Schizophrenia Research 130 (2011) 222–227

Contents lists available at ScienceDirect

Schizophrenia Research

j ourna l homepage: www.e lsev ie r.com/ locate /schres

Differential effects of chronic cannabis use on preattentional cognitive functioning inabstinent schizophrenic patients and healthy subjects

Johannes Rentzsch a,⁎,1, Elise Buntebart a,1, Ada Stadelmeier b, Jürgen Gallinat a, Maria C. Jockers-Scherübl b

a Charité—Universitätsmedizin Berlin, Department of Psychiatry and Psychotherapy, Campus Mitte, Charitéplatz 1, 10177 Berlin, Germanyb Charité—Universitätsmedizin Berlin, Department of Psychiatry and Psychotherapy, Campus Benjamin Franklin, Eschenallee 3, 14050 Berlin, Germany

⁎ Corresponding author. Tel.: +49 30 450 517 007; fE-mail addresses: johannes.rentzsch@charite.de (J. R

(E. Buntebart), ada.stadelmeier@gmx.de (A. Stadelmeie(J. Gallinat), maria.jockers@charite.de (M.C. Jockers-Sch

1 Both authors contributed equally to this work.

0920-9964/$ – see front matter © 2011 Elsevier B.V. Adoi:10.1016/j.schres.2011.05.011

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 15 January 2011Received in revised form 8 May 2011Accepted 16 May 2011Available online 31 May 2011

Keywords:MMNERPPreattentionalCognitionCannabisSchizophreniaAbstinence

Introduction: A number of studies indicate a higher risk for psychosis as well as for neurocognitive deficits inhealthy cannabis users. However, little is known about the impact of cannabis use on outcome inschizophrenia. In fact, there is growing evidence that cannabis-using schizophrenic patients may showpreserved or even better neurocognitive performance compared to schizophrenic non-users.Methods:Wemeasured mismatch negativity (MMN) to investigate preattentional neurocognitive functioningin long-term abstinent chronic cannabis users with (SZCA n=27) and without schizophrenia (COCA n=32)compared to schizophrenic patients (SZ n=26) and healthy controls (CO n=34) without any chronic druguse.Results: Healthy cannabis users showed reduced frontal MMN compared to controls (p=0.036). In contrast,cannabis-using schizophrenic patients showed increased frontal MMN compared to schizophrenic patientswithout cannabis use (p=0.038). Comparing non-cannabis users, schizophrenic patients showed reducedfrontal MMN (p=0.001). No significant differences were found between CO and SZCA (p=0.27), and COCAand SZCA (p=0.50).

Conclusion: Results suggest that chronic cannabis use may have different effects on preattentionalneurocognitive functioning in schizophrenic patients when compared to healthy subjects. This may berelated to preexisting differences in the endocannabinoid system between schizophrenic patients and healthysubjects. However, due to the naturalistic design of the study, the results must be interpreted with caution.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

Cannabis is the most widely used illicit substance worldwide. It iswidely accepted that there is a link between chronic cannabis use andpsychosis (Cohen et al., 2008). Longitudinal population-based andbirth cohort studies have shown an increased risk for psychoticsymptoms and disorders in chronic cannabis users (Moore et al.,2007). Interestingly, the main psychotropic ingredient of cannabis,delta-9-tetrahydrocannabinol (Δ-9-THC) produces transient symp-toms as well as behavioral and cognitive deficits resembling thoseseen in schizophrenia (for overview see DeLisi, 2008). It was shown,for example, that Δ-9-THC produced an increase in the clinical ratingsof schizophrenia-like symptoms for up to 2 h in healthy subjects(D'Souza et al., 2004). Furthermore, cannabis or Δ-9-THC adminis-tration reduces working memory function and leads to alterations ofthe cognitive event-related potential P300 (Roser et al., 2008); for a

ax: +49 30 450 517 921.entzsch), e-lise@web.der), juergen.gallinat@charite.deerübl).

ll rights reserved.

review see (Solowij and Michie, 2007). Both of these occurrences areconsidered to be core features of schizophrenic pathophysiology(Vadhan et al., 2009).

Apart from the acute intoxication and the short-lasting ‘drugresidue’ effects, there is also evidence for long-lasting cannabis effects(Solowij, 1995; Bolla et al., 2002; Medina et al., 2007). Recently, wewere able to show that, in chronic cannabis users, a disturbance inearly sensory gating after short-term abstinence (Patrick and Struve,2000) is also evident after an abstinence lasting at least 28 days(Rentzsch et al., 2007). This may be an important finding, as a deficitin sensory gating is considered to be one of the robust biologicalmarkers of schizophrenia (Bramon et al., 2004).

Despite the fact that these experimental and epidemiologicalfindings support the link between cannabis use and psychosis, itremains unclear which effects chronic cannabis use has on schizo-phrenic patients (Zammit et al., 2008). While some studies reportedthat cannabis use is associated with a worsened outcome (Linszen etal., 1994; Caspari, 1999; Foti et al., 2010), an increase in schizophreni-form symptoms (Turner and Tsuang, 1990) and a lower age of onset(Linszen et al., 1994; Jockers-Scherubl et al., 2003; Arendt et al., 2005),other studies were unable to confirm these findings after controllingfor potential confounders (Peralta and Cuesta, 1992; Compton et al.,

223J. Rentzsch et al. / Schizophrenia Research 130 (2011) 222–227

2004; Veen et al., 2004; Dubertret et al., 2006; Boydell et al., 2007;Selten et al., 2007). To complicate things further, recent studiesshowed that cognitive functions were preserved or even improvedin cannabis using schizophrenic patients (Coulston et al., 2007;Jockers-Scherubl et al., 2007; Sevy et al., 2007; Potvin et al., 2008;Schnell et al., 2009; Yucel et al., 2010). One difficulty in accuratelyinterpreting these results is the fact that they lack healthy cannabisusers as a control group. It therefore remains unclear whether thedifferences are specific to schizophrenia or not. This question may beof crucial importance as studies have shown differences in theendocannabinoid system between healthy and schizophrenic canna-bis users (Leweke et al., 1999; Giuffrida et al., 2004).

In 2003 and 2004 we published our first studies on schizophrenicpatients and healthy subjects with chronic cannabis use. We wereable to show that chronic cannabis use by schizophrenic patients butnot by healthy subjects was related to a significant change in serumlevels of the neurotrophic factors NGF and BDNF (Jockers-Scherublet al., 2003, 2004). Recently, we found that chronic cannabis use wasassociated with preserved or even increased neurocognitive func-tioning in abstinent schizophrenic patients compared to non-drugusing schizophrenic patients. In contrast, abstinent healthy cannabisusers showed impaired neurocognitive functioning (Jockers-Scherublet al., 2007; Rentzsch et al., 2007). These studies support the idea thatchronic cannabis use has differential effects on schizophrenic patientsand healthy individuals.

To increase the specificity of these results, we are now conductinga second study to assess the differential effect of chronic cannabis useon preattentive neurocognitive functioning using auditory evokedmismatch-negativity (MMN). The MMN is of particular interest, sinceit was found impaired not only in schizophrenic patients (Naatanenand Kahkonen, 2009) but also in healthy cannabis users (Roser et al.,2010). As MMN is thought to be a marker of preattentive neu-rophysiological processes (Naatanen, 1995) it is thus a sensitivemarker for neuronal alterations.

2. Methods

This study was approved by the ethics committee of the Charité—Universitätsmedizin Berlin and has been carried out in accordancewith the Declaration of Helsinki. Healthy controls and healthycannabis user were recruited through web pages and newspaperadvertisements. All recruited schizophrenic patients were regular in-and out- patients undergoing treatment in the Department ofPsychiatry and Psychotherapy, Charité—Universitätsmedizin Berlin.All subjects agreed to participate in the study and gave their writteninformed consent.

2.1. Study sample

We investigated 27 schizophrenic patients with chronic cannabisuse but with little or no experience of other drugs (SZCA), 26schizophrenic patients without any chronic drug use (SZ), 32otherwise healthy subjects with chronic cannabis use but with littleor no experience of other drugs (COCA), and 34 healthy subjectswithout any chronic drug use (CO). Chronic cannabis use was definedas a consumption of at least 5 days per week for at least 1 year asindicated by self-reports. In the groups without chronic drug use alife-time cannabis consumption of less than 5 times with the lastconsumption of more than 1 year prior to study participation wasaccepted. Only cannabis users with an abstinence of at least 28 dayswere included. Abstinence was ensured by urine drug screening(Mahasan-Kombi-DOA4-Test). Subjects with a consumption of addi-tional drugs of more than 30 times, intravenous drug use at any time-point and with alcohol abuse/dependence or unhealthy consumptionwere excluded. Healthy control subjects were screened with astructured clinical interview (MINI Plus, German version) to exclude

any history of a psychiatric disorder. Schizophrenic patients werediagnosed according to DSMIV criteria by the consensus of twoindependent experienced clinical psychiatrists (SZ: paranoid n=15,disorganized n=3, undifferentiated n=7, katatonic n=1; SZCA:paranoid n=19, disorganized n=4, undifferentiated n=2, katatonicn=2). The diagnosis of each individual was based on the medicalhistory, including third party history and clinical follow-up informa-tion. No SCID interviews were used. Severity of symptoms wasassessed using the Positive and Negative Syndrome Scale. Althoughnicotine can influence cognitive functions, cigarette-smoking wasallowed to avoid nicotine withdrawal states before testing and wascontrolled for. As interviews and EEG preparation took at least onehour, acute nicotine effects were minimized.

All but three schizophrenic patients were medicated. Each patients'medication-status was classified as follows: (no cannabis/formercannabis use): no antipsychotics 1/2, one antipsychotic 18/15, twoantipsychotics 7/8, three antipsychotics 0/2, additional SSRI or SNRIfor the treatment of negative symptoms 9/6, additional lorazepam(max. 1.5 mg/day) 3/2, or other additional medications (biperidene[n=4], metoprolol [n=3], opipramole [n=1], pantoprazole [n=1],pirenzepine [n=1]). Two schizophrenic patients who formerly usedcannabis, received additional treatment with carbamazepine. Onepatient who did not use cannabis received additional treatment withzopiclone and pregabaline. All but two patients (one from eachgroup) received atypical antipsychotics.

2.2. ERP recording and analysis

EEG recording took place in an electrically shielded room.During therecording subjects were seated in a slightly reclined chair with a headrest watching a soundless cartoon. Subjects were instructed to focus onthe cartoon and not to attend to the auditory MMN stimuli (standardstimuli: 90% probability, duration 80 ms, 1000 Hz; duration deviantstimuli: 5% probability, duration 40 ms, 1000 Hz; frequency deviantstimuli: 5% probability, duration 80 ms, 930 Hz). The intertrail–intervalvaried between 350 and 650 ms. EEG was recorded using an electrodecap according to the international 10/20 system with additionalelectrodes. The impedance of electrodes was kept below 10 kΩ. TheEEG was amplified with a neuroscan system (Neuroscan SynAmps)digitized at 500 Hz. Offline MMN analysis was performed using BrainVision Analyzer (Brain Products GmbH, Germany). Eye movements,electrocardiac and muscle artifacts were corrected using independentcomponent analyses (Onton et al., 2006). After referencing to averagereference data were filtered (1 to 30 Hz, 50 Hz notch), segmented(−100 to 600 ms) and baseline corrected (−100 to 0 ms). Segmentsshowing amplitudes±75 μVwere excluded. Remaining segmentswereaveraged separately for standard, duration and frequency deviantstimuli. MMN were measured by the difference between the standardand the deviant evoked potentials at electrode Fz and Cz. The MMNamplitude was defined as the most negative peak between 110 and230 ms (duration deviant MMN) and 90–210 ms (frequency deviantMMN) according to the average of all subjects.Datawere transformed toabsolute values.

2.3. Statistical methods

Statistical evaluation was performed using the Statistical Packagefor the Social Sciences. All variables showed normal distribution(Kolmogorov–Smirnov test, pN0.05) and variance homogeneity(Levene test, pN0.05). Three-way analysis of variance (ANCOVA)was conducted with the factors group (CO, COCA, SZ, SZCA), stimulus(frequency, duration) and electrode (Cz, Fz). Post hoc t-tests wereapplied for pairwise group differences. Further ANCOVA was doneusing gender, age, education level, nicotine use and relationship statusas covariates. Exploratory, we included as covariates antisocialpersonality trait scores, life-time drug-use other than cannabis and

Table 1Demographic and clinical characteristics of the study sample. Differences within the four groups were tested with ANOVA or Chi²-test and between two groups with t-tests. Datagiven as frequencies or mean (standard deviation).

SZ n=26 SZCA n=27 COCA n=32 COn=34

Age, yearsa,⁎ 34.6 (11.1)⁎ 28.1 (7.4)⁎ 29.8 (7.7) 29.5 (8.0)Gender, male/female, nb,⁎ 13/13⁎ 24/3⁎ 22/10 22/12Education level, high/low, nb,⁎ 10/16⁎ 8/19⁎ 22/10 24/10Partnership, no/yes, nb,⁎ 17/9⁎ 21/6⁎ 11/21⁎ 16/19⁎

Smoker/Nonsmoker, nb,⁎ 13/13⁎ 21/6⁎ 16/16⁎ 7/27⁎

Nicotine use, cigarettes per daya,⁎ 7.5 (9.23)⁎ 17.35 (11.52)⁎ 6.76 (8.35) 2.25 (5.48)Pre-IQa 28.9 (6.0) 29.7 (3.3) 29.9 (3.1) 31.2 (3.7)Antisocial behavior scorea,⁎ 1.5 (1.4)⁎ 3.1 (2.3)⁎ 3.4 (2.5)⁎ 1.9 (2.1)⁎

Family history, any/psychosis, n (%)a 7 (27)/5 (19) 13 (48)/6 (22) 8 (27)/1 (3) 6 (18)/2 (6)Other drug use, mean/n/max 0.1 (0.3)/3/1 7.4 (9.3)/19/30 10.8 (10.8)/25/29 0.09 (0.3)/3/1Duration of illness, yearsc 6.8 (3.7) 5.4 (5.1)Age at onset of schizophrenia, yearsc,⁎ 30.5 (9.1)⁎ 22.5 (5.8)⁎

Number of psychiatric hospital stays c 2.3 (1.5) 2.8 (2.4)PANSS-positivec 12.9 (3.8) 12.2 (3.8)PANSS-negativec,⁎ 19.9 (6.4)⁎ 16.4 (5.4)⁎

Chlorpromazine equivalentsc 689 655

SZ: schizophrenic patients without cannabis use; SZCA: schizophrenic patients with cannabis use; COCA: otherwise healthy subjects with cannabis use; CO: otherwise healthysubjects without cannabis use; Education level: “high level” corresponds to General Qualification for University Entrance/Certificate for Overall Maturity for Higher Education;Partnership: married or analogous partnership; Nicotine use: averaged smoked cigarettes per day in the last 6 months; Pre-IQ: Premorbid IQ was estimated by the multiple-choice-vocabulary test of intelligence (verbal IQ) which largely corresponds to the vocabulary subtest of the Wechsler Adult Intelligence scale (data were not available for one SZ, threeSZCA, eight COCA and six CO subjects); Antisocial behavior score: estimated by the sum of positive items of the antisocial subscale of the SCID-II screening questionnaire; Familyhistory: Self reports of positive family history for: any affective and psychotic disorder (“any”) or psychotic disorder (“psychosis”) of first, second and third-grade relatives; Other druguse: life-time use of other drugs than cannabis, mean: times used other drugs, n: number of subjects used other drugs at least once, max: maximal life-time use of other drugs in thegroup; Age at the onset of schizophrenia: age of the first medical treatment for schizophrenic symptoms; PANSS: positive and negative syndrome scale.

a ANOVA: age: F(3,115)=2.8, p=0.04; nicotine use: F(3,115)=15.5, p=0.001; Pre-IQ (data missing for 1 SZ, 3 SZCA, 9 COCA, 6 CO): F(3,96)=1.4, p=0.27; Antisocial behaviorscore: F(3,115)=5.2, p=0.002; all others p≥0.1; Other drug use, mean: F(3,115), p≤0.0001 (post-hoc LSD-test, SZCA-COCA: p=0.07).

b Chi²-Test: gender: Chi2=9.5, p=0.02; education level: Chi2=15.5, p=0.001; Partnership: Chi2=13.8, p=0.003; Smoker/Nonsmoker: Chi2=19.9, p=0.001; Family history(2 missing data for COCA), “any”: Chi2=7.1, p=0.07; Family history, “psychosis”: Chi2=7.8, p=0.05.

c T-Test: PANSS-negative: t=2.2, p=0.03; Age at onset: t=3.4, p=0.001; all others pN0.1.⁎ pb0.05.

224 J. Rentzsch et al. / Schizophrenia Research 130 (2011) 222–227

family history of affective and psychotic disorders. Correlationanalysis between MMN and cannabis-related parameters was carriedout using Pearson and partial correlation. Statistical significance wasdefined as p≤0.05 and a statistical trend as p≤0.1.

3. Results

Table 1 presents demographic and clinical characteristics of thesample. There weremarked differences with regard to gender and agebetween the groups. Schizophrenic non-cannabis users showed wereolder than the other groups. Patients with schizophrenia had lowereducation levels than healthy subjects (low level (% of subjects), SZ:62, SZCA: 70, COCA: 31, CO: 29). Most of the schizophrenic patientswere not married and had no partner (% of subjects, SZ: 65, SZCA: 78,COCA: 34, CA: 44). Most of the schizophrenic cannabis users werecigarette smokers while the healthy subjects who did not usecannabis were mostly non-cigarette-smokers (% of subjects, SZ: 50,SZCA: 78, COCA: 50, CO: 21). Premorbid-IQ's were found to be similarbetween the four groups, however it must be noted that a substantialamount of data describing the COCA and CO groups were missing. Asantisocial personality traits may be increased in cannabis users, thecorresponding subscales of the SCID-II screening questionnaire wasused to estimate self-rated antisocial behavior. Schizophrenic andhealthy cannabis users showed higher mean scores than the non-usergroups. Of both the schizophrenic and healthy groups, three subjectsin total had used a drug or drugs other than cannabis once in theirlifetimes. 19 subjects who were schizophrenic cannabis-users and 25subjects who were healthy cannabis users had used drugs other thancannabis between one and 30 times during their lifetimes. The meanvalue of life-time, non-cannabis drug-use in the last two groups wascomparable. Interestingly, schizophrenic patients with cannabis usereported more often that they had a family relative with either apsychotic, or both a psychotic and affective disorder. Schizophrenicpatients with cannabis use had lower age at onset of schizophrenia

and less negative symptoms which is in line with other studies (e.g.Peralta and Cuesta, 1992; Bersani et al., 2002; Jockers-Scherubl et al.,2004). Details of cannabis use and consumption of other drug thancannabis are presented in Table 2. Pattern of cannabis and drug use inthe cannabis groups were comparable. Schizophrenic patients withcannabis use had started regular cannabis use about 5.6 years beforedisease onset.

ANOVA showed significant differences between the four groups forfrequency but not duration deviant MMN amplitude at electrode Fz.There were no significant differences for either duration or frequencyMMN amplitude at the electrode Cz. Thus, further analyses were doneusing Fz frequency MMN.

As there were marked differences for sex, age, education level,nicotine use and partnership, these variables were used as con-founders in an ANCOVA. Analysis of covariance showed a significanteffect of group on frontal frequency MMN (F[3,110]=4.4, p=0.006,Eta2=11%) after controlling for gender (F[1,110]=1.7, p=0.19,Eta2=2%), age (F[1,110]=0.5, p=0.5, Eta2=0.4%), education level(F[1,110]=1.7, p=0.19, Eta2=2%), nicotine use (F[1,110]=0.27,p=0.61, Eta2=0.2%) and relationship status (F[1,110]=0.26,p=0.61, Eta2=0.2%). The corrected model was significant (F[8,110]=2.1, p=0.04, R2=13%, corrected R2=7%). Furthermore, lifetime use ofother drugs and antisocial behavior may be indicators for substantialgroup differences which, in turn, may influence preattentionalcognitive functioning. These variables were included in the abovemodel as additional confounders. Inclusion of the variables: lifetimeuse of other drugs other than cannabis and the antisocial behaviorscore in the analysis did not result in significant changes in thecalculated group effect on frontal frequency MMN, (F[3,109]=4.3,p=0.006, Eta2=11%) and (F[3,108]=4.5, p=0.005, Eta2=11%),respectively. However, the corrected main model failed to showstatistical significance, (F[9,109]=1.9, p=0.07, R2=13%, correctedR2=6%) and (F[10,108]=1.8, p=0.07, R2=14%, correctedR2=6%). Finally, a family history of affective and psychotic disorders

SZ SZCA COCA CO

p=0.001

p=0.038 p=0.036

1.2±0.5 1.5±0.7 1.4±0.6 1.7±0.5

0.5

1.0

1.5

2.0

2.5

3.0

µV

Fig. 1. Frequency MMN amplitude (μV) at electrode Fz. Values are given as mean andstandard deviation. P-values of pairwise t-test are shown for pb0.05. For details seeSection 3. SZ: schizophrenic patients without cannabis use, SZCA: schizophrenic patientswith cannabis use; COCA: otherwise healthy controls with cannabis use; CO: healthycontrols without cannabis use.

Table 2Details of cannabis consumption in the cannabis using groups expressed as means and(standard deviations). Differences between groups were tested using t-test.

COCA SZCA

Years of regular cannabis usea 7.9 (6.1) 6.5 (4.9)Average daily dose, in gramsa 1.1 (1.0) 1.07 (1.1)Age at onset of regular cannabis usea 18.0 (3.3) 18.5 (3.6)Duration of abstinence, in monthsa 19.3 (21.2) 27.4 (40.3)Drug use other than cannabis, meanb/nc/maxd

Amphetamine, speed 1.7 (3.1)/11/10 0.6 (1.3)/7/5MDMA, ecstasy 2.3 (3.7)/15/13 1.1 (2.4)/8/10LSD 1.1 (2.6)/10/11 1.1 (3.0)/9/15Cocaine 2.3 (3.6)/16/12 2.7 (4.9)/14/20Mushrooms 2.4 (3.5)/15/15 1.8 (2.4)/12/7Heroine, opiate 0.9 (3.9)/2/20 0.04 (0.2)/1/1

Different drugs usede, nOne to two drugs 13 9Three to four drugs 8 9Five to six drugs 4 1

Onset of regular cannabis use before onsetof schizophrenia, years (n=20)

5.6 (4.8)

Onset of first cannabis use before onset ofschizophrenia, in years (n=22)

6.3 (4.7)

SZCA: schizophrenic patients with cannabis use; COCA: otherwise healthy subjects withcannabis use.

a T-test: all p≥0.1.b Times used this drug in life, mean: for all, but amphetamine/speed (p=0.09),

p≥0.1.c Number of subjects used this drug at least once in life-time.d Maximal life-time use of this drug in the group.e Number of subjects who have used up to two, up to four or up to 6 drugs of different

classes other than cannabis (at least once).

225J. Rentzsch et al. / Schizophrenia Research 130 (2011) 222–227

was added to the analysis, but without any substantial changes:group effect (F[3,107]=4.7, p=0.004, Eta2=12%), corrected model(F[11,107]=1.7, p=0.09, R2=15%, corrected R2=6%). No con-founders in these models showed any statistically significant effecton frontal frequency MMN.

In the whole group of schizophrenic patients (SZ+SZCA: 1.37±0.56 μV), there was a statistical trend for a lower MMN compared tothe group of healthy subjects (CO+COCA: 1.56±0.56 μV, p=0.07).Pairwise comparison of MMN between the groups showed significantdifferences between CO and COCA (p=0.05), CO and SZ (p=0.001)and SZ und SZCA (p=0.04). No significant differences between COand SZCA (p=0.3), and COCA and SZCA (p=0.5) were observed (seeFig. 1). There were no statistic significant correlations between MMNand years of regular cannabis use, duration of abstinence and averagedaily dose (data not shown). There were no significant correlationsbetween medication dosages and MMN, neither in the non-cannabisnor in the cannabis using schizophrenic group (data not shown).

4. Discussion

The aim of this study was to investigate the differential effects ofchronic cannabis use on preattentive cognitive functioning (MMN) asobserved in abstinent healthy and schizophrenic subjects. To ourknowledge, this is the first study investigating MMN in abstinentschizophrenic and healthy chronic cannabis users.

The main finding of our study is that frequency-deviant MMNlocalized frontally is variable across the four groups described.Healthy cannabis users showed a reducedMMN compared to healthy,non-drug using subjects. In contrast, schizophrenic cannabis usersshowed an increased MMN compared to schizophrenic non-drugusers. These findings are consistent with our previous studies(Jockers-Scherubl et al., 2007; Rentzsch et al., 2007). Whether thedifferences in the frontal frequency-deviant MMN indicates differ-ences in frontotemporal cortex functioning between cannabis usingand non-using schizophrenic patients remains unclear. It was foundhowever, by Rasser et al. (2009), that the reduction in the frequency-

deviant MMN is correlated with frontal as well as temporal graymatter reduction in schizophrenic patients. However, as this study isbased on a naturalistic design, the direction of any causal relationshipremains unclear. Thus, it is not possible to distinguish between causeand effect. Furthermore, it cannot be excluded that other population-based confounders were responsible for the MMN group differences.We tried to control for confounders such as gender, age, educationlevel, nicotine use, relationship status, self-rating of antisocialbehavior, life-time drug-use other than cannabis and a family historyof affective and psychotic disorders. However, the group differences inthese variables could be an indicator for the possibility that factorsother than cannabis-use, e.g. genetic factors (Majic et al., 2011) or acannabis × gene interaction (Stadelmann et al., 2011), may accountfor the differences in preattentional cognitive functioning observed. Inthis context, schizophrenic patients with former chronic cannabis usemay represent a subgroup of schizophrenia patients with increasedpreattentional cognitive functioning. Thus, our conclusions have to beinterpreted with caution.

There are several possible explanations for the differential effectsof chronic cannabis use. Some explanations are based on a selectionbias presuming that schizophrenic patients who are better able tofunction socially (and have fewer negative symptoms) are thereforebetter able to obtain drugs of abuse. However, Dubertret et al. (2006)showed that schizophrenic patients using only cannabis exhibited lessnegative symptoms than those using additionally drugs of abuse. Thiscontradicts the selection-bias hypothesis if one assumes that other,less accessible drugs of abuse, such as heroin, require that the patientbe in a state of higher cognitive functioning and low negativesymptoms (Schnell et al., 2009). Other investigators have argued thatbetter cognitive functioning and less negative symptoms can beobserved in those patients with lower genetic loading for schizo-phrenia who therefore exhibit better social and cognitive functioning(Schnell et al., 2009). According to these studies, some of theseindividuals would never have developed schizophrenia had they notconsumed cannabis. Nevertheless, while this is one possible inter-pretation, it would contradict results showing that cannabis usingschizophrenic patients havemore cortical abnormalities, i.e. less gray-matter in the cingulate cortex (Szeszko et al., 2007; Bangalore et al.,2008). Furthermore, cannabis use has a negative impact on brainfunctions (Ehrenreich et al., 1999; Bolla et al., 2002; Solowij and

226 J. Rentzsch et al. / Schizophrenia Research 130 (2011) 222–227

Battisti, 2008; Battisti et al., 2010) such that abstinent healthycannabis users show cognitive deficits. One would expect that thiswould be the same in schizophrenic patients even when they have alower genetic load for schizophrenia. However, Selten et al. (2007)found no negative impact of ongoing cannabis use on schizophreniaoutcomes after controlling for gender in a 2.5 year follow-up.

Another interpretation of these results postulates that cannabis mayexhibit neuroprotective effects which would counteract a putativeneurotoxic process related to schizophrenia (Jockers-Scherubl et al.,2007; Potvin et al., 2008). Cannabinoids, including Δ-9-THC, are shownto have neuroprotective effects (Sarne andMechoulam, 2005; Stelt vander and Di, 2005). Hence, preserved neurocognitive functioning inschizophrenic cannabis users may reflect some neuroprotective effectsof cannabis. According to Schnell et al. (2009), this interpretationwouldconflict with the persistent cognitive impairments seen in chroniccannabis users. However, an explanation may be that chronic cannabisuse in healthy subjects causes neuronal dysfunctions, while havingdifferential or even beneficial effects in those subjects with pre-existingneuronal dysfunctions such as in schizophrenic patients. Whileextrapolating findings from animal to humans is limited, evidencefrom studies on rodents supports this view. Using the phencyclidine(PCP) model of schizophrenia, Spano et al. (2010) demonstrated thatchronic administration of a cannabinoid receptor 1 (CB1) agonist blockssubsequent PCP-induced behavioral abnormalities related to schizo-phrenia, i.e. disrupted prepulse-inhibition of the acoustic startle reflex,short-term memory deficits and social alterations. Furthermore, inanother animal study, the PCP-induced alterations in social interactionswere reversed after CB1 stimulation, while CB1 stimulation alonecaused the same behavioral deficit (Seillier et al., 2010). The results ofthese animal studies suggest that chronic cannabis-use may lead tobehavioral deficits in “healthy” animals, but also restore some functionsin animals with already existing neuronal “psychosis-like” abnormal-ities. It remains speculative whether the different effects of cannabis,depending of preexisting neuronal damage, may be of relevance inhumans. Thequality of the cannabis-related effectsmay further vary as afunction of brain region (Romero et al., 1995). Thus, there may becomplex interactions between the pathophysiologic state of schizo-phrenia (e.g. in the endocannabinoid system) and changes inneurotransmission induced by chronic cannabis use.

In this context it is of interest that chronic and acute cannabis usecaused “alpha hyperfrontality” and increased frontal alpha coherence(Struve et al., 2003). One can speculate that cannabis may correctfrontal dysfunction of schizophrenic patients (Pomarol-Clotet et al.,2008) as well as interhemispheric connectivity. The latter issupported by a study showing higher intermanual coordination—asan index of interhemispheric transfer—in (right handed) schizo-phrenic cannabis users who also showed less negative symptoms(Gorynia and Schwaiger, 2010).

That chronic cannabis use in schizophrenic patients may havepositive effects is contra-intuitive. However, while a negative impactof cannabis on the outcome of psychotic disorders is widely assumedby psychiatrists, Zammit et al. (2008) conclude in their meta-analysisthat there is “…little empirical evidence … currently available tosupport this view” (p. 362).

The main limitation of the study is the naturalistic design, whichcannot produce statements about the causal direction of therelationship between cannabis-use and -MMN nor can it produceany assurance that cannabis-use was the main causal factor for MMNdifferences between the groups. It remains unclear whether cannabis-use affects preattentional functioning or whether preattentionalfunctioning predisposes subjects to cannabis-use. However, theanimal studies give some support for our interpretation. Anothermain limitation is the retrospective characterization of life-time druguse obtained by self-reporting. Prospective studies will be necessary.Urinary drug test screening was used to control for abstinence. Due toaccumulation in body fat and slow excretion, Δ-9-THC can persist in

the urine of chronic users up to several weeks. Thus, it remainsunlikely that subjects used cannabis immediately prior to investiga-tion. However, long-term abstinence of drugs other than cannabiscould not be effectively controlled. Schizophrenic patients wereusually identified at the time of their first episode, including follow-updocumentation, thus rendering information about drug use andclinical parameters in a reasonably reliable manner. For the healthycannabis users, no such follow-ups were available.

Our results suggest that chronic cannabis usemay have differentialeffects on healthy and schizophrenic patients. Further research isneeded to determine whether the schizophrenic patients derived anybenefit from cannabis on neurocognition or whether they belong to asubgroup with lower genetic vulnerability.

Role of funding sourceThere was no main sponsor for this study. A. Stadelmeier was funded by a grant of

the Friedrich-Ebert Foundation, Germany.

ContributorsElise Buntebart and Johannes Rentzsch managed the literature searches and

analyses, undertook the statistical analysis and wrote the manuscript.Ada Stadelmeier undertook data analysis and organized the investigation.Jürgen Gallinat designed the electrophysiological part of the study and gives

significant impact on electrophysiological data analysis.Maria Jockers-Scherübl and Author Johannes Rentzsch designed the complete

study and wrote the study protocol. Author Maria Jockers-Scherübl wrote the seconddraft of the manuscript.

All authors contributed to and have approved of the final manuscript.

Conflict of interestAll authors report no financial interests or potential conflicts of interest that could

inappropriately influence, or be perceived to influence, this paper.

AcknowledgementsWe thank Daniela Knuth, Liane Adomat, Jan Hülsenbeck, Anna Sophia Wagner,

Nicole Mauche, Christina Tramm and David Biallowons who gave generously of theirtime and made this study possible. We particularly thank the patients and probandsparticipating in the study.

References

Arendt, M., Rosenberg, R., Foldager, L., Perto, G., Munk-Jorgensen, P., 2005. Cannabis-induced psychosis and subsequent schizophrenia-spectrum disorders: follow-upstudy of 535 incident cases. Br. J. Psychiatry 187, 510–515.

Bangalore, S.S., Prasad, K.M., Montrose, D.M., Goradia, D.D., Diwadkar, V.A., Keshavan,M.S., 2008. Cannabis use and brain structural alterations in first episodeschizophrenia—a region of interest, voxel based morphometric study. Schizophr.Res. 99 (1–3), 1–6.

Battisti, R.A., Roodenrys, S., Johnstone, S.J., Pesa, N., Hermens, D.F., Solowij, N., 2010.Chronic cannabis users show altered neurophysiological functioning on Stroop taskconflict resolution. Psychopharmacology (Berl) 212 (4), 613–624.

Bersani, G., Orlandi, V., Kotzalidis, G.D., Pancheri, P., 2002. Cannabis and schizophrenia:impact on onset, course, psychopathology and outcomes. Eur. Arch. Psychiatry Clin.Neurosci. 252 (2), 86–92.

Bolla, K.I., Brown, K., Eldreth, D., Tate, K., Cadet, J.L., 2002. Dose-related neurocognitiveeffects of marijuana use. Neurology 59 (9), 1337–1343.

Boydell, J., Dean, K., Dutta, R., Giouroukou, E., Fearon, P., Murray, R., 2007. A comparisonof symptoms and family history in schizophrenia with and without prior cannabisuse: implications for the concept of cannabis psychosis. Schizophr. Res. 93 (1–3),203–210.

Bramon, E., Rabe-Hesketh, S., Sham, P., Murray, R.M., Frangou, S., 2004. Meta-analysis ofthe P300 and P50 waveforms in schizophrenia. Schizophr. Res. 70 (2–3), 315–329.

Caspari, D., 1999. Cannabis and schizophrenia: results of a follow-up study. Eur. Arch.Psychiatry Clin. Neurosci. 249 (1), 45–49.

Cohen, M., Solowij, N., Carr, V., 2008. Cannabis, cannabinoids and schizophrenia:integration of the evidence. Aust. NZ J. Psychiatry 42 (5), 357–368.

Compton, M.T., Furman, A.C., Kaslow, N.J., 2004. Lower negative symptom scores amongcannabis-dependent patients with schizophrenia-spectrum disorders: preliminaryevidence from an African American first-episode sample. Schizophr. Res. 71 (1),61–64.

Coulston, C.M., Perdices, M., Tennant, C.C., 2007. The neuropsychological correlates ofcannabis use in schizophrenia: lifetime abuse/dependence, frequency of use, andrecency of use. Schizophr. Res. 96 (1–3), 169–184.

DeLisi, L.E., 2008. The effect of cannabis on the brain: can it cause brain anomalies thatlead to increased risk for schizophrenia? Curr. Opin. Psychiatry 21 (2), 140–150.

D'Souza, D.C., Perry, E., MacDougall, L., Ammerman, Y., Cooper, T., Wu, Y.T., Braley, G.,Gueorguieva, R., Krystal, J.H., 2004. The psychotomimetic effects of intravenousdelta-9-tetrahydrocannabinol in healthy individuals: implications for psychosis.Neuropsychopharmacology 29 (8), 1558–1572.

227J. Rentzsch et al. / Schizophrenia Research 130 (2011) 222–227

Dubertret, C., Bidard, I., Ades, J., Gorwood, P., 2006. Lifetime positive symptoms inpatients with schizophrenia and cannabis abuse are partially explained by co-morbid addiction. Schizophr. Res. 86 (1–3), 284–290.

Ehrenreich, H., Rinn, T., Kunert, H.J., Moeller, M.R., Poser, W., Schilling, L., Gigerenzer, G.,Hoehe, M.R., 1999. Specific attentional dysfunction in adults following early start ofcannabis use. Psychopharmacology (Berl) 142 (3), 295–301.

Foti, D.J., Kotov, R., Guey, L.T., Bromet, E.J., 2010. Cannabis use and the course ofschizophrenia: 10-year follow-up after first hospitalization. Am. J. Psychiatry 167 (8),987–993.

Giuffrida, A., Leweke, F.M., Gerth, C.W., Schreiber, D., Koethe, D., Faulhaber, J.,Klosterkotter, J., Piomelli, D., 2004. Cerebrospinal anandamide levels are elevatedin acute schizophrenia and are inversely correlated with psychotic symptoms.Neuropsychopharmacology 29 (11), 2108–2114.

Gorynia, I., Schwaiger, M., 2010. Effects of handedness (left vs right) and cannabis abuseon intermanual coordination and negative symptoms in schizophrenic patients ofthe paranoid type. Laterality 1–22.

Jockers-Scherubl, M.C., Matthies, U., Danker-Hopfe, H., Lang, U.E., Mahlberg, R., Hellweg,R., 2003. Chronic cannabis abuse raises nerve growth factor serum concentrations indrug-naive schizophrenic patients. J. Psychopharmacol. 17 (4), 439–445.

Jockers-Scherubl, M.C., Danker-Hopfe, H., Mahlberg, R., Selig, F., Rentzsch, J., Schurer, F.,Lang, U.E., Hellweg, R., 2004. Brain-derived neurotrophic factor serum concentra-tions are increased in drug-naive schizophrenic patients with chronic cannabisabuse and multiple substance abuse. Neurosci. Lett. 371 (1), 79–83.

Jockers-Scherubl, M.C., Wolf, T., Radzei, N., Schlattmann, P., Rentzsch, J., Gomez-Carrillo,d.C., Kuhl, K.P., 2007. Cannabis induces different cognitive changes in schizophrenicpatients and in healthy controls. Prog. Neuropsychopharmacol. Biol. Psychiatry31 (5), 1054–1063.

Leweke, F.M., Giuffrida, A., Wurster, U., Emrich, H.M., Piomelli, D., 1999. Elevatedendogenous cannabinoids in schizophrenia. NeuroReport 10 (8), 1665–1669.

Linszen, D.H., Dingemans, P.M., Lenior, M.E., 1994. Cannabis abuse and the course ofrecent-onset schizophrenic disorders. Arch. Gen. Psychiatry 51 (4), 273–279.

Majic, T., Rentzsch, J., Gudlowski, Y., Ehrlich, S., Juckel, G., Sander, T., Lang, U.E.,Winterer, G., Gallinat, J., 2011. COMT Val108/158Met genotype modulates humansensory gating. NeuroImage 55 (2), 818–824.

Medina, K.L., Hanson, K.L., Schweinsburg, A.D., Cohen-Zion, M., Nagel, B.J., Tapert, S.F.,2007. Neuropsychological functioning in adolescentmarijuana users: subtle deficitsdetectable after a month of abstinence. J. Int. Neuropsychol. Soc. 13 (5), 807–820.

Moore, T.H., Zammit, S., Lingford-Hughes, A., Barnes, T.R., Jones, P.B., Burke, M., Lewis,G., 2007. Cannabis use and risk of psychotic or affective mental health outcomes: asystematic review. Lancet 370 (9584), 319–328.

Naatanen, R., 1995. The mismatch negativity: a powerful tool for cognitiveneuroscience. Ear Hear. 16 (1), 6–18.

Naatanen, R., Kahkonen, S., 2009. Central auditory dysfunction in schizophrenia asrevealed by themismatch negativity (MMN) and its magnetic equivalent MMNm: areview. Int. J. Neuropsychopharmacol. 12 (1), 125–135.

Onton, J., Westerfield, M., Townsend, J., Makeig, S., 2006. Imaging human EEG dynamicsusing independent component analysis. Neurosci. Biobehav. Rev. 30 (6), 808–822.

Patrick, G., Struve, F.A., 2000. Reduction of auditory P50 gating response in marihuanausers: further supporting data. Clin. Electroencephalogr. 31 (2), 88–93.

Peralta, V., Cuesta, M.J., 1992. Influence of cannabis abuse on schizophrenicpsychopathology. Acta Psychiatr. Scand. 85 (2), 127–130.

Pomarol-Clotet, E., Salvador, R., Sarro, S., Gomar, J., Vila, F., Martinez, A., Guerrero, A.,Ortiz-Gil, J., Sans-Sansa, B., Capdevila, A., Cebamanos, J.M., McKenna, P.J., 2008.Failure to deactivate in the prefrontal cortex in schizophrenia: dysfunction of thedefault mode network? Psychol. Med. 38 (8), 1185–1193.

Potvin, S., Joyal, C.C., Pelletier, J., Stip, E., 2008. Contradictory cognitive capacities amongsubstance-abusing patients with schizophrenia: a meta-analysis. Schizophr. Res.100 (1–3), 242–251.

Rasser, P.E., Schall, U., Todd, J., Michie, P.T., Ward, P.B., Johnston, P., Helmbold, K., Case,V., Soyland, A., Tooney, P.A., Thompson, P.M., 2009. Gray matter deficits, mismatchnegativity, and outcomes in schizophrenia. Schizophr. Bull. 37 (1), 131–140.

Rentzsch, J., Penzhorn, A., Kernbichler, K., Plockl, D., Gomez-Carrillo, D.C., Gallinat, J.,Jockers-Scherubl, M.C., 2007. Differential impact of heavy cannabis use on sensorygating in schizophrenic patients and otherwise healthy controls. Exp. Neurol. 205 (1),241–249.

Romero, J., Garcia, L., Fernandez-Ruiz, J.J., Cebeira, M., Ramos, J.A., 1995. Changes in ratbrain cannabinoid binding sites after acute or chronic exposure to theirendogenous agonist, anandamide, or to delta 9-tetrahydrocannabinol. Pharmacol.Biochem. Behav. 51 (4), 731–737.

Roser, P., Juckel, G., Rentzsch, J., Nadulski, T., Gallinat, J., Stadelmann, A.M., 2008. Effectsof acute oral Delta9-tetrahydrocannabinol and standardized cannabis extract onthe auditory P300 event-related potential in healthy volunteers. Eur. Neuropsy-chopharmacol. 18 (8), 569–577.

Roser, P., Della, B., Norra, C., Uhl, I., Brune, M., Juckel, G., 2010. Auditory mismatchnegativity deficits in long-term heavy cannabis users. Eur. Arch. Psychiatry Clin.Neurosci. 260 (6), 491–498.

Sarne, Y., Mechoulam, R., 2005. Cannabinoids: between neuroprotection andneurotoxicity. Curr. Drug Targets CNS Neurol. Disord. 4 (6), 677–684.

Schnell, T., Koethe, D., Daumann, J., Gouzoulis-Mayfrank, E., 2009. The role of cannabisin cognitive functioning of patients with schizophrenia. Psychopharmacology(Berl) 205 (1), 45–52.

Seillier, A., Advani, T., Cassano, T., Hensler, J.G., Giuffrida, A., 2010. Inhibition of fatty-acid amide hydrolase and CB1 receptor antagonism differentially affect behaviouralresponses in normal and PCP-treated rats. Int. J. Neuropsychopharmacol. 13 (3),373–386.

Selten, J.P., Veen, N.D., Hoek, H.W., Laan, W., Schols, D., van d., T.I., Feller, W., Kahn, R.S.,2007. Early course of schizophrenia in a representative Dutch incidence cohort.Schizophr. Res. 97 (1–3), 79–87.

Sevy, S., Burdick, K.E., Visweswaraiah, H., Abdelmessih, S., Lukin, M., Yechiam, E.,Bechara, A., 2007. Iowa gambling task in schizophrenia: a review and new data inpatients with schizophrenia and co-occurring cannabis use disorders. Schizophr.Res. 92 (1–3), 74–84.

Solowij, N., 1995. Do cognitive impairments recover following cessation of cannabisuse? Life Sci. 56 (23–24), 2119–2126.

Solowij, N., Battisti, R., 2008. The chronic effects of cannabis on memory in humans: areview. Curr. Drug Abuse Rev. 1 (1), 81–98.

Solowij, N., Michie, P.T., 2007. Cannabis and cognitive dysfunction: parallels withendophenotypes of schizophrenia? J Psychiatry Neurosci. 32 (1), 30–52.

Spano, M.S., Fadda, P., Frau, R., Fattore, L., Fratta, W., 2010. Cannabinoid self-administration attenuates PCP-induced schizophrenia-like symptoms in adultrats. Eur. Neuropsychopharmacol. 20 (1), 25–36.

Stadelmann, A.M., Juckel, G., Arning, L., Gallinat, J., Epplen, J.T., Roser, P., 2011.Association between a cannabinoid receptor gene (CNR1) polymorphism andcannabinoid-induced alterations of the auditory event-related P300 potential.Neurosci. Lett. 496 (1), 60–64.

Stelt van der, M., Di, M.V., 2005. Cannabinoid receptors and their role in neuroprotec-tion. Neuromolecular. Med. 7 (1–2), 37–50.

Struve, F.A., Manno, B.R., Kemp, P., Patrick, G., Manno, J.E., 2003. Acutemarihuana (THC)exposure produces a “transient” topographic quantitative EEG profile identical tothe “persistent” profile seen in chronic heavy users. Clin. Electroencephalogr. 34(2), 75–83.

Szeszko, P.R., Robinson, D.G., Sevy, S., Kumra, S., Rupp, C.I., Betensky, J.D., Lencz, T.,Ashtari, M., Kane, J.M., Malhotra, A.K., Gunduz-Bruce, H., Napolitano, B., Bilder, R.M.,2007. Anterior cingulate grey-matter deficits and cannabis use in first-episodeschizophrenia. Br. J. Psychiatry 190, 230–236.

Turner, W.M., Tsuang, M.T., 1990. Impact of substance abuse on the course and outcomeof schizophrenia. Schizophr. Bull. 16 (1), 87–95.

Vadhan, N.P., Serper, M.R., Haney, M., 2009. Effects of delta-THC on working memory:implications for schizophrenia? Prim. Psychiatry 16 (4), 51–99.

Veen, N.D., Selten, J.P., van d., T.I., Feller, W.G., Hoek, H.W., Kahn, R.S., 2004. Cannabisuse and age at onset of schizophrenia. Am. J. Psychiatry 161 (3), 501–506.

Yucel, M., Bora, E., Lubman, D.I., Solowij, N., Brewer, W.J., Cotton, S.M., Conus, P., Takagi,M.J., Fornito, A., Wood, S.J., McGorry, P.D., Pantelis, C., 2010. The impact of cannabisuse on cognitive functioning in patients with schizophrenia: a meta-analysis ofexisting findings and new data in a first-episode sample. Schizophr. Bull. AdvanceAccess published July 25, 2010. doi:10.1093/schbul/sbq079.

Zammit, S., Moore, T.H., Lingford-Hughes, A., Barnes, T.R., Jones, P.B., Burke, M., Lewis,G., 2008. Effects of cannabis use on outcomes of psychotic disorders: systematicreview. Br. J. Psychiatry 193 (5), 357–363.