Magnetic Resonance Spectroscopy in Adolescent Cannabis Users: Metabolites in the Anterior Cingulate Cortex Reflects Individual Differences in Personality Traits and can Affect Rehabilitation Compliance
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.288984
Source of Support: None, Conflict of Interest: None
Keywords: Adolescents, anterior cingulate cortex, cannabis users, glutamate, magnetic resonance spectroscopy, personality traits
Addictive disorders include conditions related to substance abuse, including alcohol. Addiction occurs when individuals continue to take harmful substances despite unfavorable consequences. In DSM-5, substance use disorder (SUD) is considered the most common psychiatric condition and cause of serious cognitive and behavioral impairment ; it is a true disease process with a chronic relapsing course. Addiction is a progressive disorder characterized by pervasive and long-lasting disturbances in a complex variety of neurotransmitter systems, including glutamate, GABA, serotonin, noradrenaline and endogenous opioids, as well as dopamine. It has a complex and multidimensional etiopathogenesis, with variable degrees of contributing factors, ranging from intersubject variability, enviroment, and addictive substance. Environmental and genetic factors combine to produce risk factors that increase the likelihood of drug use. They may also combine to produce protective factors, which decrease the likelihood of drug use. Vulnerability refers to the total sum of an individual's risk and protective factors and defines the overall possibility of drug use. Individuals with high-risk factors and few protective elements are more predisposed to drug abuse. The biological and behavioral heterogeneity demonstrated in SUDs are only partially identified by current diagnostic classifications, such as the DSM 5 or ICD-10. Therefore, efforts are underway to define new classification schemes for SUD, based on genetic/biological, physiological, and behavioral endophenotypes.,
The search for phenotypic markers at risk for SUD has led to the identification of several biological and behavioral traits known to predispose individuals to addiction, such as impulsivity, novelty/sensation seeking and harm avoidance. Impulsivity is a core symptom in patients with attention-deficit/hyperactivity disorder (ADHD), a neurodevelopmental disorder characterized by hyperactivity and cognitive deficits, often associated to drug use.,
Studies support the hypothesis that ADHD is characterized by a neurochemical alteration of the fronto-striatal glutamate-signaling circuit, including the anterior cingulate cortex (ACC) and the dorsolateral prefrontal cortex (DLPFC)., Longitudinal prospective and familiar high-risk studies have evaluated the possibility that premorbid personality traits may predict SUDs. The most consistent findings are that individuals who show higher scores on impulsiveness, exploratoriness, neuroticism, irritableness, disinhibition, and extraversion, have an increased SUD risk. Based on the psychobiological model of personality criteria, Cloniger developed the temperament and character inventory-revisited test (TCI-R), which measures four biological and heritable dimensions of temperament: harm avoidance (HA; cautiousness and pessimism), novelty seeking (NS; impulsiveness and thrill seeking), reward dependence (RD; forming attachments with others and/or sentimentality), and persistence (PE; perseverance and ambitiousness), as well as three dimensions of character acquired through social learning (self-directedness, SD; Co-operativeness, CO, and self-transcendence, ST). Cross-sectional studies have established a relation between the TCI temperament dimensions and various substance-related behaviors.,,,
In particular, one of the most common and consistent findings is that individuals with high SUDs score on the NS scale compared with controls are associated with dependance various substances  and pathological gambling. These personality traits are heritable, stable across time, and largely genetically determined  but might be modulated by normal variances in neurotransmitter systems, especially the central monoamine systems. In order to study the link between cerebral metabolites distribution and personality, we used 1H-MRS, a sophisticated non invasive magnetic resonance imaging (MRI) technique, that has allowed to obtain cerebral metabolic information in a group of adolescent drug users, by measuring the correlation of TCI scores and metabolites ratios in the ACC. This neuroimaging approach has had a major impact on the elucidation of personality traits associated with addiction.,,
Due to its important role in behavior, its high glutamate receptor density, and the well-known effects of glutammatergic neurotransmission on cerebral blood flow, we have chosen the ACC as region of interest to investigate the role of Glx and the personality traits. The Glx is involved in important cognitive functions as learning, memory, and the long-term potentiation effect of neuroplasticity, a phenomenon that takes place in the glutammatergic synapses of hippocampus, neocortex, and other cerebral regions. Metabolic abnormalities or imbalances of Glx may cause neurocognitive deficits and impulsive drug seeking behavior or drug abuse.,
Just like the orbitofrontal cortex, the ACC also has an important role in emotional and social behavior.,
It is well known that the glutamate neurotransmission influence impulsive choices and actions  and that there is a relationship between brain morphometry and personality traits. Glutamate measurements with 1H-MRS in the ACC negatively correlate with sensation seeking temperament. In the “sensation seeking behavior,” the intensity of the lack of gratification may lead to the development of assumptive behaviors.
We investigated the relationship between personality and brain metabolism in addictive behaviors to define which temperamental traits (innate) or dimensions of character (role of environment and experiences of life) could make adolescents more vulnerable to drugs use and which metabolic pattern may be associated to drug seeking behavior. These results could represent a crucial step to understanding the psychobiological mechanisms of addiction in adolescence. An early knowledge of the cerebral metabolic pattern that predisposes to drugs and the associated personality traits can help in identify adolescents at risk for addiction and/or prevent the first assumption of substances with the development of tailored behavioral rehabilitative programs.
All procedures were approved by the Local Institutional Review Board. Twenty-eight male cannabis users with different history of substance intake were enrolled in this study [mean age = 18.54 ± 2.80, [Table 1]. Initially, the sample was constituted by 32 subjects (28 male and 4 females). We decided to discard the small sample of females and to consider a group of male-only individuals, even though we know that are polysubstance users, in the last few years relying on hashish consumption (N = 21), a derived resin of cannabis with a large concentration (also doubled) of tetrahydrocannabinol (THC, the main active ingredient of cannabis) and associated with alcohol and nicotine use (N = 28) [Table 2]. The subjects were recruited from the Addiction Department of Verona, ULSS 20 (Italy) and underwent the MRI procedure at the Department of Neuroradiology, University Hospital of Verona (Italy). Subject's selection was based on the following criteria: (i) age between 15 and 22 years (ii) daily or weekly drug use; a maximum of 30 months of abuse; (iii) absence of psychosis or risk to develop psychotic or other neurological disorders; (iv) no contraindication to perform MRI. The selection of potential subjects was carried out by the researchers based on the presence of addictive behaviors as described in the DMS-IV criteria. The toxicological data were collected for each drug user [Table 3]. Individuals not fulfilling the aforementioned inclusion criteria were excluded. Eleven healthy “no drugs-users” (three males and eight females) were recruited as a control sample. Control participants' ages ranged from 13 to 24 years, with a mean age and standard deviation of 20.16 ± 1.95 years [Table 1].
The TCI-R of Cloninger was used to assess the personality characteristics of subjects. In addition to the TCI-R, the Symptom Checklist-90 (SCL-90) was used to further assess psychological disorders and symptoms (depression, anxiety, hostility, etc.). A personal interview with anamnestic information (age, gender, education, occupation, and handedness) and questions about the pattern and severity of substance use (type, age of first use, lifetime use, and abstinence) further delineated each participant's story of consumption [Table 1] and [Table 2]. The structured interview administered to the participants was the Structured Clinical Interview for DSM IV Axis II Disorders (SCID-II), the Zung-A Scale (Anxiety), Zung-D scale (Depression), Boredom PS (Boredom), the Routinary Reward Behavioral System (SCAR), and the Visual Analogue Scale (VAS) for the evaluation of abstinence and craving [Table 4] and [Table 5]. A psychological interview was held with the parents of the participants on the behavioral/relational aspects of the teen and on the family educational models (money management, night return, etc.). All participants underwent urine screening to verify the state of drug intake.
Magnetic resonance spectroscopy acquisition techniques
The volume of interest was selected within the frontal lobes white matter, including 8 voxels; the final matrix available for calculation of the metabolite spectra was a grid of 2 × 4 voxels (volume of interest = 27 cm3) for each hemisphere. The metabolites peaks were measured as parts of million (ppm) and at the end of each acquisition the metabolite ratios of the Glx/Cr, Cho/NAA, and Ins/Cr were automatically calculated using the scanner software (Spectroscopy Card, Siemens). With this method, it was possible to quantify the neurochemical damage in a single volume of interest within the frontal lobes and identify the reference metbolites. NAA is a neuronal marker and it is positioned at 2.0 ppm; Cho is a cell membrane marker and it is located at 3.2 ppm, Glx is a neurotransmitter and it positioned at 2.2 ppm, Cr is an index of metabolism energy and it is found at 3.0 ppm, whereas Ins is a glial cell marker and it is positioned at 3.5 ppm. The metabolite concentrations were examined after scaling to Cr (no statistical differences between groups in Cr values was observed when using water as the reference). In the Cho/NAA ratio, we used NAA as reference to investigate the neuronal integrity, as NAA measurement may be a valid and indirect marker of the cerebral energetic state., Metabolite ratios of all selected voxels were then averaged and presented as mean ratios in order to quantify the neurochemical values in a single volume of interest of the ACC. Normative reference data were obtained from our normative reference values of healthy subjects [Table 1].
Analysis of spectroscopic data
The volume of interest was selected within the white matter of the frontal lobes, including 8 voxels, and the final matrix available for calculation of metabolite intensity was a grid of 2 × 4 voxels (volume of interest = 27 cm3) for each hemisphere. We identified the metabolites peaks of interest express as parts of million (ppm), and at the end of each acquisition, the metabolite ratios of the Glx/Cr, Cho/NAA, and Ins/Cr were automatically calculated using the scanner software (Spectroscopy Card, Siemens). Using this method, it was possible to quantify the neurochemical damage in a single discrete volume of interest of the frontal lobes. The NAA is a neuronal marker and it is positioned at 2.0 ppm, Cho is a cell membrane marker, and it is positioned at 3.2 ppm, Glx is a neurotransmitter and it is positioned at 2.2 ppm, Cr is an index of metabolism energy and it is positioned at 3.0 ppm, whereas Ins is a glial cell marker and it is positioned at 3.5 ppm. The metabolite concentrations were examined after scaling to Cr (no statistical differences between groups in Cr values was observed when using water as the reference). In the Cho/NAA ratio, we used NAA as reference to investigate the neuronal integrity since NAA measurement may be a valid indirect biomarker of the brain energy state., Metabolite ratios of all selected voxels were then averaged and presented as mean ratios in order to quantify the neurochemical values in a single volume of interest of the ACC. Normative reference data were obtained from our normative reference values of healthy subjects [Table 1].
All data analyses and calculations of sample size were performed using Statistical Package for the Social Sciences version 13.0 (SPSS, Chicago, IL, USA). Scores on personality scales and psychopathological symptoms were analyzed for each subject. The spectroscopy data were compared by considering the ratio values versus the reference standard. Descriptive statistics for quantitative continuous variables were presented as mean (M) ± standard deviation (SD). In order to analyze the level of significance of differences between the mean values of the results of the TCI-R between the experimental group and the control group, the Mann–Whitney U-test was performed. The threshold of statistical significance was set at 5% (P < 0.05).
The Pearson's correlation analysis was used to test correlation effects between the TCI-R values and metabolites distribution in adolescent drug users. The results of the TCI-R scales as well as Glx/Cr, Cho/NAA, and Ins/Cr levels calculated in the sampled area of the ACC were also considered. Spearman's rho correlations between the indices of personality and concentrations of Glx/Cr, Cho/NAA, and Ins/Cr (expressed in ppm) were calculated. The significance threshold was set at the level P < 0.05.
Initially, the statistical analysis was performed on 32 subjects (28 males and 4 females), using age and sex as covariates to compare metabolite concentrations in the drugs users group. Our sample included polydrug users [see [Table 2]. In the statistical analysis, we have considered the main substance used in the last year (cannabis) in association with alcohol and nicotine, considering for each substance the age of the first use, the frequency of use and the abstinence. The statistical analysis (P < 0.05) was not significative for alcohol and nicotine. Because of the small number of females (4/32), we excluded them from the analysis; the decision was taken in order to create a more homogeneous sample group. The correlations between the control subjects and TCI traits were not found significant [Table 1].
In our SUD study, the spectroscopy analysis showed a reduction of Glx and an increased Cho (index of membrane turnover) with a cerebral metabolic pattern similar to the one found in depressive disease. The personological analysis showed an increased NS value, a temperamental factor that explains the tendency to use psychotropic substances, and a reduction of CO, which is a reduction of positive interactions (heat/affection), interactive learning, and social norms. The correlation analysis between the TCI-R values and metabolites distribution in SUD, showed that the CO1 scores correlated negatively with Glx concentrations in the ACC, suggesting that low levels of Glx in the ACC are closely related to values of CO in adolescent cannabis users. Several studies have demostrated the roles of ACC and prefrontal cortex in cognitive control (”top-down control”). In particular, the ACC seems to be recruited when cognitive demands increase for the maintenance and the implementation of cognitive functions in case of conflictual situation (i.e., decision process). At the same time, we found a negative correlation between cho and NS. Adolescent drugs users with higher level of cho present less novelty seeker traits compare to the low Cho level subjects in the same population. Control subjects are metabolically and behaviorally similar to normative data (except for a reduction of the scores to the ST and an increased CO in the TCI-R). In cannabis users, even though the scores of CO were significantly reduced compared with control subjects and normative data, we found a negative correlation between the lowest values of Glx/Cr and the value of CO. Therefore, adolescent drug users with a marked cerebral reduction of Glx tend to be more collaborative, although always below the normal range. However, this may also depend on a specific feature of our subjects, as they participated in a rehabilitation program and therefore were more collaborative.
Our correlation analysis showed a metabolic pattern compatible with a depressive behavioral state (reduction of Glx and increase of Cho) associated to increased CO and low NS. These results are in line with the neurobiological model that explains the tendency of our sample to use substances to reduce activity and independent choices with needs of external consensus, to accept and rely on the other judgment. In support of this hypothesis, the SCL-90 analysis showed significative score for Interpersonal Sensitivity, Depression, Phobic anxiety, Psychoticism, Schizophrenia characterized by withdrawal, isolation, and lifestyle schizoid.
Adolescence is the first phase of life in which the individual is intentionally in contact with psychoactive substances. It is characterized to a greater tendency to search new sensations (a peculiarity in the processes of gratification typical of the age), to an altered sensitivity to drugs and to a lack of effect “full,” as a result of the first approach to substances use that could push teenagers to try again “to see what happens.” Adolescents may also be more vulnerable to the effects of drugs to generate dependence due to cerebral development imbalances between the prefrontal control areas and the subcortical impulsivity. The combined action of these factors could characterize adolescence as an age psychobiologically at risk for the development of addiction. It was recently showed the association between cannabis use and genetic risk of develop psychiatric traits as schizophrenia.
Metabolic and personality patterns seem to be associated with the risk of substance intake in adolescents. The combination of 1H-MRS and Neuropsychology data represents a new neurobiological, noninvasive approach to asses the risk behaviors of addiction in adolescence. Our study support the hypothesis that brain metabolites are prognostic markers of brain function, emotions, and experience of gratification in adolescence.
In terms of prevention, it is necessary to consider various predisposing factors such as environment, availability of the substances, role of cultural and social conditioning, peer pressure, positive attitude of the group toward the use of substances, and time of exposure to these factors. Additionally, there is the tendency in teenagers to use therapeutic drugs as a self-medication in order to solve psychological problems and emotional states.
Drug-prevention programs in adolescents should be based on early identification of teens at risk of conduct disorder and substance abuse. We think that our results can contribute to better understand the biological bases, which characterize inter-individual differences in drug-user adolescence. There is a general difficulty in recruiting data in teen cannabis users and although our study presents limitations (a small sample size, polysubstance consumption, and the partecipation in drug detoxification programs), it provides new indications on cerebral markers and the neurobiological processes of adolescents cannabis use. As it is not a long-term study, further investigations in terms of rehabilitative outcome would be needed to identify adolescents with a more rehabilitation compliance and a subsequent greater success in the therapeutic drug detox.
Financial support and sponsorship
There are no financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]