Purpose: Neuropsychological, brain imaging: To investigate mood, psychopathology, cognitive deficits, and serotonin system function in current ecstasy users, former ecstasy users, polydrug users and non-drug users. Specific Hypotheses Tested - That when compared with polydrug users and non-drug users, current and former ecstasy users would exhibit elevated psychopathology, selective deficits in cognitive performance, and reduced serotonin transporter site (SERT) levels.

Design: Retrospective (non-experimental) between subjects design, with drug use (current ecstasy use, previous ecstasy use, polydrug use or no drug use) serving as a between-subjects factor, with all participants completing measures of mood, psychological problems, and cognitive function, and undergoing PET scans with (+)-McN5652.

Subjects: 30 current ecstasy users, 31 former ecstasy users, 29 polydrug users and 30 drug-naive controls probably residing in or near Hamburg (Germany). All participants were recruited through advertisements in local newspapers and magazines, and through leaflets distributed at dance events and educational institutions in the Hamburg area. Matching - Groups were matched on age, gender, pre-morbid IQ, and (in drug users) use of drugs other than ecstasy. Groups were approximately matched on education. (Imaging data from this sample has already appeared in another publication (Buchert et al. 2003)).

Criteria for Inclusion - Having used ecstasy for at least 20 weeks at a self-reported frequency of at least weekly use of 2 or more tablets within a 48 h time period. Former Ecstasy Users - Self-reported lifetime consumption of at least 250 tablets in women and 400 tablets in men, having used ecstasy for at least 3 years (36 months) but having ceased ecstasy use for at least 20 weeks prior to study date. Polydrug Users - No reported use of ecstasy, but use of other substances in amounts similar to those reported by ecstasy users. Non-drug Users - Never having used any illicit substance, but similar to other groups in age, gender, premorbid IQ and education. All groups - Ages 18-30, absence of major acute medical or psychiatric illness, no signs of epilepsy, not pregnant, and abstinent from all illicit substances for at least 6 days prior to study days, with abstinence verified through urinary analyses. Drug histories were also verified through hair analyses.

Drug Use Parameters - Current ecstasy users reported an average lifetime consumption of 817.1 +/- 1133.86 tablets, with an average dose per use of 3.95 +/- 2.53 tablets, and a maximum dose per use of 8.51 +/- 11.81 tablets. Current ecstasy users reported an average duration of use, in months, of 53.01 +/- 32.45 mo (range = 5-120 mo). Average time since last ecstasy use in current ecstasy users, in days, was 23 +/- 16.14 days. Onset of ecstasy use was at 19,97 +/- 3.77 years. Former ecstasy users reported lifetime consumption of 767.87 +/- 579.04 tablets, at an average dose per use of 4.06 +/- 3.88 tablets per use, and a maximum dose per use of 8.09 +/- 6.99 tablets. Former ecstasy users reported average duration of use, in months, as 54.52 +/- 28.13 mo. Average time elapsed from last ecstasy use to study day, in days, in former ecstasy users was 515 +/- 495 days. Former ecstasy users reported onset of ecstasy use at 18.16 +/- 2.76 years. Use of Other Drugs. Cannabis: Current ecstasy users reported lifetime consumption of 566.78 +/- 1187.98 joints, former ecstasy users reported an average lifetime consumption of 2132,91 +/- 2199.77 joints and polydrug users an average lifetime consumption of 1247.66 +/- 1290.57 joints. Cannabis was used by 29/30 current ecstasy users, 31/31 former ecstasy users and 28/29 polydrug users. Monthly cannabis consumption, in grams, was 7.76 +/- 18.29 g, in former ecstasy users was 18.46 +/- 28.57 g). Yearly consumption, in grams, in current ecstasy users, was 88.22 +/- 229.21 grams, 281.35 grams in former ecstasy users, and 141.91 +/- 150.76 grams in polydrug users. Amphetamine: Current ecstasy users reported an average lifetime consumption of 67.9 +/- 105.77 g, former ecstasy users an average lifetime consumption of 77.37 +/- 115.35 g and polydrug users an average lifetime consumption of 4.24 +/- 9.07 g. Amphetamine was used by 27/30 current ecstasy users, 27/31 former ecstasy users and 12/29 polydrug users. Monthly amphetamine consumption was 2.65 +/- 1046 g in current ecstasy users, none listed for former ecstasy users and 0.07 +/- 0.37 g for polydrug users, and yearly amphetamine consumption, in grams, was 15.15 +/- 34.55 g for current users, 8.91 +/- 45.26 g for former ecstasy users, and 0.98 +/- 4.64 g for polydrug users. LSD: Current ecstasy users reported average lifetime consumption of 1608.9+/- 5178.09 mcg, former ecstasy users an average lifetime consumption of 2436.88 +/- 5127.33 mcg, and polydrug users an average lifetime consumption of 160.29+/- 375.8 mcg. LSD was used by 22/30 current ecstasy users, 27/31 former ecstasy users and 11/29 polydrug users. Monthly LSD consumption was (in mcg) 17.92 mcg in current ecstasy users, and not used in former ecstasy users or in polydrug users. Yearly LSD use, in mcg, was 190.42 +/- 809.87 mcg in current users, 19.76 +/- 49.77 mcg in former ecstasy users, and 8.62 +/- 37.96 mcg in polydrug users. Psilocybin: Current ecstasy users reported average lifetime consumption of 0.85 +/- 3.19 g, former ecstasy users an average lifetime consumption of 5.05 +/- 15.98 g, and polydrug users an average lifetime consumption of 14.41 +/- 32.52 g. Psilocybin, was used by 3 current ecstasy users, 6 former ecstasy users and 10 polydrug users. Monthly psilocybin consumption was 0.067 g in current ecstasy users, 0.31 +/- 1.23 g in former ecstasy users, and 0.069 +/- 0.37 in polydrug users. Yearly psilocybin consumption was 0.55 +/- 1.6 g, 2.12 +/- 8.4 g, and 1.47 +/- 3.7 g in polydrug users. Cocaine: Current ecstasy users reported an average lifetime consumption of 39.10 +/- 76.43 g, average lifetime consumption in former ecstasy users of 101.41 +/- 218.86 g, and polydrug users an average lifetime consumption of 254.66 +/- 708.53 g. Cocaine was used by 28/30 current ecstasy users, 30/31 former ecstasy users and 22/29 polydrug users. Monthly cocaine consumption was 0.39 +/- 0.88 g in current ecstasy users, 0.27 +/- 1.01 g in former ecstasy users and 1.85 +/- 5.49 g in polydrug users. Yearly cocaine consumption was 6.22 +/- 15.11 g in current ecstasy users, 5.332 +/- 14.11 g in former ecstasy users, and 16.10 +/- 36.69 g in polydrug users. Alcohol: Current users reported an average weekly use of 87.88 +/- 96.19 g, former ecstasy users an average weekly use of 137.09 +/- 190.63 g, polydrug users an average weekly use of 195.42 +/- 194.28 g, and non-drug users average weekly use of 61.35 +/- 73.17 g. Tobacco: Current ecstasy users reported an average weekly use of 51.20 +/- 57.94 cigarettes, former ecstasy users, an average weekly consumption of 94.02 +/- 73.44 cigarettes, polydrug users a weekly consumption of 125.31 +/- 83.05 cigarettes and non-drug users a weekly use of 14.13 +/- 35.73 cigarettes.

Group Demographics and Matched Variables - The researchers matched all groups on age, gender, pre-morbid IQ, and (approximately) education, and they matched all drug-using groups on use of drugs other than ecstasy. Gender, as M/F Ratio - Current ecstasy users, 15/15; former ecstasy users, 16/15; polydrug users, 15/14; non-drug users, 15/15. Age - Average age of current ecstasy users was 24.5 +/- 4, average age of former ecstasy users was 24.13 +/- 4.21, average age of polydrug users was 24.41 +/- 4.55 and average age of non-drug users was 23.13 +/- 3.67. Education - (Originally presented as number of people who received basic (9 years), intermediate (10 years) or advanced (13 years or more) education). Current ecstasy users had attained approximately 10.9 years education (8 basic, 10 intermediate, 12 advanced), former ecstasy users attained approximately 10.6 years education (8 basic, 11 intermediate, 12 advanced), polydrug users attained approximately 10.9 years education (5 basic, 13 intermediate, 11 advanced), and non-drug users attained approximately 11.1 years education (7 basic, 10 intermediate, 13 advanced). Pre-morbid IQ (as measured via German-language vocabulary test, MWT-B, and German Wilde Intelligence Test (WIT)) - Estimated IQ for current ecstasy users was 102.5 +/- 9.64, for former users, 106.48 +/- 13.94, for polydrug users, 104.28 +/- 9.9, and for non-drug users, 104.97 +/- 13.47.

Measures: Mood and Self-Reported Psychological Problems - Measured via SCL-90-R, a self-report measure of psychological problems (e.g. anxiety, depression, interpersonal sensitivity, somatisation, etc.) Cognitive Function, Attention - Measured with the "Go/No Go" and "Divided Attention" tests of the TAP. Psychomotor Speed - Assessed with Trail Making Test-A. Executive Function - Assessed via Wisconsin Card Sorting Test (WCST). Learning and Memory - Assessed via "telephone numbers" and "company signs," two sub-tests from a German test battery (Lernund Gedachtnistest 3, or LGT-3), requiring retention of information over 30 minutes. Verbal memory was assessed through AVLT (list learning and recall), and Rivermead Behavioral Memory Test (RBMT) -Story, (story recall).

Imaging with McN5652 - (Also previously described in Buchert et al. 2003). All participants underwent a full-body PET scan with 466 +/- 76 MBq [11-C]-(+)-McN5652 with an infusion time of 4 minutes, and with a scan time of 90 minutes. Mesencephalon, putamen, caudate and thalamus were selected as volumes of interest (VOIs) in order to test hypotheses concerning serotonin transporter availability. White matter served as control tissue, and cerebellar gray matter served as a reference site for kinetic modeling. Distribution volume ratios for (+)-McN5652 were calculated to reflect receptor density.

Analyses: A one-way analysis of variance (ANOVA) was used to analyze all group differences (mood, cognitive function and imaging), with drug use serving as the dependent variable (current ecstasy user, former ecstasy user, polydrug user and non-user control). Analyses of covariance (ANCOVAs) were performed to see whether participant gender and selected drug use parameters were associated with scores or values on dependent measures, and whether controlling for covariates, once identified, altered findings of group differences. Drug use parameters were yearly cannabis, amphetamine, cocaine, LSD, psilocybin and ecstasy exposure, and average ecstasy dose per use (referred to in the paper as typical number of ecstasy exposures, or TE). Measures in current and former ecstasy users were compared with measures in polydrug users via planned comparisons. Regression Analyses - A stepwise regression was performed on all measures of mood, cognitive function and imaging data, using each score or value as a dependent measure, and with the predictors being maximum and average ecstasy dose per use and lifetime consumption of ecstasy, in tablets, yearly drug consumption of cannabis, amphetamine, cocaine, LSD, psilocybin and ecstasy.

Results - Significant Differences Found: Mood and Self-Reported Psychological Problems - All drug-using groups had significantly higher Global Severity scores on the SCL-90-R than non-drug users (Current ecstasy = former ecstasy = polydrug user > non-drug user). Current ecstasy users, former ecstasy users and polydrug users had higher global, interpersonal sensitivity, and obsessive-compulsive scores than non-drug user controls, and current and former ecstasy users had higher depression scores than non-user controls. Only former ecstasy users had higher anxiety scores than non-user controls. Average ecstasy dose per session was a significant covariate for global, anxiety, and psychoticism scores. Both gender and average ecstasy dose per use served as covariates for somatisation. Yearly cannabis use was a significant covariate for obsessive-compulsive and aggression scores.

Regression Analysis - Mood: Average ecstasy dose per use best predicted somatisation, anxiety, depression and paranoid ideation. Average ecstasy dose per use and yearly cannabis use best predicted SCL-90-R general score, phobic anxiety, obsessive-compulsive and phobic anxiety scores. Yearly cannabis use, yearly LSD use and average ecstasy dose per use best predicted aggression scores.

Cognitive Function - Planned comparisons found that polydrug users performed less well (higher perseveration scores) than all other groups on the WCST. Former ecstasy users had lower RBMT-Story scores than non-drug users. Former ecstasy users performed less well on both measures of learning and memory in the AVLT compared to non-drug users, and former ecstasy users' performance was below that of current ecstasy users on one AVLT sub-test degree of progress in list learning (ALVT 5-1). Polydrug users had lower AVLT 6 scores (memory for new list, interference) than non-drug using controls. Gender and amount of cocaine consumed in the last year were covariates for performance on AVLT sum of first 5 trials (AVLT1-5).

Regression Analysis - Immediate recall (ALVT 1 was best predicted by average ecstasy dose per use. ALVT 8 (delayed recall) was best predicted by lifetime ecstasy consumption. Yearly cannabis consumption was the best predictor for ALVT 6 (new list), and both RBMT-Story scores (immediate and delayed). Yearly cocaine use best predicted ALVT 5-1 (list learning). Imaging with McN5652 - As previously reported (Buchert et al. 2003), current ecstasy users had lower McN5652 binding in mesencephalon than subjects in all other groups (current ecstasy users < former ecstasy users = polydrug users = non-drug users). Current ecstasy users also had lower caudate McN5652 binding than polydrug users. Average ecstasy dose per use served as a covariate predicting binding in caudate; an interaction between average dose per use, gender and yearly LSD use served as covariates predicting binding in thalamus. Gender and yearly LSD use served as covariates for binding in mesencephalon.

Regression analysis - Average ecstasy dose per use best predicted McN5652 binding in thalamus and caudate. Yearly ecstasy use was the best predictor of mesencephalon McN5652 binding.

Results - No Significant Differences: Mood and Self-reported Psychological problems - When SCL-90-R scores for current and former ecstasy users were compared with polydrug user controls, no significant differences were found. When analysis of covariate treated all drug use parameters as covariates, between-group differences in obsessive compulsion, aggression and psychoticism scores were eliminated.

Regression analysis - Lifetime ecstasy consumption and maximum dose per use did not significantly predict mood scores. Yearly cannabis use alone did not significantly predict mood scores.

Cognitive Function - Members of all four groups performed similarly on intelligence tests (used for matching). All four groups performed similarly on the TAP (attention), the Trail-Making Test-A (psychomotor speed), and all other WCST scores save perseverance. There were no significant between-group differences in performance on two tests of memory, "telephone numbers" and "company signs." Current ecstasy users did not perform significantly differently from ex-ecstasy users, polydrug users or non-drug users on all AVLT scores. Average ecstasy dose per use, yearly cannabis use, amphetamine use, LSD use and psilocybin use were all not significantly associated with performance on AVLT, RBMT-Story or WCST.

Regression analysis - Average ecstasy dose per use did not predict RBMT Story, ALVT 8 or AVLT 5-1 scores. Yearly cannabis use did not predict ALVT 1, ALVT 8 or either RBMT Story score.

Imaging with McN5652 - (See also Buchert et al. 2003) Former ecstasy users did not have significantly different levels of McN5652 binding in mesencephalon, caudate or thalamus when compared to either polydrug using or non-drug using controls. When an analysis of covariance including other drug use parameters (i.e. yearly LSD, amphetamine, cannabis use) and gender, group differences in McN5652 binding in caudate were no longer significant. Regression Analysis - The following drug use parameters did not predict McN5652 binding in any selected brain region (mesencephalon, caudate, thalamus); yearly cannabis, LSD, amphetamine and cocaine use) Neither lifetime ecstasy consumption nor maximum dose per use predicted McN5652 biding in any of these brain regions.

Overall Effects: The effects of ecstasy use on mood, cognition and markers of serotonin transporter binding were explored in a large sample of current and abstinent ecstasy users polydrug user and non-drug user controls, with all participants recruited from the dance event community. All three drug-using groups had higher scores on a measure of various negative mood states and psychological problems. However, only current and former ecstasy users, and not polydrug user controls, had higher depression scores than non-drug users. Analyses of covariance and regression analysis suggest that use of other drugs (including yearly cannabis and LSD consumption) and average ecstasy dose per use all influence self-reported mood and psychological problems across groups. Though all groups performed similarly on some tests of memory, former users had lower scores on the AVLT and RBMT-Story when compared with non-drug user controls. On the other hand, current ecstasy users' performance on ALVT and RBMT-Story was not significantly different from that of polydrug users or non-drug user controls. Current and former ecstasy users and controls performed similarly on tests of attention and psychomotor speed, and polydrug users performed worse than all other groups on WCST perseverance, while all groups performed similarly on other WCST scores. Analyses of covariance and regression analysis suggest that both ecstasy use parameters and use of other drugs (cannabis, cocaine) may affect neurocognitive test performance. Imaging with the radioactive ligand McN5652 detected lower ligand binding (marker of lower serotonin transporter levels) in mesencephalon and caudate in current, but not former, ecstasy users. A regression analysis found that when other variables were used as predictors, differences in caudate binding disappeared. The specific hypotheses tested were only partially confirmed. While current ecstasy users had lower McN5652 binding than controls, and former users had lower scores on measures of learning and memory, few changes in mood, cognitive function or serotonin transporter binding were found only in current and former ecstasy users. Furthermore, mood and psychopathology were generally associated with drug use in general, and not ecstasy use in specific.

Comments: This study of brain serotonin function, cognitive performance and mood in current and abstinent ecstasy users employed one of the largest sample sizes to date, and combined assessments from three areas of interest. Taken together, the results demonstrate the difficulty of establishing associations between use of any one drug, and group differences in any one domain. It does appear that ecstasy use uniquely reduces brain serotonin transporter levels, but levels are no longer lower after approximately 4.5 months' abstinence. On the other hand, while this study failed to find impaired verbal memory in current ecstasy users, former users did have impaired verbal memory when compared to non-user controls, and in one case, scores on a measure of learning and memory were lower than those attained by current ecstasy users. Findings of elevated negative mood in former ecstasy users contradicts previous findings indicating a decline in self-reported negative mood and psychological problems in abstinent ecstasy users (Morgan et al. 2002). Since this study found an association between general drug use and psychological problems, this finding may be the result of greater drug use in at least some former ecstasy users. Findings on memory contradict previous findings of impaired verbal recall in current users, often detected through the same assessments used here (i.e. Fox et al. 2001; McCann et al. 1999; Morgan et al. 1999; Rodgers et al. 1999). (It should be noted that not all papers have found these differences (Simon et al. 2002).) While imaging findings are in agreement with earlier reports employing the same ligand (McCann et al. 1998), the degree of reduction reported in this study is far less pronounced (between 4% and 6%, versus between 20% and 40%, and potentially much higher, in the work of McCann and colleagues (see Kish 2002). As has been seen in other studies (Gijsman et al. 2002 versus Verkes et al. 2000), impaired verbal recall is not associated with presumed markers of serotonin function. There are numerous ways of interpreting study data. It is possible that reduced McN5652 binding reflects either harm to serotonin axons, downregulation of serotonin transporter sites, or both, with return to control levels after abstinence a result of compensation, upregulation and normalization, or a mixture of both processes. Impaired verbal recall may be affected by use of other drugs, or it may result from upregulation or compensation in serotonin function. An examination of this paper strongly suggests that in future, investigators should take care to assess and control for effects produced by use of other drugs. Like most other research in this area, the study was retrospective, and it still appears that groups were not matched on some parameters of cannabis and amphetamine use. However, this is one of the few studies that sought to control for a number of variables, including confirming self-reported drug use with hair and urinary analysis, recruiting subjects from the same population and seeking to match subjects for drug use in all applicable groups.