Gustavo L. Ottoni

Temperament and character traits associated with the use of alcohol, cannabis, cocaine, benzodiazepines, and hallucinogens: evidence from a large Brazilian web survey

OBJECTIVES To evaluate how personality traits are associated with occasional use, abuse, and dependence of alcohol, cannabis, cocaine, benzodiazepines, and hallucinogens in a large availability sample of adults via online questionnaires. METHODS The sample consisted of 8,646 individuals (24.7% men and 75.3% women) who completed an anonymous web survey. Involvement with drugs and temperament/character traits were assessed through the Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) and the Temperament and Character Inventory - Revised (TCI-R), respectively. Interactions among variables were analyzed using MANOVA with Bonferroni adjustment. RESULTS Novelty seeking was the trait most associated with increased involvement with alcohol, cannabis, and cocaine. There was a significant association between harm avoidance and benzodiazepine use. Persistence was lower in cannabis-, benzodiazepine-, and cocaine-dependent subjects, as well as in hallucinogen abusers. Self-directedness was reduced in dependents of all drug classes. No strong relationships were found between other temperament or character dimensions and the severity of drug use. CONCLUSIONS Novelty seeking was associated with increased involvement with all drugs studied in this sample, although to a lesser extent with benzodiazepines and hallucinogens. The temperament and character profile for benzodiazepine use was different from that of other drugs due to the relationship with higher harm avoidance and self-transcendence and lower self-directedness.

The relationship between affective temperaments, defensive styles and depressive symptoms in a large sample.

BACKGROUND Affective temperaments may represent heritable subclinical manifestations of mood disorders. The concept of ego defense mechanisms also has provided a model for the comprehension of mood psychopathology. The relationships between affective temperaments, defense styles and depressive symptoms remain unknown. METHODS We obtained data from a subsample of the Brazilian Internet Study on Temperament and Psychopathology (BRAINSTEP). Socio-demographic information was collected and participants completed the Affective and Emotional Temperament Composite Scale (AFECTS), the defense style questionnaire (DSQ-40) and the Symptom Checklist-90-Revised (SCL-90-R). RESULTS Among 9937 participants (4472 male; 45%), individuals with hyperthymic or euthymic temperaments were more likely to present a mature defense style, whereas an immature defensive style was predominantly observed in individuals with cyclothymic, volatile, depressive, dysphoric, euphoric and disinhibited temperaments. Higher immature and lower mature defense style scores were independently associated with depressive symptoms. Participants with either euthymic or hyperthymic temperaments were less likely to endorse depressive symptoms. Euthymic and hyperthymic temperaments moderated the correlations of mature/immature defenses with depressive symptoms. LIMITATIONS The data was collected from a convenience web-based sample. The study was cross-sectional. CONCLUSIONS Affective temperaments are associated with distinct defense styles. These two personality theories provide distinct but interacting views for comprehension of depressive psychopathology.

ZDHHC8 gene may play a role in cortical volumes of patients with schizophrenia

ZDHHC8 rs175174 polymorphism is located in 22q11.2 region and its role in brain volume has not been fully addressed. A total of 282 schizophrenia patients and 379 controls were genotyped. A sample of 138 patients underwent brain MRI scan. No association was found between schizophrenia and genotypes. Nevertheless, GG-genotype carriers presented gray matter volume (GMV) reduction in frontal lobe compared to A-allele carriers, and cerebellar hemispheres GMV reductions were found in G-allele carriers compared to AA-genotype. Moreover, A-allele carriers presented posterior brain GMV reductions when compared to GG-genotype. These data suggest that ZDHHC8 may play a role in cortical volumes.

Childhood trauma is associated with maladaptive personality traits

The association between childhood trauma and personality traits has been poorly characterized and reported. Our aim was to evaluate whether distinct types of childhood abuse and neglect are associated with various personality dimensions using data from a large web-based survey. A total of 12,225 volunteers responded anonymously to the Internet versions of the Temperament and Character Inventory-Revised (TCI-R) and the Childhood Trauma Questionnaire (CTQ) via our research website, but only 8,114 subjects (75.7% women, mean age 34.8±11.3yrs) who met the criteria for validity were included in the analysis. Childhood trauma was positively associated with harm avoidance and was negatively associated with self-directedness and, to a lesser extent, with cooperativeness. The associations were robust with emotional abuse and neglect but were non-significant or mild with physical trauma. Emotional neglect was associated with reduced reward dependence and persistence. All types of abuse, but not neglect, were associated with increased novelty seeking scores. Reporting of childhood trauma, especially of an emotional nature, was associated with maladaptive personality traits. Further investigation of the effects of different types of childhood trauma on psychological and neurobiological parameters is warranted.

Thinking About Dying and Trying and Intending to Die: Results on Suicidal Behavior From a Large Web-Based Sample

OBJECTIVE Suicide is an important worldwide public health problem. The aim of this study was to characterize risk factors of suicidal behavior using a large Web-based sample. METHOD The data were collected by the Brazilian Internet Study on Temperament and Psychopathology (BRAINSTEP) from November 2010 to July 2011. Suicidal behavior was assessed by an instrument based on the Suicidal Behaviors Questionnaire. The final sample consisted of 48,569 volunteers (25.9% men) with a mean ± SD age of 30.7 ± 10.1 years. RESULTS More than 60% of the sample reported having had at least a passing thought of killing themselves, and 6.8% of subjects had previously attempted suicide (64% unplanned). The demographic features with the highest risk of attempting suicide were female gender (OR = 1.82, 95% CI = 1.65 to 2.00); elementary school as highest education level completed (OR = 2.84, 95% CI = 2.48 to 3.25); being unable to work (OR = 5.32, 95% CI = 4.15 to 6.81); having no religion (OR = 2.08, 95% CI = 1.90 to 2.29); and, only for female participants, being married (OR = 1.19, 95% CI = 1.08 to 1.32) or divorced (OR = 1.66, 95% CI = 1.41 to 1.96). A family history of a suicide attempt and of a completed suicide showed the same increment in the risk of suicidal behavior. The higher the number of suicide attempts, the higher was the real intention to die (P < .05). Those who really wanted to die reported more emptiness/loneliness (OR = 1.58, 95% CI = 1.35 to 1.85), disconnection (OR = 1.54, 95% CI = 1.30 to 1.81), and hopelessness (OR = 1.74, 95% CI = 1.49 to 2.03), but their methods were not different from the methods of those who did not mean to die. CONCLUSIONS This large Web survey confirmed results from previous studies on suicidal behavior and pointed out the relevance of the number of previous suicide attempts and of a positive family history, even for a noncompleted suicide, as important risk factors.

Polymorphisms in schizophrenia candidate gene UFD1L may contribute to cognitive deficits

We aimed to investigate UFD1L polymorphisms in schizophrenia and in relation to cognition. A total of 299 cases and 363 controls were genotyped, and 130 patients completed nine neuropsychological tests. We found that rs5992403 AA-genotype carriers showed lower scores on the set-shifting task. Therefore, UFD1L may participate in the core cognitive deficits observed in schizophrenia.

Distinct sensitivity to caffeine-induced insomnia related to age:

Caffeine acts by antagonizing the effect of the endogenous homeostatic sleep factor adenosine. In the current study we aimed to evaluate the pattern of caffeine-induced insomnia and its relation to age and sex in a general population sample derived from a web survey. The sample included 75,534 participants (28.1% men) from 18 to 75 years who answered self-report questionnaires by accessing a website in Brazilian Portuguese (BRAINSTEP project). In our sample, 3620 (17.0%) men and 9920 (18.3%) women reported insomnia due to caffeine intake. Caffeine-induced insomnia increased with aging in both men and women. This difference remained after adjusting for sociodemographic, psychiatric and sleep related variables as well as caffeine intake. Women showed higher proportion of caffeine-induced insomnia than men, but this difference did not remain after controlling for covariates. Also, individuals with caffeine-induced insomnia reported poorer sleep quality, higher latency to fall asleep and a higher proportion of psychiatric diagnoses and daily use of hypnotic drugs. In conclusion, our results show an age-associated increase in caffeine-induced insomnia and poorer mental health indicators among people with caffeine-induced insomnia complaints.

Candidate genes for schizophrenia in a mixed Brazilian population using pooled DNA

To the Editors: Some studies have reported rare and de novo copy number variations (CNVs) associated with schizophrenia (International Schizophrenia Consortium, 2008). However, common CNVs may also contribute to its genetic liability (Glessner et al., 2010). DNA pooling may be a cost-effective way to discover common CNVs (Breen et al., 2000); this may reduce detection of interindividual CNV variability unrelated to the disease (de Cid et al., 2009). In this study, we investigated CNVs in people with schizophrenia using microarray analysis of DNA pools. We recruited 138 patients and 120 healthy controls from Sao Paulo (SP; in the south-east of Brazil). Additionally, we selected 74 cases of schizophrenia from Rio Grande do Sul (RS; in the south of Brazil) for replication in a distinct population. In both groups we only included stable patients being followed in outpatient clinics, who had been diagnosed by Structured Clinical Interview for DSM-IV. The Research Ethics Committee of UNIFESP approved the research protocol and all participants gave informed consent. The characteristics of the study population are described in Supplementary Table 1. The population structure was assessed as previously described (Santos et al., 2010). We constructed six pools by mixing equal amounts of DNA isolated from blood: a) SP pool: a total of 138 samples from patients with schizophrenia from SP; these samples were then randomly distributed among three other pools comprising 46 samples each (SP-A, SP-B and SP-C). Clinical characteristics such as severity were not taken into consideration; b) CON pool: a total of 120 samples from controls; and c) RS pool: a total of 74 samples from patients with schizophrenia from RS. Each pool was hybridised to one array, except for the SP and CON pools, which were hybridised to two each, making a total of eight arrays (Supplementary Fig. 1). Samples were analysed on a single-nucleotide polymorphism (SNP) array 6.0 (Affymetrix) and analysed with the Partek Genomics Suite (PGS; Partek Inc.) and PennCNV (http://www. openbioinformatics.org/penncnv/). CNVs were scored using the Hidden Markov Model (for PennCNV) or segmentation (for PGS) and calculated according to a set of 270 normal samples from the HapMap database. For PGS, only non-polymorphic probes were considered. We considered relevant the regions that were identified as CNVs by both methods, between both replicates of the SP pool and in at least one subdivided DNA pool (SP-A, SP-B or SP-C). In addition, we validated the CNVs in the RS pool. Finally, we compared the identified CNVs with those in the CON pool (Supplementary Fig. 1). After calculating all the overlaps, regions on 2p11.2, 10q11.21, and 14q32.33 were found to be duplicated (Table 1) in patients