Purpose: Neurochemical, neurocognitive: to see whether Ecstasy users were differentially affected by altering brain serotonin levels through manipulating available tryptophan, and to assess mood and cognitive function at baseline and after tryptophan manipulation. Specific hypotheses tested -- Former, but not current, Ecstasy users will have a similar reaction to tryptophan challenge as controls if Ecstasy effects are reversible, but if Ecstasy use produces irreversible effects, both current and former users would react differently to tryptophan challenge than controls. Alternately, former users might respond differently than current users and controls if Ecstasy-induced changes are only apparent after cessation of use (last hypothesis possibly added after data analysis).

Design: Retrospective (non-experimental) between-subjects /within subjects design, with drug use (current Ecstasy user, former Ecstasy user, or non-Ecstasy user) serving as one between-groups variable, tryptophan manipulation (tryptophan depletion, tryptophan augmentation) serving as another between-groups variable, and time of measurement (baseline, after manipulation) serving as a within-subjects variable. Half the subjects received a tryptophan-free amino acid drink (tryptophan depletion), the other half received a tryptophan-containing drink (tryptophan augmentation), and all underwent assessments of mood and cognitive function.

Subjects: 32 current Ecstasy users, 32 former Ecstasy users and 32 non-drug using controls probably residing in or near London (England) and recruited via advertisements placed in unspecified locations. Matching -- Groups matched on gender, general IQ, and body mass index (BMI).

Criteria for Inclusion, Ecstasy Users, -- Reported having taken Ecstasy on more than 20 occasions in the past year – Former Ecstasy Users,-- Having taken Ecstasy on at least 20 occasions in one year, and abstinence from Ecstasy use for at least one year prior to the study Non-drug users, -- No reported Ecstasy use, but regular use of cannabis and alcohol permitted, All Groups -- Male, aged 18-40 years old, normal range BMI, not taking psychotropic medication or receiving psychological treatment, fewer than 32 errors on NART (measure of pre-morbid verbal IQ), BDI scores lower than 30 and STAI scores lower than 60, abstention from MDMA or LSD for three weeks prior to study day, with abstinence verified via self-report only, and abstinence from all psychoactive substances for 24 hours prior to study day, with abstinence verified through urinary analysis.

Drug Use Parameters -- Lifetime consumption is not reported for current or former Ecstasy users. However, computation using other drug use parameters indicates that current Ecstasy users probably consumed Ecstasy on 180.82 ± 63.46 occasions, and took 466.52 ± 57.75 tablets over a lifetime. Average dose per use for current Ecstasy users was 2.58 ± 0.91 tablets, frequency of use, in times per month, was 3.48 ± 1.83 times per month, and duration of use (in months), was 51.96 ± 34.68 months. Time since last Ecstasy use for current users, in days, was 39 ± 22.63 days. Lifetime consumption in former Ecstasy users is estimated to be 291.1 ± 134.76 occasions, and 707.37 ± 185.97 tablets. Former Ecstasy users’ average dose per use was 2.43 ± 1.38 tablets per use, frequency of use was 6.95 ± 4.27 times per month, and duration of use was 41.88 ± 31.56 months. Time since last Ecstasy use reported by former users, in days, was 841.3 ± 637.4 days. Other drugs -- (No information provided on lifetime consumption or number of individuals per group reporting use, only monthly use and duration of use). Cannabis: Current users’ average duration of use was 80.4 ± 50.4 months, average frequency of use (days per month) was 17.5 ± 12 days per month, and dose per month (in ounces) was 0.6 ± 0.5 oz. Former Ecstasy users’ average duration of use was 86,4 ± 61.2 months, average frequency of use was 14.1 ± 12.2 days per month, and average dose per month (in ounces) was 0.6 ± 0.9 oz. Non-user controls’ average duration of use was 88.8 ± 80.4 months, average frequency of use was 17.1 ± 11.2 days per month, and average dose per month (in ounces) 0.7 ± 0.7 oz. Alcohol: Current users’ average duration of use was 111.6 ± 67.2 months, average frequency of use was 18.2 ± 9.3 days per month, and average dose per use (in units) was 6.7 ± 3.6 units. Former users’ average duration of use was 114 ± 64.8 months, frequency of use was 10.6 ± 9.3 days per month, and average dose per use (in units) was 5.6 ± 4.4 units. Non-drug users’ average duration of use was 92.4 ± 78 months, average frequency of use was 14.1 ± 8.4 days per month, and average dose per use (in units), was 7.2 ± 4.8 units. Tobacco: Current users’ average duration of use was 81.6 ± 72 months, average frequency of use was 18.3 ± 14.5 days per month, and average dose (g per session) was 7.4 ± 7.2 g. Former Ecstasy users’ average duration of use was 100.8 ± 96 months, average frequency of use was 18.7 ± 14.7 days per month, and average dose per use (in g) was 7.8 ± 7.8 g. Non-drug users’ average duration of use was 72 ± 97.2 months, average frequency of use was 17.7 ± 14.3 days per month, and average dose per use (in g) was 7.1 ± 7.0 g. Cocaine: Current Ecstasy users’ average duration of use was 24 ± 25.2 months, average frequency of use was 0.9 ± 1.7 days per month, and average dose per use (in g) was 0.2 ± 0.3 g. Former Ecstasy users’ average duration of use was 37.2 ± 42 months, average frequency of use was 2.7 ± 6.3 days per month, and average dose per use (in g) was 0.2 ± 0.6 g. Non-drug users did not report cocaine use. Amphetamine: Current users’ average duration of use was 43.2 ± 38.4 months, average frequency of use was 0.6 ± 1.1 days per month, and average dose per use, in g, was 0.2 ± 0.5 g. Former Ecstasy users’ average duration of use was 40.8 ± 33.6 months, average frequency of use was 2.3 ± 3.5 days per month, and average dose per use, in g, was 0.4 ± 0.7 g. Non-drug users did not report amphetamine use. LSD: Current users’ average duration of use was 49.2 ± 56.4 months, average frequency of use was 0.6 ± 1.7 days per month, average dose per use not listed. Former Ecstasy users’ average duration of use was 22.8 ± 24 months, average frequency of use was 1.9 ± 3.6 days per month, and no average dose per use listed. Non-drug users did not report LSD use. Ketamine: Current users reported duration of use of 3 ± 1.7 months, and average frequency of use of 0.9 ± 2.6 days per month. Neither former Ecstasy users nor non-drug users reported using ketamine. Benzodiazepines: Current users’ average duration of use not listed (probably less than 6 months or sporadic), and average frequency of use was 0.4 ± 2.1 times per month. Neither former Ecstasy users nor non-drug users reported using benzodiazepines. GHB: Current Ecstasy users do not report using GHB. Former Ecstasy users’ average duration of use not reported (presumably under 6 months or sporadic), and average frequency of use was 0.5 ± 2.2 times per month, with no average dose per use provided. Non-drug users also did not report any GHB use. Lastly, "20-30%" of non-drug users had previously consumed amphetamines, cocaine or LSD at least once, but no information presented concerning amount used or frequency of use.

Group Demographics and Matched Variables -- The researchers matched all subjects by gender and estimated verbal IQ (NART score). Gender, as M/F ratio -- Current Ecstasy users, 32/0; former Ecstasy users, 32/0; Non-drug users, 32/0. Age – Average age of current Ecstasy users was 25.22 ± 5.91 years, 27.83 ± 5.92 years in former users, and in non-drug users, 23.97 ± 6.10 years. Education -- Educational information coded as having university education, completing university-level exams, or passing "general," non-university level exams. Current Ecstasy users had attained approximately 13.4 years of education (58% university degree level education, 23% at least university-admittance exams, and 19% "non-university exams). Former Ecstasy users had attained 12.91 years of education (52% university degree level education, 12% had completed pre-university exams, and 36% had completed non-university exams), and non-drug users had 13.85 years of education (72% university level education, 12% completed pre-university exams, 16% completed non-university exams.) Verbal IQ - Current Ecstasy users had an estimated verbal IQ of 111.38 ± 7.82, former Ecstasy estimated verbal IQ was 111.27 ± 8.93, and non-drug users’ estimated verbal IQ was 113.54 ± 9.8. BMI figures not supplied for any of the samples.

Measures: Tryptophan measures -- Free (not bound to cell membranes or proteins, more able to cross blood-brain barrier) and total (bound and unbound) plasma tryptophan was assessed in blood collected "before" (presumably immediately before) and 5 h after administering tryptophan manipulation. Plasma tryptophan detected with HPLC.

Cognition -- Memory was assessed via immediate and delayed RBMT story recall, and with the Buschke Selective Reminding Task, an RAVLT-like list learning task (16 words), but with selective reminders for any words not recalled in a previous trial. Three trials were presented, and a delayed recall trial 30 minutes later. Working memory and concentration -- Measured via Serial Sevens task (count backward in sevens). Executive Function – assessed via verbal fluency task (word generation) , with H and L as starting consonants and fruits and vegetables as categories. Attention -- Focused attention was assessed via "cancellation tasks" (cross out one or two targets, scored by completion time and number of errors), and sustained attention was assessed via rapid visual information processing (RVIP) task. Involving response on viewing three consecutive even or odd numbers among 100 digits per minute for 10 minutes, and then performing the same task while listening to and counting a sequence of low and high tones, and counting number. RVIP is scored by number of correct responses and response time (RT).

Mood and personality -- Baseline depression and anxiety assessed with BDI and Trait anxiety subtest of STAI. Baseline impulsiveness was assessed with Baratt Impulsiveness Scale (BIS), and self-reported aggression was assessed with the Aggression Questionnaire. Current mood at baseline and after tryptophan manipulation was measured via 16-item visual analog scales (VAS) assessing alertness, contentedness and calm. Current depressive mood (at baseline and post-tryptophan manipulation) was measured via modified BDI (instructions changed to refer to current state), and anxiety was measured with State STAI score. Physical effects of tryptophan manipulation (drink) were assessed via another VAS.

Analyses: A repeated measures analysis of variance (ANOVA) was used to examine differences in most variables, including measures of plasma tryptophan, cognitive function and mood, with drug use (current Ecstasy user, former Ecstasy user, non-Ecstasy user) serving as one between-group variable, tryptophan manipulation (tryptophan depletion, tryptophan augmentation) as the other between-group variable, and time of measure (before or after tryptophan manipulation) as a within-subjects variable. P. was set at 0.05, and post-hoc comparisons employed Bonferroni corrections. In self-paced tasks, errors were covaried with time to completion. If group differences were detected prior to tryptophan manipulation, than an analysis of covariance (ANCOVA) was performed, with baseline scores used as covariates for post-manipulation scores, but otherwise replicating the original analysis of variance (drug use and tryptophan manipulation as between-group variables)..

Drug Use Parameters and Outcome Variables -- Correlations were performed to detect relationships between Ecstasy use parameters and differences in measures of plasma tryptophan levels, cognitive function and mood.

Results-Significant Differences Found: Tryptophan Measures After tryptophan augmentation, former Ecstasy users had higher levels of total and free plasma tryptophan than either current Ecstasy users or non-users (Former ecstasy users > current Ecstasy users = non-Ecstasy users).

Cognitive Function Former Ecstasy users, but not current Ecstasy users or non-Ecstasy users, had poorer delayed prose recall after tryptophan depletion and improved delayed prose recall after tryptophan augmentation. (See "Results-No significant differences" on trends relating to immediate recall). Current and former Ecstasy users had lower immediate recall scores post-manipulation than non-Ecstasy users on the Buschke Selective Reminding Task, regardless of manipulation (tryptophan depletion or augmentation), indicating shallower learning curves in the Ecstasy user groups. Former Ecstasy users had lower baseline delayed recall scores on the BSRT and more errors on the serial sevens task than either current Ecstasy users or non-Ecstasy users. Non-Ecstasy users had faster RTs on double digit cancellation after tryptophan manipulation (that is, they had a practice effect), while current Ecstasy users showed less of a practice effect, and former users had even less of a practice effect than subjects in the other two groups (Non-drug users > current users > former users). At baseline, former Ecstasy users made a greater number of errors on the RVIP task than current Ecstasy users or non-Ecstasy users. Tryptophan depletion increased number of correct RVIP responses for all groups, and tryptophan augmentation reduced number of correct answers for all groups. Former Ecstasy users had slower RTs on the RVIP than current Ecstasy users or controls prior to tryptophan manipulation. (See "Results-No Significant Differences" on trends for RVIP-plus-tones performance prior to tryptophan manipulation).

Mood and Personality -- Current Ecstasy users, former Ecstasy users and non-Ecstasy users rated themselves less alert and more calm on VAS measures after tryptophan manipulation, regardless of whether they underwent tryptophan depletion or tryptophan augmentation. Only current Ecstasy users reported increased visual sensitivity after tryptophan augmentation, and reduced visual sensitivity after tryptophan depletion. (No other groups showed any changes in these symptoms after tryptophan manipulation). Tryptophan augmentation produced higher nausea scores than tryptophan depletion, though both manipulations increased nausea from baseline.

Drug Use Parameters and Measures -- Baseline (pre-tryptophan manipulation) plasma total tryptophan levels were positively correlated with duration of Ecstasy use in former Ecstasy users; the longer someone reported using Ecstasy, the higher their plasma total tryptophan. Frequency of Ecstasy use was positively correlated with higher baseline levels of plasma total tryptophan in former Ecstasy users, with higher plasma total tryptophan associated with greater frequency of use. There was a negative association between changes in free and total plasma tryptophan levels after tryptophan augmentation and baseline immediate and delayed prose recall scores in former Ecstasy users, with lower plasma tryptophan associated with higher scores. Changes in free and total plasma tryptophan after tryptophan augmentation were also negatively associated with post-augmentation immediate and delayed story recall in former Ecstasy users. There was a negative association between changes in free tryptophan levels after tryptophan augmentation and changes in verbal fluency score (from baseline to post-manipulation) in current Ecstasy users, with a greater change in free tryptophan after augmentation associated with less change in verbal fluency. (The following correlations were found only for former Ecstasy users). Duration of Ecstasy use was negatively associated with degree of BSRT (AVLT-like task) learning over trials from baseline to post-manipulation, meaning that the longer duration lasted, the less improvement was seen on list learning.) Average dose per use was negatively associated with the baseline consonant-start (phonemic) verbal fluency test only, with higher average dose per use associated with lower word generation rates. Frequency of use was associated with fewer correct responses for visual targets on baseline RVIP-and-tones task. (The following correlations were significant for current users only). There was a negative correlation between duration of Ecstasy use and immediate and delayed story recall at baseline. Duration of Ecstasy use was also negatively associated with baseline letter (phonemic) verbal fluency performance, with longer duration of use associated with lower word generation rates. A negative correlation was found between time since last use of Ecstasy and hostility scores on the Aggressiveness Questionnaire for former Ecstasy users, with longer periods of abstinence associated with lower Hostility scores. There was a positive association between time since last use of Ecstasy and physical aggression scores on the Aggressiveness Questionnaire in current Ecstasy users, meaning that longer periods of abstinence (in current users) were associated with higher physical aggressiveness scores.

Results-No Significant Differences: Tryptophan Measures -- There were no significant differences in baseline plasma free or total tryptophan levels in current Ecstasy users, former Ecstasy users or non-drug users. There were no significant differences between plasma total and free tryptophan levels after tryptophan depletion across all three groups (current Ecstasy users, former Ecstasy users, non-drug users).

Cognitive Function -- At baseline, current Ecstasy users, former Ecstasy users and non-drug user controls performed similarly on immediate prose recall. (There were trends for better immediate prose recall across all groups after tryptophan augmentation than after depletion. There was also a trend for lower immediate prose recall scores in former Ecstasy users than in current Ecstasy users or non-Ecstasy users, both at baseline and post-manipulation, a trend for lower delayed prose recall scores in former Ecstasy users than in current Ecstasy users or non-Ecstasy users). Baseline BSRT immediate recall scores were similar in current Ecstasy users, former Ecstasy users and non-Ecstasy users. Tryptophan manipulation did not affect delayed prose recall for either current Ecstasy users or non-drug users. Tryptophan manipulation did not affect immediate or delayed recall on the BSRT in all three groups. Likewise, tryptophan manipulation did not affect delayed recall on the BSRT. At baseline, current Ecstasy users, former Ecstasy users and non-drug users performed similarly on verbal fluency. There were no group differences in verbal fluency after tryptophan manipulation. Tryptophan manipulation did not alter serial sevens task performance for any of the three groups (that is, current and former Ecstasy users and non-Ecstasy users receiving tryptophan augmentation did not perform better or worse than those receiving tryptophan depletion.) At baseline, there were no group differences on digit cancellation, tryptophan manipulation did not significantly alter digit cancellation performance, and none of the groups showed performance sensitivity to tryptophan manipulation. Number of correct responses and RT on the RVIP in all groups (current and former Ecstasy users, non-Ecstasy users) was unaffected by tryptophan manipulation. (There was a trend for former Ecstasy users to make a greater number of errors on the RVIP-and-tones prior to tryptophan manipulation.) Tryptophan manipulation did not affect RVIP-and-tones performance, either for correct responses to targets, RTs or correct counting of tones, and none of the groups were more sensitive to the effects of tryptophan manipulation on RVIP-and-tones performance, whether scored as RTs or number of correct responses.

Mood and Personality -- Current Ecstasy users, former Ecstasy users and non-Ecstasy users all had similar baseline scores on the following measures: BDI, Trait anxiety, BIS, and Aggressiveness Questionnaire (AQ only administered at baseline). Tryptophan manipulation did not alter current depression or anxiety in any of the three groups (current and former Ecstasy users, non-Ecstasy users). Except for changes in visual sensitivity in current Ecstasy users, tryptophan depletion did not significantly alter self-reported symptoms in all three groups.

Drug Use Parameters and Measures -- Frequency of Ecstasy use was not associated with changes in plasma free or total tryptophan after tryptophan augmentation. There were no correlations between parameters of Ecstasy use (duration or frequency of use, tablets per use, time since last use) and plasma free or total tryptophan in current users. There were no significant correlations between duration of Ecstasy use, frequency of Ecstasy use, or time since last use and baseline plasma free tryptophan in former Ecstasy users. There were no significant relationships between drug use parameters and baseline serial sevens, digit cancellation or category verbal fluency scores in current or former Ecstasy users. Current Ecstasy users’ immediate or delayed BSRT scores were not significantly associated with any drug use parameter assessed (though there was a trend for duration of use to be negatively correlated with total BSRT recall after tryptophan manipulation). Former Ecstasy users’ parameters of Ecstasy use were not correlated with immediate or delayed prose (story) recall. There were no significant correlations between baseline mood scores and plasma tryptophan values in current Ecstasy users, former Ecstasy users or non-Ecstasy users. (There was a trend for an association between duration of Ecstasy use and higher total Aggressiveness Questionnaire scores for former Ecstasy users).

Overall Effects: At baseline, free and total plasma tryptophan levels were similar in samples of male current Ecstasy users, former Ecstasy users and non-Ecstasy user controls, but former Ecstasy users had higher free and total tryptophan levels after tryptophan augmentation than did current Ecstasy users or non-user controls. Former Ecstasy users’ prose recall scores were also sensitive to tryptophan manipulation, with a decline in performance seen under tryptophan depletion, and with performance improving under tryptophan augmentation. Current users’ prose recall was not affected by tryptophan manipulation. At baseline, former Ecstasy users had lower scores than current Ecstasy users or non-Ecstasy users on a verbal recall task, a measure of working memory, and a measure of attention. Former Ecstasy users also had the slowest reaction times on the RVIP (attention measure) than current Ecstasy users or non-user controls. Yet all groups performed better (had more correct responses) on the RVIP attention task after tryptophan depletion than tryptophan augmentation. After completing the BSRT again, current and former Ecstasy users did not have as high scores as non-Ecstasy users, an indication of a shallower learning curve in both groups. Former Ecstasy users showed the smallest practice effect on double digit cancellation (measure of focused attention), followed by current users, with non-drug users showing the strongest practice effect. Though baseline scores for a number of tasks were generally lower in former Ecstasy users than for current Ecstasy users or non-Ecstasy users, these tasks were not sensitive to tryptophan manipulation. These included BSRT (verbal recall, list learning), serial sevens (working memory), verbal fluency (executive function), digit cancellation (attention), and RVIP with tones (attention). Only prose recall (verbal recall) and RVIP (attention) were sensitive to tryptophan manipulation, with opposite responses (tryptophan depletion reduced performance on prose recall, but increased number of correct answers on RVIP). However, former Ecstasy users with greater changes in free and total plasma tryptophan after tryptophan augmentation performed less well on prose recall, and greater changes in free (but not total) plasma tryptophan after augmentation were associated with less improvement in verbal fluency for current Ecstasy users. None of the groups were more sensitive to changes in mood or alertness after tryptophan manipulation, as all reported greater calmness and less alertness after either tryptophan depletion or augmentation. Only current Ecstasy users reported greater visual sensitivity after tryptophan augmentation, and reduced visual sensitivity after tryptophan depletion. Baseline plasma tryptophan levels were associated with duration and frequency of Ecstasy use in former users, with longer duration and greater frequency of use predicting higher baseline plasma tryptophan (though there were no significant between-group differences in baseline tryptophan levels). Duration of Ecstasy use was associated with lower BSRT (verbal recall) scores in former Ecstasy users only, and frequency of Ecstasy use was associated with fewer correct RVIP responses. Though no between-group differences were seen on verbal fluency, average dose per use was associated with lower consonant-(letter) verbal fluency in former users. However, duration of use was negatively associated with letter verbal fluency score, and immediate and delayed story recall in current Ecstasy users. In former Ecstasy users, time since last Ecstasy use was negatively associated with self-reported hostility (longer periods of abstinence associated with lower hostility scores), yet in current Ecstasy users, time since last use was positively associated with self-reported physical aggression (greater aggression after longer periods of abstinence). Taken together, it appears that former, but not current, Ecstasy users had different responses to tryptophan manipulation, performance on some measures of verbal recall and attention, though improvement in list recall after practice was lower in current and former Ecstasy users. Only the last of three possible hypotheses were confirmed; neither the hypothesis based on reversible Ecstasy-associated changes, nor the hypothesis based on irreversible changes were confirmed. Instead, only former Ecstasy users showed differences in outcome measures after tryptophan manipulation, suggesting that changes are only apparent after cessation of Ecstasy use. The cumulative study findings also suggest dissociations between the effects of drug use parameters on cognitive function and mood, and their effects on tryptophan levels, implying independence from any changes in serotonin function.

Comments: To date, this is the first report that uses tryptophan depletion and augmentation as means of studying serotonin function in Ecstasy users and controls. The researchers may have performed so many statistical tests and correlations to incur Type 1 Error (detecting a relationship that does not in fact exist). This seems especially likely when correlations between Ecstasy use and cognitive performance are not reflected in actual between-group differences in cognitive performance, as is the case for current users and prose recall. It is notable that baseline differences in cognitive performance are not reflected in differences in baseline plasma tryptophan levels, and that between-group differences exist in tasks found to be insensitive to tryptophan manipulation. This suggests that changes in cognitive function or mood may not be associated with changes in plasma tryptophan produced by regular Ecstasy use, and, by implication, that these changes cannot serve as indicators of changed or impaired serotonin function. It is important to note that this is the second investigation finding greater reductions in cognitive function in former Ecstasy users, and not in current Ecstasy users (Thomasius et al. 2003). Measures of cognitive function in both studies reporting this finding were similar to those used in studies that found impaired performance in current Ecstasy users (see, for example, Gouzoulis-Mayfrank et al. 2000; Morgan et al. 1999). Possible explanations for impaired cognitive function in former Ecstasy users include long-term changes arising from regular Ecstasy use, pre-existing factors leading to abstinence and reduced cognitive function, and effects of other drugs used. Since the study only examined findings in men, it is unclear whether similar effects should be expected in women, as gender differences in serotonin transporter density after Ecstasy use (but not cognitive function) have been previously reported (Reneman et al. 2001). Other limitations include retrospective sample design and the already noted large number of analyses performed in this study.