8-12 Increased NMDA re- EVIDENCE OF ETHANOL-LIKE Center, West Haven, Conn ceptor function produced by long-term EFFECTS (Drs Krystal, Petrakis, Cooney, ethanol administration contributes to with- Karper, Trevisan, and Charney drawal-related seizures10 and neurotoxic ef- Sensation Scale and Mss Webb and 13 Namanworth); and the Upjohn fects. Research Institute, Department The NMDA antagonists ketamine hy- Ketamine produced significant dose- of Pharmacology, University of drochloride, phencyclidine (PCP), and related ethanol-like effects as assessed by the Michigan Medical School, Ann dizocilpine maleate (MK-801) substitute for Sensation Scale (Figure 1; RMANOVA, Arbor (Dr Stetson). ethanol in preclinical drug discrimination dose ϫ time interaction: F12,228=12.1;

ARCH GEN PSYCHIATRY/ VOL 55, APR 1998 354

©1998 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/26/2021 PATIENTS AND METHODS Subjects were inpatients at the Substance Abuse Treat- ment Research Unit of the Veterans Affairs Connecticut Healthcare System, West Haven. They participated in test- PATIENTS ing for a mean±SD of 17.6±4.2 days (range, 10-27 days) after consuming their last alcoholic beverage. Fourteen patients Twentymaleinpatients(mean±SDage,44.0±10.5years;weight, completed detoxification with pharmacologic supports prior 74.7±9.0 kg) who met criteria for alcohol dependence18 as de- to study entry (benzodiazepines, n=10; nimodipine, n=4). termined by the Structured Clinical Interview for DSM-III-R 19 The mean±SD period between the administration of the last participated in testing. Patients began drinking at a mean±SD benzodiazepine dose and the first pharmacologic test day of 15.3±2.9 years of age, began regular drinking at 17.8±5.5 was 15.7±5.9 days (range, 7-26 days). On their first test day, years of age, began regular drinking to intoxication at 21.1±7.1 patients received placebo (n=5), 0.1 mg/kg ketamine hydro- yearsofage,andtheirheaviestlevelofdrinkingwasat32.7±13.8 chloride (n=9), or 0.5 mg/kg ketamine hydrochloride (n=6). years of age. Patients had a 23.0±10.1-year history of alcoholism. They had undergone a mean±SD of 5.8±9.7 inpatient al- TESTING PROCEDURE coholdetoxifications(range,0-40).Theirmeandailyconsump- This research protocol was approved by the Human Sub- tion of alcohol was equivalent to 391.5±170 mL of absolute jects Subcommittee of the Veterans Affairs Connecticut Health- alcohol per day. The mean±SD Michigan Alcoholism Screen- care System and the Human Investigations Committee of the ing Test score20 was 38.7±6.5. Sixteen (80%) of the 20 patients Yale University School of Medicine, New Haven, Conn. Af- in this study met the von Knorring et al21 criteria for type 2 alter giving informed consent for human investigation, each coholism, defined as age of onset before 25 years of age and 2 patient completed 3 test days separated by 48 to 96 hours in or more social consequences of alcoholism. Twelve (60%) of a randomized order under double-blind conditions. The in- the 20 patients had a first-degree relative with a history of al- formation presented to patients while obtaining consent in- coholism. Patients were medically stable at study entry based cluded a warning that the effects of ketamine might re- on medical history, physical examination, and routine labo- semble ethanol intoxication and might stimulate craving for ratory testing. alcohol. On each test day, patients received a 40-minute in- Patients were excluded if they met the criteria for another travenous infusion containing either saline solution, 0.1 mg/kg substance use disorder other than nicotine dependence in the ketamine hydrochloride, or 0.5 mg/kg ketamine hydrochlo- year prior to testing. Fifteen patients (75%) reported lifetime ride (Ketalar, Parke-Davis, Kalamazoo, Mich). This method marijuana use, but no use occurred in the year prior to test- of administration was similar to that reported previously in ing. Ten patients (50%) had lifetime cocaine use. Of these pa- healthy subjects.22 For each test session, participants fasted tients, 1 used cocaine 6 months prior to testing at a subabuse overnight and remained in a fasting state during the test ses- level and the remainder had not used cocaine for at least 1 year sion. They presented for testing at approximately 8:30 AM and prior to testing. The absence of other current substance abuse an intravenous line was placed at that time. Blood was drawn wassupportedbynegativeresultsofurinetoxicologicalscreens to determine ketamine levels at 10 and 80 minutes after the prior to testing. Subjects were also excluded if they had an- initiation of ketamine infusion. other DSM-III-R Axis I diagnosis during a period that was free of alcohol consumption. Continued on next page

PϽ.001). Post hoc RMANOVAs revealed that 0.5 mg/kg (Figure 2;RMANOVA,doseϫtimeinteraction:F12,228=13.1; ketamine hydrochloride produced greater Sensation Score PϽ.001). Post hoc RMANOVAs revealed that 0.5 mg/kg ket- increases than both 0.1 mg/kg ketamine hydrochloride (dose amine hydrochloride was perceived as similar to a greater ϫ time interaction: F6,114=13.1; PϽ.001) and saline solu- number of standard ethanol drinks than were both 0.1 mg/ tion (dose ϫ time interaction: F6,114=12.4; PϽ.001). How- kg ketamine hydrochloride (dose ϫ time interaction: ever, 0.1 mg/kg ketamine hydrochloride effects were not F6,114=12.4; PϽ.001) and saline solution (dose ϫ time in- significantly different from saline solution. teraction: F6,114=14.3; PϽ.001). There was also a nonsignificant trend for 0.1 mg/kg ketamine hydrochloride effects to Self-reported High be rated as similar to more ethanol drinks than saline solution (dose ϫ time interaction: F6,114=3.4; P=.06). As depicted in Figure 1, ketamine increased self-rated high in a dose-related manner (RMANOVA, dose ϫ time inter- SPECIFICITY OF ETHANOL-LIKE EFFECTS action: F12,228=8.6; PϽ.001). Post hoc RMANOVAs revealed that 0.5 mg/kg ketamine hydrochloride produced Differential Similarity to the Ascending and greater euphoria than both 0.1 mg/kg ketamine hydrochlo- Descending Limbs of Ethanol Intoxication ride (dose ϫ time interaction: F6,114=7.3; PϽ.001) and saline solution (dose ϫ time interaction: F6,114=15.3; PϽ.001). Ketamine increased total scores on the Biphasic Alcohol Effects of 0.1 mg/kg ketamine hydrochloride were not sig- Effects Scale (RMANOVA, dose ϫ time interaction: nificantly different than saline solution. F12,204=6.0; P=.001). The ketamine dose effect was ex- plored with post hoc within-subjects contrasts, which in- Number of Drinks Scale dicated that 0.5 mg/kg ketamine hydrochloride had greater ethanol-like effects relative to both 0.1 mg/kg and saline Ketamine produced a dose-dependent increase in the per- solution (F1=5.1; P=.03), but no significant difference be- ceived number of standard ethanol drinks administered tween the 0.1 mg/kg dose and saline solution.

ARCH GEN PSYCHIATRY/ VOL 55, APR 1998 355

©1998 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/26/2021 Ratings were performed to characterize subjective re- ous study.23 Assessments of ethanol-like effects, mood states, sponses to ketamine that mirrored a previous psychophar- and craving were completed at 60 and 15 minutes prior to macologic study.23 Patients completed visual analog scales ketamine infusion and 10, 40, 80, 110, 170, and 230 min- of anger, anxiety, high, nervousness, and sadness.22,23 Anxi- utes after the initiation of ketamine infusion. ety was defined as “a mental awareness of worry.” Ner- vousness was defined as a “physical feeling of jitteriness, KETAMINE LEVELS tension, heart throbbing, breathlessness, or other similar symptoms.” These scales consisted of 100-mm lines Plasma ketamine levels were determined by gas chroma- (0=none; 100=maximum possible) marked proportion- tography–mass spectrometry by 1 of the authors (P.S.) ac- ately to the perceived intensity of the experience. cording to methods reported previously.26 Triplicate qual- Patients completed measures of ethanol-like subjec- ity control samples were assayed on each of 3 consecutive tive effects: the Sensation Scale24; self-rated visual analog days. The calibration curve was calculated for ketamine con- scales measuring similarity to acute behavioral effects of centrations ranging between 20 ng/mL and 500 ng/mL. alcohol, cocaine, and marijuana (0=not at all similar; The concentration means for seeded control samples con- 100=identical); a scale measuring the number of standard taining 50 ng/mL and 200 ng/mL were found to be within drinks of ethanol comparable to their drug responses; and 1.3% and 3.1% of the theoretical values. The assay was the Biphasic Alcohol Effects Scale.25 The Biphasic Alcohol found to have coefficients of variation ranging between Effects Scale measures stimulant effects associated with the 3.7% and 4.9%. ascending limb of ethanol intoxication and sedative effects associated with the descending limb of ethanol in- DATA ANALYSIS toxication. Items associated with the ascending limb of ethanol intoxication include energized, excited, stimulated, Data were evaluated initially using repeated-measures analy- talkative, “up,” and vigorous. Items associated with the de- sis of variance (RMANOVA) with within-subjects factors scending limb include difficulty concentrating, “down,” of drug (placebo, 0.1 mg/kg ketamine hydrochloride, or 0.5 heavy head, inactive, sedated, slow thoughts, and slug- mg/kg ketamine hydrochloride) and time. These gish. Inspection of the data on the visual analog scales mea- RMANOVAs were tested for lack of sphericity and Huynh- suring similarity to ethanol, cocaine, and marijuana re- Feldt adjustments were made to the degrees of freedom to vealed no clear discriminative effects of ketamine beyond reduce type I error. Significant RMANOVAs were followed 80 minutes after the initiation of drug infusion. To sim- with post hoc RMANOVAs comparing the responses fol- plify the analysis and reduce lack of sphericity, time points lowing the low and high doses of ketamine to placebo. Sig- up to 80 minutes after the initiation of drug infusions were nificant main effects were also followed with post hoc within- analyzed. For comparison purposes, standard drinks were subjects contrasts with significance adjusted for multiple defined as equivalents of 15 mL of absolute ethanol, ap- comparisons using Bonferroni corrections. The means of the proximately comparable to 12 oz of beer, 4 oz of wine, 1.25 baseline values were compared between test days using paired oz of 80-proof alcohol, or 1 oz of 100-proof alcohol. Etha- t tests with Bonferroni correction made to decrease the ef- nol craving was assessed using a self-rated visual analog fect of multiple comparisons. No significant baseline differ- scale evaluating “desire to drink alcohol” used in a previ- ences emerged in these analyses.

As depicted in Figure 3, ketamine increased scores previous experience with the effects of ethanol, mari- on the sedative or descending limb items on the Bipha- juana, and cocaine compared the similarity of ketamine ef- sic Alcohol Effects Scale, but not the stimulant or as- fects with each of these drugs of abuse. The RMANOVA cending limb items. The overall RMANOVA performed performed on these data revealed significant effects of the on subjects completing the Biphasic Alcohol Effects Scale reference drug of abuse (ethanol, marijuana, or cocaine; (n=18) revealed significant effects (dose ϫ time inter- F2,18=3.8; P=.04), ketamine dose (F2,18=7.3; P=.01), and the action: F12,204=6.2; PϽ.001; limb ϫ dose ϫ time interac- ketamine dose ϫ time interaction (F6,54=6.0; P=.002). There tion: F12,204=4.1; P=.001). There were no significant main was a nonsignificant trend toward significance for the drug effects or interactions in the RMANOVA performed on of abuse ϫ ketamine dose ϫ time interaction (F12,108=1.9; ascending limb data. However, the RMANOVA con- P=.1). A post hoc within-subjects contrast revealed that ket- ducted on descending limb data revealed highly signifi- amine effects were significantly more similar to ethanol than cant ketamine effects (dose ϫ time interaction: F12,204=8.4; to both marijuana and cocaine (F1=6.7; P=.02). PϽ.001). Post hoc contrasts revealed that the 0.5 mg/kg ketamine hydrochloride dose effect was greater then both Ethanol 0.1 mg/kg and placebo effects (F1=7.9; P=.009) and the 0.1 mg/kg ketamine hydrochloride dose effect was greater Ketamine effects showed a dose-related similarity to etha- than the placebo effect (F1=4.6; P=.04). nol effects (RMANOVA, dose ϫ time interaction: F6,114=12.1; PϽ.001). Post hoc RMANOVAs revealed that Differential Similarity to Ethanol, 0.5 mg/kg ketamine hydrochloride was more similar to Cocaine, and Marijuana ethanol than both 0.1 mg/kg ketamine hydrochloride (dose ϫ time interaction: F6,114=12.9; PϽ.001) and saline solu- Ketamine effects were rated as more similar to those of etha- tion (dose ϫ time interaction: F6,114=16.6; PϽ.001). How- nol than to marijuana or cocaine (Figure 4, visual ana- ever, 0.1 mg/kg ketamine hydrochloride effects were not log scale). In the initial analysis, the 10 patients who had significantly different from saline solution.

ARCH GEN PSYCHIATRY/ VOL 55, APR 1998 356

©1998 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/26/2021 50 A 50 Placebo B 0.1 mg/kg Ketamine Hydrochloride 0.5 mg/kg Ketamine Hydrochloride 40 40

30 30

20 20 Sensation Scale Score High Analog Scale Score

10 10

0 0 –30030 60 90 120 150 180 210 240 –30030 60 90 120 150 180 210 240 Time, min Time, min Figure 1. Effects of placebo, 0.1 mg/kg ketamine hydrochloride, and 0.5 mg/kg ketamine hydrochloride on Sensation Scale Scores (A) and on self-rated “high” (B) in recently detoxified alcoholic patients (N=20). Values are expressed as mean±SEM. See “Patients and Methods” and “Results” sections for explanation of statistical analyses.

Marijuana 10 Placebo 0.1 mg/kg Ketamine Hydrochloride 9 In patients reporting a history of marijuana use, ket- 0.5 mg/kg Ketamine Hydrochloride amine had dose-related marijuana-like effects. The 8

RMANOVA performed on data from the visual analog 7 scale assessing similarity to marijuana revealed a signifi- 6 cant ketamine dose ϫ time interaction (F6,84=4.6; P=.02). 5 Cocaine 4 In patients reporting a history of cocaine use, there was a 3 trend for ketamine to produce cocaine-like effects. The 2

RMANOVA performed on data from the visual analog scale No. of Standard Ethanol Drinks Perceived measuring similarity to cocaine revealed a nonsignificant 1 ketamine dose ϫ time interaction (F6,54=3.1; P=.09). 0 –30030 60 90 120 150 180 210 240 Time, min SELF-RATED VISUAL ANALOG SCALES OF Figure 2. The number of standard ethanol drinks that recently detoxified CRAVING AND MOOD STATES alcoholic patients (N=20) determined were similar to the effects of placebo, 0.1 mg/kg ketamine hydrochloride, and 0.5 mg/kg ketamine hydrochloride. There was no significant increase in self-rated desire to Values are expressed as mean±SEM. See “Patients and Methods” and drink alcohol following high doses of ketamine (base- “Results” sections for explanation of statistical analyses. line craving [mean±SD], 19.5±7.0 mm; 10 minutes after detoxifiedtype2alcoholicsacrossseveralresponsemeasures. initiating the infusion, 25.8±7.8 mm) or low doses of ket- Ketamine effects were rated more similar to items associated amine (baseline, 18.2±6.8 mm; 10 minutes after initiat- with the sedative or descending limb than the stimulant or ing the infusion, 23.2±7.1 mm) relative to placebo (base- ascending limb of the Biphasic Alcohol Effects Scale. These line, 15.0±4.7 mm; 10 minutes after initiating the infusion, data suggested a possible differential contribution of NMDA 18.9±5.6 mm). The RMANOVA performed on these data receptors to the stimulant and sedative effects of ethanol. found a significant time effect (F6,114=3.9; P=.03), but no Ketamine hydrochloride produced effects similar to 1.5±2.5 other main effects or interactions. No significant ket- standard alcohol drinks at the 0.1 mg/kg dose and 8.7±8.1 amine effects were found on the visual analog scales for standard alcohol drinks at the 0.5 mg/kg dose. As predicted anger, anxiety, drowsiness, nervousness, or sadness. by the preclinical literature,15 ketamine doses associated with greater similarity to ethanol produced effects that were at- PLASMA KETAMINE LEVELS tributed to higher levels of ethanol consumption. Ketamine effects were rated more similar to ethanol Ketamine blood levels increased in a dose-dependent fash- than to either marijuana or cocaine. Thus, NMDA recep- ion (Figure 5; RMANOVA, dose ϫ time interaction: tor antagonism may figure more prominently in the behav- F4,64=30.7; PϽ.001). ioral effects of ethanol than marijuana or cocaine. In contrast, patients found mCPP effects comparably similar to 23 COMMENT ethanol, marijuana, and cocaine. Although more modest than its ethanol-like effects, ketamine effects did show some The principal finding of this study was that ketamine pro- similaritytotheeffectsofmarijuanaandcocaineinourstudy. duced dose-related ethanol-like subjective effects in recently The NMDA antagonists also showed cocaine-like effects in

ARCH GEN PSYCHIATRY/ VOL 55, APR 1998 357

©1998 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/26/2021 24 A 24 Placebo B 0.1 mg/kg Ketamine Hydrochloride 0.5 mg/kg Ketamine Hydrochloride 20 20

16 16

12 12

8 8

4 4

0 0 Baseline40 80 120 160 200 240 Baseline40 80 120 160 200 240 Biphasic Alcohol Effects Scale; Stimulant–Ascending Limb Score Time, min Biphasic Alcohol Effects Scale; Sedative–Descending Limb Score Time, min

Figure 3. Effects of placebo, 0.1 mg/kg ketamine hydrochloride, and 0.5 mg/kg ketamine hydrochloride on scores for the stimulant or ascending limb of the Biphasic Alcohol Effects Scale (A) and on the sedative or descending limb of the Biphasic Alcohol Effects Scale (B) in recently detoxified alcoholic patients (N=20). Values are expressed as mean±SEM. See “Patients and Methods” and “Results” sections for explanation of statistical analyses.

50 50 50 A B Placebo C 0.1 mg/kg Ketamine Hydrochloride 0.5 mg/kg Ketamine Hydrochloride 40 40 40

30 30 30

20 20 20

10 10 10 Similarity to Ethanol, Visual Analog Scale Score Similarity to Cocaine, Visual Analog Scale Score 0 Similarity to Marijuana, Visual Analog Scale Score 0 0 Baseline10 20 30 40 50 60 70 80 Baseline10 20 30 40 50 60 70 80 Baseline10 20 30 40 50 60 70 80 Time, min Time, min Time, min

Figure 4. Similarity of the effects of placebo, 0.1 mg/kg ketamine hydrochloride, and 0.5 mg/kg ketamine hydrochloride to ethanol (A), marijuana (B), and cocaine (C) in recently detoxified alcoholic patients (N=20). Values are expressed as mean±SEM. See “Patients and Methods” and “Results” sections for explanation of statistical analyses.

preclinical studies.27,28 However, the current study design ther, ketamine and PCP abuse has been a significant clini- may have biased the results in favor of finding ethanol-like cal problem.34 However, ketamine may have failed to pro- effects. For example, this study evaluated a patient group duce dysphoric emotional states that have been linked to the who identified ethanol as their primary substance of abuse. elicitationofcravinginotherstudies.22,35 Inaddition,thelower Also, the comparisons of ketamine with cocaine were lim- ketamine dose may not have been sufficiently similar to etha- ited to a smaller subsample of alcoholics with cocaine use nol to facilitate the induction of craving.36-38 In contrast, the histories, reducing the statistical power of this analysis. In higher ketamine dose may have sated the desire for further addition, the slow intravenous ketamine infusion used in consumption of an NMDA antagonist–like compound. It is the current study may have minimized the stimulant effects possible that an intermediate ketamine dose might have been and enhanced the sedative or descending limb effects of ket- more effective in stimulating ethanol craving. amine. Stimulant effects in this study may have been more The inability of ketamine to prime craving in the pa- prominent had ketamine been administered as a rapid in- tient group may also have been related to the similarity travenous bolus.22,26 of its effects to the sedative effects of ethanol that emerge Ketamine did not stimulate craving relative to placebo as blood alcohol levels plateau or decline.25 Stimulant ef- in our study. However, both ketamine and placebo infusion fects associated with the ascending limb of ethanol in- briefly increased craving, suggesting that test day instruc- toxication are more closely tied to the development of tions may have created an expectancy that craving would craving than are the sedative or descending limb ef- develop. The current findings contrasted with previous stud- fects.39,40 Currently, there is no clear evidence implicat- ies of mCPP,23,29 in which the production of ethanol-like sub- ing NMDA antagonism in the stimulant effects of etha- jective effects was accompanied by craving. The failure to nol. Instead, clinical studies have implicated both produce craving was not likely caused by the absence of re- catecholamine41,42 and opiate43 systems in these effects. warding effects of ketamine. In animals, NMDA antagonists The neuropharmacology of the ethanol-like effects of produceconditionedplacepreference,30 enhancebrainstimu- ketamine remains to be clarified. The generalization between lation reward,31 and are usually self-administered.32,33 Fur- ethanol and the other NMDA antagonists is not symmetri-

ARCH GEN PSYCHIATRY/ VOL 55, APR 1998 358

©1998 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/26/2021 500 Modulation of dopamine systems also may have con- Placebo 0.1 mg/kg Ketamine Hydrochloride tributed to the current findings. Ketamine has direct do- 450 0.5 mg/kg Ketamine Hydrochloride † paminergic effects via its low affinity for dopamine reup- 400 take sites as well as its NMDA receptor–mediated modulation of dopaminergic neuronal activity.57,58 However, 350 ketamine effects on dopamine neurons do not correlate well 59 300 with its NMDA antagonist-like discriminative properties. Also, the euphoric effects of ketamine in humans seem to 250 be insensitive to haloperidol pretreatment.60 200 Our data support the hypothesis that the capacity of † ethanol to block NMDA receptors contributes significantly 150 to its subjective effects in humans. Future studies should

Plasma Ketamine Level, ng/mL ∗ 100 explore a wider range of ketamine doses and rates of administration. Further, the similarity of ketamine to drugs 50 of abuse should be evaluated in populations primarily de- 0 pendent on cocaine or marijuana. These studies should also consider employing training doses of ethanol, cocaine, and –50 0401020 30 50 60 70 80 marijuana to facilitate the accuracy of interpreting the simi- Time, min laritybetweenketamineeffectsandthoseoftheseotherdrugs. Figure 5. Plasma ketamine levels in recently detoxified alcoholic patients Comparisons of ketamine and sedative-hypnotic agents (n=17) following the administration of placebo, 0.1 mg/kg ketamine would also provide insights into the specificity of the simi- hydrochloride, and 0.5 mg/kg ketamine hydrochloride. Values are expressed larity between ketamine effects and the sedative effects of as mean±SEM. Asterisk indicates PϽ.05 by within-subjects contrasts with ethanol. In addition, the dependence on subjective report Bonferroni adjustments for multiple comparisons; dagger, PϽ.001 by within-subjects contrasts with Bonferroni adjustments for multiple is a potential limitation of our study. Physiologic measures comparisons. See “Patients and Methods” and “Results” sections for might aid the evaluation of neurobiological contributions explanation of all other statistical analyses. to the acute behavioral effects of ethanol. The NMDA antagonist properties of ethanol may have cal, reflecting the multiplicity of mechanisms contributing therapeutic implications in humans. For example, acam- to the discriminative properties of ethanol.14,23,44,45 Animals prosate reduced ethanol consumption in clinical trials. 61,62 trained to discriminate ethanol from other drugs recognize This drug has both NMDA agonist and antagonist-like ef- the ethanol-like properties of NMDA antagonists. In con- fects, making it difficult to extrapolate a therapeutic mecha- trast, ethanol effects on other systems are sufficiently promi- nism at this time.63 One potential strategy would be to ex- nent to animals trained to discriminate NMDA antagonists plore agents that reduce ethanol effects at the NMDA re- from other drugs to make ethanol seem like a different type ceptor. One class of candidate agents to serve this function of drug.15,46 Consistent with this view, ethanol also shows wouldbeagonistsofthestrychnine-insensitiveglycinemodu- asymmetrical generalization with selective agents acting at latory site of the NMDA receptor complex. These drugs re- other sites of ethanol action, such as ␥-aminobutyric acid duce ethanol effects in some preclinical studies.64,65 Prelimi- and serotonin receptors.45,47 Thus, it is possible that an ap- nary human data also suggest that the strychnine-insensitive propriate combination of drugs acting at NMDA, serotonin, glycine partial agonist, D-cycloserine, exacerbates ethanol ␥-aminobutyricacid,andotherreceptorsmightproducesym- intoxication at doses associated with NMDA antagonistlike metrical generalization with ethanol. amnestic and euphoric effects.62 Thus, NMDA receptors may Ketamine also appears to be a more complex stimu- become an important focus for future drug development lus in animals than the selective noncompetitive NMDA in the alcoholism field. antagonist dizocilpine (MK-801),48 as suggested by the asymmetrical generalization between these drugs. Ani- mals trained to discriminate ketamine recognize dizo- Accepted for publication June 9, 1997. cilpine as ketamine-like,33 while animals trained to iden- This research was supported by grant NIAAA 1 R01 tify dizocilpine do not recognize ketamine as a similar AA10121-01 from the National Institute on Alcohol Abuse agent.49 The complexity of the ketamine stimulus may and Alcoholism, Bethesda, Md (Dr Krystal), and the De- arise from its differential relative affinity for NMDA re- partment of Veterans Affairs, Washington, DC, through fund- ceptor subunits,50 its agonism of the µ−opiate receptor, 51,52 ing of the VA-Yale Alcoholism Research Center and a Merit and its blockade of dopamine transporters.53 Review Grant (Dr Krystal). The µ−agonist actions of ketamine may be of limited We wish to acknowledge the critical contributions to importance to its ethanol-like effects. The discriminative this research made by the clinical and research staff of the properties of NMDA antagonists, particularly their ethanol- Biostudies Unit and Substance Abuse Treatment Unit of the like effects, are not dependent on their affinity for µ−re- West Haven Veterans Affairs Medical Center. We also thank ceptors.15,54 Similarly, µ−receptor agonism does not ap- Christine Boose for her assistance in data collection and pear to contribute to the discriminative properties of ethanol analysis. in animals.55 However, µ−antagonists reduce aspects of hu- Corresponding author: John H. Krystal, MD, Depart- man ethanol intoxication.43,56 Thus, future studies will be ment of Psychiatry, Schizophrenia Biology Research Cen- needed to more fully assess the contributions of µ−recep- ter, Yale University, 950 Campbell Ave, VAMC 116A, West tors to the ethanol-like effects of ketamine. Haven, CT 06516.

ARCH GEN PSYCHIATRY/ VOL 55, APR 1998 359

©1998 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/26/2021 32. Young AM, Woods JH. Maintenance of behavior by ketamine and related com- REFERENCES pounds in rhesus monkeys with different self-administration histories. J Phar- macol Exp Ther. 1981;218:720-727. 1. Tsai G, Gastfriend DR, Coyle JT. The glutamatergic basis of human alcoholism. 33. Koek W, Woods JH, Winger GD. MK-801, a proposed noncompetitive antago- Am J Psychiatry. 1995;152:332-340. nist of excitatory amino acid neurotransmission, produces phencyclidine-like be- J Pharmacol Exp Ther. 1988; 2. Nakanishi S. Molecular diversity of glutamate receptors and implications for brain havioral effects in pigeons, rats and rhesus monkeys. 245:969-974. function. Science. 1992;258:597-603. 3. Lovinger DM, White G, Weight FF. Ethanol inhibits NMDA-activated ion current 34. Siegel RK. Phencyclidine and ketamine intoxication: a study of four populations in hippocampal neurons. Science. 1989;243:1721-1724. of recreational users. In: Petersen RC, Stillman RC, eds. Phencyclidine (PCP) 4. Lovinger DM, White G, Weight FF. NMDA receptor-mediated synaptic excitation Abuse: An Appraisal. Rockville, Md: National Institute on Drug Abuse; 1978:119- selectively inhibited by ethanol in hippocampal slices from adult rat. J Neurosci. 147. NIDA Research Monograph 21. 1990;10:1372-1379. 35. Litt MD, Cooney NL, Kadden RM, Gaupp L. Reactivity to alcohol cues and in- Addict Behav. 1990;15:137-146. 5. White G, Lovinger DM, Weight FF. Ethanol inhibits NMDA-activated current but duced moods in alcoholics. 36. Engle KB, Williams TK. Effect of an ounce of vodka on alcoholics’ desire for al- does not alter GABA-activated current in an isolated adult mammalian neuron. cohol. Q J Stud Alcohol. 1972;33:1099-1105. Brain Res. 1990;507:332-336. 37. Ludwig AM, Wikler A, Stark L. The first drink. Arch Gen Psychiatry. 1974;30:539-547. 6. Dildy-Mayfield JE, Leslie SW. Mechanism of inhibition of N-methyl-D-aspartate– stimulated increases in free intracellular Ca2+ concentration by ethanol. J Neu- 38. Hodgson R, Rankin H, Stockwell T. Alcohol dependence and the priming effect. Behav Res Ther. 1979;17:379-387. rochem. 1991;56:1536-1543. 7. Simson PE, Criswell HE, Johnson DB, Hicks RE, Breese GR. Ethanol inhibits NMDA- 39. Newlin D, Thomson J. Alcohol challenge in sons of alcoholics: a critical review and analysis. Psychol Bull. 1990;108;383-402. evoked electrophysiological activity in vivo. J Pharmacol Exp Ther. 1991;257: 40. Earleywine M. Confirming the factor structure of the Anticipated Biphasic Alco- 255-231. hol Effects Scale. 8. Trevisan L, Fitzgerald LW, Brose N, Gasic GP, Heinemann SF, Duman RS, Nes- Alcohol Clin Exp Res. 1994;18:861-866. tler EJ. Chronic ingestion of ethanol up-regulates NMDA R1 subunit immuno- 41. Perez-Reyes M, McDonald SA, Hicks RE. Interaction between ethanol and dex- troamphetamine. Alcohol Clin Exp Res. 1992;16:75-81. reactivity in rat hippocampus. J Neurochem. 1994;62:1635-1638. 42. McDougle CJ, Krystal JH, Price LH, Heninger GR, Charney DS. Noradrenergic 9. Danysz W, Jankowska E, Glazewski S, Kostowski W. The involvement of NMDA response to acute ethanol administration in healthy subjects: comparison with receptors in acute and chronic effects of ethanol. Alcohol Clin Exp Res. 1992; intravenous yohimbine. Psychopharmacology. 1995;118:127-135. 16:499-504. 43. Swift RM, Whelihan W, Kuznetsov O, Buongiorno G, Hsuing H. Naltrexone-induced 10. Grant KA, Valverius P, Hudspith M, Tabakoff B. Ethanol withdrawal seizures and alterations in human ethanol intoxication. Am J Psychiatry. 1994;151:1463-1467. the NMDA receptor complex. Eur J Pharmacol. 1990;176:289-296. 44. Overton DA. Comparison of ethanol, pentobarbital, and phenobarbital using drug 11. Iorio KR, Reinlib L, Tabakoff B, Hoffman PL. Chronic exposure of cerebellar gran- vs drug discrimination training. Psychopharmacology. 1977;53:195-199. ule cells to ethanol results in increased N-methyl-D-aspartate receptor function. 45. Signs SA, Schechter MD. The role of dopamine and serotonin receptors in the me- J Pharmacol Exp Ther. 1992;41:1142-1148. diation of the ethanol interoceptive cue. Pharmacol Biochem Behav. 1985;30:55-64. 12. Fidecka S, Langwinski R. Interaction between ketamine and ethanol in rats and 46. Garcha HS, Stolerman IP. Discrimination of a drug mixture in rats: role of train- mice. Pol J Pharmacol. 1989;41:23-32. ing dose, and specificity. Behav Pharmacol. 1989;1:25-31. 13. Chandler LJ, Newsom H, Sumner C, Crews F. Chronic ethanol exposure potentiates 47. De Vry J, Slanger JF. Effect of training dose on discrimination and cross- NMDA excitotoxicity in cerebral cortical neurons. J Neurochem. 1993;60:1578-1581. generalization of chlordiazepoxide, pentobarbital, and ethanol in the rat. Psy- 14. Grant KA, Knisely JS, Tabakoff B, Barrett JE, Balster RL. Ethanol-like discrimi- chopharmacology. 1986;88:341-345. native stimulus effects of non-competitive N-methyl-D-aspartate antagonists. Be- 48. Wong EHF, Kemp JA, Priestley T, Knight AR, Woodruff GN, Iversen LL. The an- hav Pharmacol. 1991;2:87-95. ticonvulsant MK-801 is a potent N-methyl-D-aspartate antagonist. Proc Natl Acad 15. Grant KA, Colombo G. Discriminative stimulus effects of ethanol. J Pharmacol SciUSA.1986;83:7104-7107. Exp Ther. 1992;254:1241-1247. 49. France CP, Moerschbaecher JM, Woods JH. MK-801 and related compounds in 16. Butelman ER, Baron SP, Woods JH. Ethanol effects in pigeons trained to dis- monkeys. J Pharmacol Exp Ther. 1991;257:727-773. criminate MK-801, PCP or CGS-19755. Behav Pharmacol. 1993;4:57-60. 50. Sucher NJ, Awobuluyi M, Choi Y-B, Lipton SA. NMDA receptors from genes to 17. Schechter MD, Meehan SM, Gordon TL, McBurney DM. The NMDA receptor an- channels. Trend Pharmacol Sci. 1996;17:348-355. tagonist MK-801 produces ethanol-like discrimination in the rat. Alcohol. 1993; 51. Smith DJ, Westfall DP, Adams JD. Ketamine interacts with opiate receptors as 10:197-201. an agonist [abstract]. Anesthesiology. 1980;53:S-5. 18. American Psychiatric Association. Diagnostic and Statistical Manual of Mental 52. Herling S, Coale EH, Hein DW, Winger G, Woods JH. Similarity of the discrimi- Disorders, Third Edition, Revised. Washington, DC: American Psychiatric Asso- native stimulus effects of ketamine, cyclazocine, and dextrorphan in the pigeon. ciation; 1987. Psychopharmacology. 1982;73:286-291. 19. Spitzer RI, Williams JBW, Gibbon M, First MB. The Structured Clinical Interview 53. Smith DJ, Azzaro AJ, Zaldivar SB, Palmer S, Lee HS. Properties of the optical for DSM-III-R, I: history, rationale, and description. Arch Gen Psychiatry. 1992; isomers and metabolites of ketamine on the high-affinity transport and catabo- 49:624-629. lism of monoamines. Neuropharmacology. 1981;20:391-396. 20. Selzer ML. The Michigan Alcoholism Screening Test: the quest for a new diag- 54. Shannon HE. Pharmacological analysis of the phencyclidine-like discriminative prop- nostic instrument. Am J Psychiatry. 1971;127:1653-1658. erties of narcotic derivatives in rats. J Pharmacol Exp Ther. 1982;222:146-151. 21. von Knorring AL, Bohman M, von Knorring L, Oreland L. Platelet MAO activity as a 55. Winter JC. The stimulus properties of morphine and ethanol. Psychopharmaco- biological marker in subgroups of alcoholism. Acta Psychiatr Scand. 1985;72:51-58. logia. 1975;44:209-214. 22. Krystal JH, Karper LP, Seibyl JP, Freeman GK, Delaney R, Bremmer JD, Heninger 56. Volpicelli JP, Watson NT, King AC, Sherman CE, O’Brien CP. Effect of naltrexone GR, Bowers MB Jr, Charney DS. Subanesthetic effects of the NMDA antagonist, on alcohol “high” in alcoholics. Am J Psychiatry. 1995;152:613-615. ketamine, in humans. Arch Gen Psychiatry. 1994;51:199-214. 57. Javitt DC, Zukin SR. Recent advances in the phencyclidine model of schizophre- 23. Krystal JH, Webb E, Cooney NL, Kranzler H, Charney DS. Specificity of ethanol-like nia. Am J Psychiatry. 1991;148:1301-1308. effects elicited by serotonergic and noradrenergic mechanisms. Arch Gen Psychia- 58. Irufine M, Shimizu T, Nomoto M. Ketamine-induced hyperlocomotion associ- try. 1994;51:898-911. ated with alteration of presynaptic components of dopamine neurons in the nucleus 24. Maisto SA, Conners GJ, Tucker JA, McCollam JB, Addesso VJ. Validation of the accumbens of mice. Pharmacol Biochem Behav. 1991;40:399-407. Sensation Scale, a measure of subjective physiological responses to alcohol. Be- 59. Snell LD, Mueller ZL, Gannon RL, Silverman PB, Jonson KM. A comparison behav Res Ther. 1980;18:37-43. tween classes of drugs having phencyclidine-like behavioral properties on do- 25. Martin CS, Earleywine M, Musty RE, Perrine MW, Swift RM. Development and vali- pamine efflux in vitro and dopamine metabolism in vivo. J Pharmacol Exp Ther. dation of the Biphasic Alcohol Effects Scale. Alcohol Clin Exp Res. 993;17:140-146. 1984;231:261-269. 26. KrystalJH,KarperLP,D’SouzaDC,Abi-DarghamA,MorrisseyK,BremnerJD,Heninger 60. Krystal J, Karper L, Bennett A, Abi-Saab D, D’Souza C, Abi-Dargham A, Charney GR, Bowers MB, Suckow R, Charney DS. Interactive effects of subanesthetic ketamine D. Modulating ketamine-induced thought disorder with lorazepam and haloperi- and subhypnotic lorazepam in humans. Psychopharmacology. In press. dol in humans [abstract]. Schiz Res. 1995;15:156. 27. Colpaert FC, Niemegeers CJE, Janssen PAJ. Discriminative stimulus properties 61. Sass H, Soyka M, Mann K, Zieglgansberger W. Relapse prevention by acampro- of cocaine. Pharmacol Biochem Behav. 1979;10:535-546. sate: results from a placebo-controlled study on alcohol dependence. Arch Gen 28. Koek W, Colpaert FC, Woods JH, Kamenka J-M. The phencyclidine (PCP) ana- Psychiatry. 1996;53:673-680. log N-(1-[2-benzo(B)thiophenyl]cyclohexyl) piperidine shares cocaine-like but 62. Whitworth AB, Fischer F, Lesch OM, Nimmerrichter A, Oberbauer H, Platz T, Pot- not other characteristic behavioral effects with PCP, ketamine, and MK-801. gieter A, Walter H, Fleischhacher WW. Comparison of acamprosate and placebo J Pharmacol Exp Ther. 1989;250:1019-1027. in long-term treatment of alcohol dependence. Lancet. 1996;347:1438-1442. 29. Benkelfat C, Murphy DL, Hill JL, George DT, Nutt D, Linnoila M. Ethanollike prop- 63. Lovinger DM, Zieglgansberger W. Interactions between ethanol and drugs act- erties of the serotonergic partial agonist m-chlorophenylpiperazine in chronic al- ing on the NMDA-type glutamate receptor. Alcohol Clin Exp Res. 1996;20(suppl): coholic patients. Arch Gen Psychiatry. 1991;48:383. 187A-191A. 30. Layer RT, Kaddis FG, Wallace L. The NMDA receptor antagonist MK-801 elicits con- 64. Rabe CS, Tabakoff B. Glycine site-directed agonists reverse the actions of etha- ditioned place preference in rats. Pharmacol Biochem Behav. 1993;44:245-247. nol at the N-methyl-D-asparate receptor. Mol Pharmacol. 1990;38:753-757. 31. Heberg LJ, Rose IC. The effect of MK-801 and other antagonists of NMDA-type 65. Woodward JJ, Gonzales RA. Ethanol inhibition of N-methyl-D-aspartate– glutamate receptors on brain-stimulation reward. Psychopharmacology. 1989; stimulated endogenous dopamine release from rat striatal slices: reversal by gly- 99:87-90. cine. J Neurochem. 1990;54:712-715.

ARCH GEN PSYCHIATRY/ VOL 55, APR 1998 360