- Email: [email protected]
The hypothalamic-pituitary-thyroid axis and personality in a sample of healthy subjects
Psychoneuroendocrinology 87 (2018) 181–187
Contents lists available at ScienceDirect
Psychoneuroendocrinology journal homepage: www.elsevier.com/locate/psyneuen
The hypothalamic-pituitary-thyroid axis and personality in a sample of healthy subjects
T
⁎
Aleksei Piskunova, , Adela Fustéa, Nadezhda Teryaevab, Aleksei Moshkinb, José Ruiza a b
Department of Personality, Assessment and Psychological Treatment, Faculty of Psychology, University of Barcelona, Spain N.N. Burdenko Neurosurgical Institute, Moscow, Russia
A R T I C L E I N F O
A B S T R A C T
Keywords: TSH Thyroxin Triiodothyronine Personality Health Neuroticism Depression
Thyroid hormones influence various brain pathologies, including psychiatric disorders. However, the relationship between these hormones and the psychological state in the normal, non-clinical population is poorly understood. We aimed to test whether serum levels of thyroid hormones are associated with personality in the healthy population. Thyrotropin (TSH), free T3 (FT4), total and free T4 (TT4 and FT4) concentrations were measured in the blood of 104 healthy participants (44% men) aged 18–59 (M = 35 ± 9). Personality traits were assessed using the revised short versions of the Temperament and Character Inventory (TCI-140) and the NEO Five-Factor Inventory (NEO-FFI). The data were analysed by correlational, regression, extreme groups and graphical techniques, which showed significant correlations between inter-individual variations in serum thyroid hormone levels and specific aspects of personality. In particular, high serum TSH was strongly associated with higher Persistence and Self-Directedness, and negatively correlated with Harm Avoidance on the TCI-140 and Neuroticism on the NEO-FFI, thus representing a more adaptive personality profile. Furthermore, increased FT4 was associated with lower Reward-Dependence, and increased TT4 was associated with lower Cooperativeness, representing a deficit in social attachment. Our data indicate that the relationship between thyroid hormones and personality in the healthy population might be rather more complex than the results obtained in clinical samples suggest.
1. Introduction The study of the relationship between hormones and personality dates back to Antiquity, when Hippocrates and Galen proposed the first notions of humoral typologies based on the deficit or excess of body fluids. Today, according to Netter (2004), the hormones most frequently studied in relation to personality are the ones associated with the hypothalamo-pituitary-adrenal (HPA) axis such as cortisol, the ones in the gonadal-pituitary-adrenal (HPG) axis such as testosterone, and the ones in the adrenomedullary axis such as the catecholamines noradrenaline, adrenaline and dopamine. The most significant results of the study of HPA axis hormones and personality show a positive relationship between cortisol and Neuroticism (anxiety and depression), and a negative relationship with Novelty/Sensation seeking and Aggressiveness. For its part, testosterone has also been associated with Dominance, Aggressiveness and Novelty/sensation seeking. Likewise, higher levels of noradrenaline have been associated with Anger and Aggressiveness, while adrenaline levels are more related to Fear and serotonin levels to Anxiety (Gray, 1993). In contrast, the dopaminergic
⁎
system has been fundamentally associated with Novelty/Sensation seeking, and the serotoninergic system with Harm Avoidance (Cloninger, 2000). In addition, the adrenomedullary axis (the catecholaminergic system) in interaction with the HPG axis has been considered as the regulatory mechanism of emotions, social behavior and clinical conditions (Zuckerman, 1997). Thyroid hormones appear to be less studied in the relation to personality. However, there is a lot of background data. A variety of mechanisms of thyroid hormone uptake into brain tissues and hormone activation and their influences on neurotransmitter generation have been described (for a review, see Schroeder and Privalsky, 2014). Thyroid hormones are also involved in neurogenesis and neuron-astrocyte communication (Morte and Bernal, 2014; Préau et al., 2014; Sánchez-Huerta et al., 2016). Thyroid-stimulating hormone (TSH) serum level predicts response to selective serotonin reuptake inhibitors in depressive patients, suggesting a relationship between thyrotrophic hormone and serotoninergic system susceptibility (Gitlin et al., 2004). TSH secretion is exerted by classical negative feedback loop, thus elevated TSH is believed to indicate low peripheral thyroid function. An
Corresponding author. E-mail address: [email protected] (A. Piskunov).
http://dx.doi.org/10.1016/j.psyneuen.2017.10.023 Received 6 June 2017; Received in revised form 6 September 2017; Accepted 29 October 2017 0306-4530/ © 2017 Elsevier Ltd. All rights reserved.
Psychoneuroendocrinology 87 (2018) 181–187
A. Piskunov et al.
«did not sleep at all») and how they felt at the present moment. Seven participants were excluded from the data analysis: one who reported feeling unwell, four who had drunk alcohol the day before, and two who had slept very poorly or not at all the previous night.
association between hypothyroidism and depression has been long accepted by scientists and clinical specialists. Indeed, many studies suggest that the hypothalamic–pituitary–thyroid axis (HPTA) is involved in pathogenesis of psychiatric disorders, especially mood disorders such as depression (Hein and Jackson, 1990) and bipolar disorder (Sierra et al., 2014; Bauer et al., 2014). Hypothyroidism and depression share some symptoms (Jackson, 1998) and many patients with depression have abnormal thyroid hormone levels (Kirkegaard and Faber, 1998). However, larger studies on the relationship of thyroid hormones and depression conducted over the last 15 years in normal populations, including studies with prospective and community approaches, have unexpectedly shown serum TSH to be inversely associated with anxiety and depression (Forman-Hoffman and Philibert, 2006; Williams et al., 2009; Medici et al., 2014). Furthermore, a negative correlation has been found between serum TSH and Neuroticism (Frey et al., 2007). Regarding peripheral thyroid hormones, several studies (Roca et al., 1990; Sim et al., 2002; Steiblienė et al., 2012; Akiibinu et al., 2012) have reported elevated T4 in patients with cluster A personality disorders (e.g., Schizoid and Schizotypal). Stalenheim et al. (1998; cited by Sinai et al., 2009) found that elevated T3 levels were associated with psychopathy and antisocial disorder, whereas serum levels of FT4 were negatively related to these disorders. Sinai et al. (2009) also mention other studies (Arqué et al., 1987; Balada et al., 1992) suggesting a negative correlation between TSH and T4 with sensation seeking, and a positive correlation between T4 and scores on depression-anxiety scales in healthy females with no psychiatric history. As can be seen from the above, despite the long history of investigation, the relationship between thyroid hormones and psychological and psychiatric functions is not clearly defined. Most of the studies in the area dealt with thyroid pathology or psychiatric illness. However, we believe that better and more fundamental understanding of the problem might be obtained by studying the role of HPTA in health, i.e. by investigating the role of thyroid hormones in normal regulation of psychological functions, such as personality, in subjects with no thyroid disease. In the present study we aimed to explore in detail whether personality dimensions in mentally healthy individuals might be related to physiological concentrations of serum thyroid hormones.
2.2. Psychological measures Two models for assessing personality were used: the NEO Five Factor Inventory (NEO-FFI; Costa and McCrae, 1992) and the short version (TCI-140) of the revised Cloninger’s Temperament and Character Inventory (Cloninger, 1999). The NEO-FFI is based on a lexical analysis of trait adjectives and the TCI-140 is based on a psychobiological model that accounts for individual differences in personality traits by integrating neurochemical systems, learning, and social influences. Personality measures from these models correlate to problem behaviors and psychiatric diagnoses. Personality dimensions assessed with both the NEO-FFI and the TCI-R demonstrate broad-sense heritability in excess of 30% (Johnson et al., 2008). All participants completed the short version of the revised 5-point Likert scale version of TCI-140, which allows valid personality assessment over four dimensions of temperament (NS: Novelty Seeking, HA: Harm Avoidance, RD: Reward Dependence, PS: Persistence) and three dimensions of character (SD: Self-Directedness, С: Cooperativeness, ST: Self-Transcendence). The NEO-FFI was performed in 71 subjects (70%). The participants completed the short 60-item version of the NEO-FFI, which allows reliable and valid assessment of personality along the dimensions Neuroticism (N), Extraversion (E), Openness to experiences (O), Agreeableness (A), and Conscientiousness (C) (Kudryashev, 1992; Baturina and Aidman, 2010). If not indicated as a raw score, each dimensional score is presented as a T-score with a mean of 50 and a standard deviation of 10. Table 1 shows descriptive statistics and reliability data for TCI-140, and Table 2 shows descriptive statistics for NEO-FFI. Since we used a commercial version of the questionnaire that does not provide scoring keys or raw items data, we were unable to calculate Cronbach’s alphas and so reliability data on NEO-FFI are not presented here. 2.3. Thyroid hormone measurements
2. Methods All analyses were performed in accordance with accredited routines at the Laboratory for Clinical Biochemistry of the Burdenko Neurosurgical Institute. Blood was collected in the morning between 8:00 a.m. and 10:30 a.m. after an overnight fast using separating gel tubes. Serum was separated by centrifugation at 3000g for 15 min. The levels of thyrotropin (TSH), total and free thyroxin (TT4 and FT4) and free triiodothyronine (FT3) were determined using automated chemiluminescence immunoassay (“Immulite2000”, Siemens, USA).
2.1. Participants A total of 104 healthy volunteers (46 male, 58 female, aged 18–55 years, M35 ± 12) from the Moscow region in Russia were recruited for this study via a social network advertisement containing a brief description of the study, which was posted in the online communities of students at the biology and psychology faculties and on the personal webpages of researchers. The advertisement invited students and researchers to participate in the study and asked them to share the invitation with their peers. As an incentive to take part, prospective participants were told that they would be given information on their blood hormone levels and psychological measures. Before enrollment, participants completed an online medical history in order to assess the presence of any clinical syndrome (DSM, Axis I; American Psychiatric Association, 2013), medical condition or any thyroid or psychotropic drug intake, all of which were considered as exclusion criteria. After a complete description of the study to the subjects, written informed consent was obtained. The study was approved by the local ethics committee of the N.N. Burdenko Neurosurgical Institute. Participants were asked not to come for blood sampling if they had slept poorly the night before, and not to drink alcohol in the preceding days before in order to mitigate the influence of these factors on thyroid hormone plasma levels (Ylikahri et al., 1980; Kessler et al., 2010). On the day of blood sampling, participants were asked to state when they had last drunk alcohol, how they had slept the previous night (recorded on a 1–5 Likert scale with «5» corresponding to «slept perfectly» and «1»
2.4. Statistical analysis We used correlations, extreme groups, graphical and regression techniques to investigate the relationships between serum thyroid hormones and personality traits. Each type of analysis complemented Table 1 TCI-140 descriptive statistics and alpha internal consistency.
Novelty Seeking Harm Avoidance Reward Dependence Persistence Self-Directedness Cooperativeness Self-Transcendence
Mean
SD
S
K
alpha
62.3 55.2 62.3 68.1 65.7 72.3 46.7
9.08 12.7 10.9 10.0 10.4 11.2 8.2
0.19 0.13 −0.24 0.08 −0.55 −0.70 −0.22
0.16 −0.47 0.25 0.72 0.44 0.34 0.04
0.71 0.87 0.81 0.84 0.80 0.84 0.71
K − kurtosis, S − skewness, n = 104.
182
Psychoneuroendocrinology 87 (2018) 181–187
A. Piskunov et al.
Table 2 NEO-FFI descriptive statistics.
Table 4 Partial correlations between thyroid indices and NEO-FFI scores controlled for age.
Neuroticism Extraversion Openness Agreeableness Conscientiousness
Mean
SD
26.1 28.2 32.2 29.1 29.8
7.61 6.77 5.81 7.22 9.91
Neuroticism Extraversion Openness Agreeableness Conscientiousness
TSH
FT3
FT4
TT4
−0,30** 0,22 0,01 0,19 0,13
0,11 −0,01 −0,07 −0,19 0,00
0,00 −0,21 −0,16 −0,10 0,10
0,06 −0,11 −0,10 −0,14 −0,15
n = 71.
** − p < 0.01. n = 71.
the other. The distributions of variables were initially plotted to identify pathological cases. Cases with TSH values beyond normal range (0.4–4 μIU/mL) were excluded (four cases in total) in order to form a clinically euthyroid sample. Gender effects were tested using the T-test for independent samples, and age effects were determined with Pearson's correlation test. Thyroid indices were correlated with NEOFFI and TCI-140 scores using a partial correlation test. The results are presented as M ± 1 SD. The p-values are two-tailed. The variance inflation factor (VIF) and tolerance statistic indicated no problem with multicollinearity. In all analyses, p values less than 0.05 were considered significant. All calculations were carried out using the IBM SPSS (version 22). Nonlinear relationships were analysed through the LOcally WEighted Scatter-plot Smoother (LOESS), non-parametric, local area, polynomial regression procedure (Fox, 2000) to produce data points for the TCI-140 and NEO-FFI scores. This method involves a series of local regression analyses that allows the shape of a curve to vary across the variable continua. The resulting curves are the best unbiased depictions of the patterns in the data. The curves are much more accurate than the lines (linear or quadratic) that are imposed on the data in familiar parametric analyses (Fox, 2000).
(r = −0.24, p < 0.05) and positively correlated to Persistence (r = 0.25, p < 0.05) and Self-Directedness (r = 0.27, p < 0.01). FT3 showed no significant correlations with TCI-140 scores. FT4 was negatively correlated to Reward Dependence (r = −0.19, p < 0.05). TT4 was negatively correlated to Cooperativeness (r = −0.22, p < 0.05). For the NEO-FFI dimensions, only the correlation between TSH and Neuroticism was significant (r = −0.30, p < 0.01). No further significant correlations were found for the other personality dimensions and thyroid hormones. 3.3. Regression To further examine the relationships between thyroid hormones and personality factors, we performed series of multivariate stepwise regression with thyroid hormones as dependent variable and personality scores as predictors. Age and sex were included as covariates. First, we tested the relationship between thyroid hormones and TCI-140 dimensions, separately for Temperament and Character scales. In the case of Temperament, only the model for TSH reached statistical significance (R = 0.36, R2 = 0.12, F = 7.69, p < 0.001) with standardized regression coefficient for Persistence (βPS) equal to 0.22. For Character dimensions, the TSH serum level was positively related to Self-Directedness with βSD = 0.22 (R = 0.38, R2 = 0.14, F = 7.83, p < 0.001), and TT4 negatively related to Cooperativeness with βC = −0.25 (R = 0.31, R2 = 0.10, F = 6.40, p < 0.01). Using stepwise regression, the NEO-FFI scales could not predict thyroid hormone levels with β statistically different from zero. However, the regression model with forced entry of predictors showed Neuroticism along with Age to be significant predictors of TSH with βN = −0.32 and βAge = 0.23 (R = 0.42, R2 = 0.17, F = 4.5, p < 0.01). The regression model with forced entry of Extraversion and Age as TSH level predictors was also significant, with βE = 0.25 and βAge = 0.23 (R = 0.34, R2 = 0.12, F = 2.8, p < 0.05).
3. Results 3.1. Descriptive data on thyroid hormones All subjects were clinically euthyroid: TSH range 0.42–3.98 M 1.74 ± 0.71, median 1.67 μlU/L, (normal reference range 0.4–4 μIU/ mL); FT3 range 1.24–6.14, M 4.84 ± 0.63, median 4.87 pmol/L (normal reference range 2.76–6.45 pmol/L); FT4 range 11.6–18.1, M 13.30 ± 1.35 pmol/L, median 13.39 (normal reference range 11.5–22.7 pmol/L); TT4 range 52–159, M 96 ± 13 nmol/L, median 94 (normal reference range 50–150 nmol/L). TSH was correlated to age (r = 0.28, p = 0.005), but not to other indices (p > 0.5), as is generally the case in people with no obvious thyroid disease (Bremner et al., 2012; Hadlow et al., 2013).
3.4. Extreme groups analysis Next, we compared personality profiles between extreme groups with relatively high and low levels of thyroid hormones. Classical cutoff values for extreme group analyses include M ± 1 SD, M ± 0.5 SD or use centiles such as deciles and quartiles (Preacher et al., 2005). More extreme cut-off values provide greater differences in dependent variable scores, but their statistical validity is reduced due to the decrease in the number of observations in each group. As our sample size was in the medium range (n = 104), we chose the moderate option for cut-off values, M ± 0.5 SD, which places almost a third of the observations (about 30.9%) in each extreme group. High TSH subjects scored higher in Persistence (56.3 ± 2.3 vs 47.8 ± 2.3, p = 0.02). With regard to the NEO-FFI, low TSH subjects scored higher on Neuroticism (52.9 ± 1.5 vs 46.7 ± 1.7, p = 0.016). Low and high FT3 groups showed no differences in TCI-140 or in NEOFFI scores. High FT4 subjects scored lower on Reward Dependence (46.1 ± 2. vs 52.4 ± 1.8, p = 0.041). High and low TT4 subjects showed no difference in TCI-140 scores. Neither FT4 nor TT4 extreme groups showed statistically significant differences in NEO-FFI scores.
3.2. Correlations Partial correlations between thyroid indices and personality scores controlling for age are shown in Table 3 for TCI-140 and in Table 4 for NEO-FFI. TSH was negatively correlated to Harm Avoidance Table 3 Partial correlations between thyroid indices and TCI-140 scores controlled for age.
Novelty Seeking Harm Avoidance Reward Dependence Persistence Self-Directedness Cooperativeness Self-Transcendence
TSH
FT3
FT4
TT4
−0,12 −0,24* 0,06 0,25** 0,27** 0,16 0,14
−0,04 0,11 −0,16 −0,01 −0,01 −0,12 0,08
0,11 −0,07 −0,19* −0,02 0,08 −0,05 0,10
0,03 0,04 −0,06 −0,18 −0,08 −0,22* −0,12
* − p < 0.05, ** − p < 0.01. n = 104.
183
Psychoneuroendocrinology 87 (2018) 181–187
A. Piskunov et al.
Fig. 1. LOESS graphs for TCI Character scores and TSH (a), FT3 (b), FT4 (b), TT4 (d). NS − Novelty Seeking, HA − Harm Avoidance, RD − Reward Dependence, PS − Persistence. The information to the publisher: graphs should be printed in color.
significant predictor of TSH level in regression analysis with a forced entry of predictors. FT3 was negatively related to Reward Dependence and positively related to Novelty Seeking, but only for values above the mean (Fig. 1b). These tendencies were also shown by correlation analysis but did not reach statistical significance. FT3 showed no expressed trends for TCI-140 Character and NEO-FFI dimensions, except for a relatively noticeable negative relationship between Conscientiousness (Fig. 3b) in agreement with the results of the other analyses. As in the correlational and extreme groups analyses, LOESS showed a negative relationship between FT4 and Reward Dependence (Fig. 1c), but only for subjects with FT4 around and above mean values. In addition, there were positive relationships between FT4 and both Cooperativeness and Self-Directedness for values below the mean (Fig. 2c) which were not revealed by the other analyses. FT4 was not associated with the NEO-FFI dimensions, except perhaps for a slight decrease in Extraversion with increased FT4 (Fig. 3c); this was also seen in the correlation analysis, although it was not statistically significant. TT4 was positively related with Novelty Seeking and negatively related with Persistence (for values below mean) (Fig. 1d), with both tendencies being insignificant in the other analyses. TT4 was also negatively related to all TCI-140 Character dimensions (Fig. 2d), especially to Cooperativeness, confirming the results of both correlational and regression analyses. TT4 according to the NEO-FFI dimensions plot (Fig. 3d) showed moderate negative relationships between the hormone level and Agreeableness, Extraversion and Conscientiousness which were also seen in correlation analysis, but were not statistically significant. Graphical analysis mostly confirmed the results of the other analyses on data processing, with a greater slope corresponding to a greater r-value in correlational and regression analyses. It also identified a number of relationships between personality and thyroid hormones that were not revealed by linear types of analyses, such as a positive relation between TSH and Cooperativeness for subjects with the hormone level above mean values, or a negative relationship between Persistence and
3.5. Graphical analysis The graphical analysis procedure using LOESS plots was performed to illustrate the data from other sections, and also to explore possible non-linear relationships between thyroid hormone levels and personality dimensions. In contrast to linear types of analysis, LOESS plots do not provide easily comparable statistical measures and do not exclude the influence of covariates such as age and sex. For this reason, this section is mostly illustrative and explorative. In order to standardize the graphs, personality raw scores were transformed into T-scores and thyroid hormone levels into z-scores. For the hormone axis, mean ± 2SD intervals were plotted since they contained the highest density of observations. According to the graphical analysis for TCI-140 temperamental scores (Fig. 1a), TSH was positively and almost linearly related to Persistence, confirming the results of the correlation, stepwise regression and extreme groups analyses. Subjects with TSH levels below the mean exhibited a negative relationship between hormone level and Novelty Seeking that was not evident during other types of analysis. Subjects with TSH serum levels above mean value showed a negative relationship between Harm Avoidance and hormone level that coincides with the results of the correlational analysis. Subjects with TSH above the mean value presented a positive relationship between the hormone level and Self-Directedness (Fig. 2a) which was previously revealed by both correlational and stepwise regression analyses. These subjects also showed a strong tendency towards above mean Cooperativeness and Self-Transcendence (Fig. 2a), which was observed during correlational analysis but did not reach statistical significance. In respect to NEO-FFI, TSH was negatively related to Neuroticism in a linear manner, i.e., with the same slope throughout the graph (Fig. 3a), this confirming the results of all the other types of analysis performed. Extraversion, Agreeableness and Conscientiousness were positively related to TSH level in a non-linear manner, with the slopes changing significantly along the graphs. These trends did not reach statistical significance in the other analyses except for Extraversion, which was a 184
Psychoneuroendocrinology 87 (2018) 181–187
A. Piskunov et al.
Fig. 2. LOESS graphs for TCI Temperament scores and TSH (a), FT3 (b), FT4 (b), TT4 (d). SDR − SelfDirectedness, C − Cooperativeness, ST − SelfTranscendence. The information to the publisher: graphs should be printed in color.
Fig. 3. LOESS graphs for NEO FFI scores and TSH (a), FT3 (b), FT4 (b), TT4 (d). O − Openness, C − Conscientiousness, E − Extraversion, A − Agreeableness, N − Neuroticism. The information to the publisher: graphs should be printed in color.
185
Psychoneuroendocrinology 87 (2018) 181–187
A. Piskunov et al.
Probably, these observations are related to a greater incidence of hyperthyroidism in psychiatric illness. However, as mentioned above, these data for TT4 were obtained from graphical analysis only and should be interpreted with caution. As the present study is based on correlational analysis techniques, it is difficult to investigate the cause-and-effect relationships. However, drawing on previously published data from animal and clinical studies we can propose some mechanisms for the action of the thyroid hormones on the brain. Both depression and personality dimensions of Neuroticism and Harm Avoidance have been associated with lower serotoninergic system activity (Cloninger, 2000; Jacobsen et al., 2012). Interestingly, in some studies the basal TSH level predicted susceptibility to selective serotonin reuptake inhibitors in depressive patients (Gitlin et al., 2004) with low TSH values associated with more clinical improvement. 5-HT activity in the brain may be reduced by the thyrotropin-releasing hormone (TRH) (Heal and Smith, 1988); this action might also explain the relationship observed between depression-related personality traits and low TSH due to a compensatory increase in TRH recorded in depressive patients (Banki et al., 1988; Frye et al., 1999). Indeed, we cannot rule out blunted TSH diurnal rhythm in depression (Peteranderl et al., 2002) as one of the possible reasons for the relationships observed in this study.
TT4 in subjects with relatively low TT4. However, these data need additional validation using other statistical methods, and, probably, a larger sample as well. 4. Discussion In our sample of normal human subjects, we found a relation between individual variation in serum thyroid hormone concentrations and specific aspects of personality functioning. TSH was the hormone with the strongest relationship with personality scores. TSH serum concentrations were positively associated with Persistence (in all types of analyses) and Self-Directedness (in partial correlations and regression analyses), and negatively associated with Harm Avoidance (by partial correlations and graphical analyses of TCI-140) and Neuroticism on the NEO-FFI (in all types of analysis). These results corroborate those obtained by Forman-Hoffman and Philibert (2006), Williams et al. (2009), and Medici et al. (2014) who showed that serum TSH is inversely associated with anxiety and depression (Neuroticism). Furthermore, our data are in full agreement with those of Frey et al. (2007) who revealed a negative correlation between Neuroticism and serum TSH in a sample of 121 healthy participants using the same psychometrical tool as the present article (NEO-FFI). These relationships suggest that higher TSH is associated with a more adaptive personality profile: that is, it facilitates optimal coping with daily problems, and promotes an adequate sense of self-efficacy and satisfaction with one’s life and one’s relations with others. According to Cloninger’s psychobiological model of temperament and character (Cloninger and Svrakic 1997; Svrakic et al., 2002; Svrakic and Cloninger, 2010), the adaptive personality profile is basically characterized by high scores on Self-Directedness and Cooperativeness dimensions, and low scores on Harm Avoidance (Spittlehouse et al., 2014). These findings may appear to challenge the common belief that low thyroid function is associated with affective disorders such as depression. In our study, however, we tested a normal population with no evidence of mental or thyroid illness. Moreover, the relationships observed between TSH level and personality traits were independent of the levels of peripheral thyroid hormones (FT3, FT4, TT4). Therefore, the correlations between TSH and personality obtained here are more likely to be associated with central than with peripheral thyroid function. One of the strengths of the present study is its use of the TCI-140, which provides much more detailed psychobiological information on the relationship between depressive symptoms and thyroid hormones. For instance, the present data suggest that the relationship between depression and low Self-Directedness, low Persistence, or low Cooperativeness observed in a number of studies (Matsudaira and Kitamura, 2006; Spittlehouse et al., 2014) may be partially mediated by low TSH. This study also showed a negative association between Reward Dependence and serum FT4 (in the partial correlation, extreme groups and graphical analyses) which was independent of age and TSH. Low Reward Dependence is known to be associated with cluster A personality disorders of DSM-5 (American Psychiatric Association, 2013) such as Schizoid and Schizotypal Personality Disorders (Ohi et al., 2012). Interestingly, several studies (Roca et al., 1990; Sim et al., 2002; Steiblienė et al., 2012; Akiibinu et al., 2012) have reported elevated T4 in these patients. Finally, by partial correlation, regression and graphical analyses we found a negative association between total T4 and Cooperativeness (by partial correlation, regression and graphical analysis), a personality dimension which has been shown to be decreased in almost all mental disorders and is indicative of problems in the interpersonal domain. Graphical analysis also revealed a relation between high TT4 and lower scores for Persistence and for all character dimensions (SelfDirectedness, Cooperativeness and Self-Transcendence), thus representing a maladaptive personality profile (Spittlehouse et al., 2014).
5. Conclusions The present study shows that the relationship between personality and thyroid hormones may be more complex than previously supposed. The most significant findings are the positive association of serum TSH with Persistence and Self-Directedness (maturity and self-regulation) and its negative association with Harm Avoidance and NEO-FFI Neuroticism (emotional dysregulation). Thus, our study supports the hypothesis that reduced serum TSH may represent the endophenotype associated with a maladaptive personality profile. Increased FT4 was associated with lower Reward-Dependence and increased TT4 with lower Cooperativeness (both suggesting a deficit in social attachment). Studies with larger sample sizes and facet level analyses may broaden our understanding of the relationship between thyroid hormones and personality. 6. Limitations The unavailability of Cronbach's alphas for NEO-FFI was one of the limitations of the study. Other limitations were the lack of information on TSH diurnal rhythm and the inability to fully control all the factors that might influence TSH level (in particular, individual sleep patterns). Finally, because of the sample size used we were unable to study the relationship between specific personality profiles (resulting of the combination of main dimensions) and the dependent variables studied. Contributors AP, NT, AF & JR designed the study, wrote the protocol, and conducted literature searches. AP & NT conducted the study. AP & AF analysed the data. AP, NT, AF, AM & JR interpreted the data. AP & AF wrote the first draft of the manuscript. All authors contributed to the manuscript and have approved the final version. Conflict of interest The authors have no conflict of interest to declare regarding this manuscript. References Akiibinu, M.O., Ogundahunsi, O.A., Ogunyemi, E.O., 2012. Inter-relationship of plasma markers of oxidative stress and thyroid hormones in schizophrenics. BMC Res. Notes
186
Psychoneuroendocrinology 87 (2018) 181–187
A. Piskunov et al.
Kudryashev, A.F. (Ed.), 1992. Luchshie psichologicheskiye testi dlya profotbora i proforientacii [Best Psychological Tests for Vocational Assessment and Orientation]. Petrozavodsk University, Petroza-vodsk, Russia. Matsudaira, T., Kitamura, T., 2006. Personality traits as risk factors of depression and anxiety among Japanese students. J. Clin. Psychol. 62 (1), 97–109. Medici, M., Direk, N., Visser, W.E., Korevaar, T.I., Hofman, A., Visser, T.J., Tiemeier, H., Peeters, R.P., 2014. Thyroid function within the normal range and the risk of depression: a population-based cohort study. J. Clin. Endocrinol. Metab. 99 (4), 1213–1219. http://dx.doi.org/10.1210/jc.2013-3589. Morte, B., Bernal, J., 2014. Thyroid hormone action: astrocyte-neuron communication. Front. Endocrinol. (Lausanne) 5, 82. http://dx.doi.org/10.3389/fendo.2014.00082. Netter, P., 2004. Personality and hormones. In: Stelmack, R.M. (Ed.), On the Psychobiology of Personality: Essays in Honor of Marvin Zuckerman. Elsevier, Ltd., Oxford, pp. 353–377. Ohi, K., Hashimoto, R., Yasuda, Y., Fukumoto, M., Yamamori, H., Iwase, M., Kazui, H., Takeda, M., 2012. Personality traits and schizophrenia: evidence from a case-control study and meta-analysis. Psychiatry Res. 198 (1), 7–11. http://dx.doi.org/10.1016/j. psychres.2011.12.018. Peteranderl, C., Antonijevic, I.A., Steiger, A., Murck, H., Held, K., Frieboes, R.M., Uhr, M., Schaaf, L., 2002. Nocturnal secretion of TSH and ACTH in male patients with depression and healthy controls. J. Psychiatr. Res. 36, 189–196. Préau, L., Fini, J.B., Morvan-Dubois, G., Demeneix, B., 2014. Thyroid hormone signaling during early neurogenesis and its significance as a vulnerable window for endocrine disruption. Biochim. Biophys. 1874–9399 (14), 00167–00169. http://dx.doi.org/10. 1016/j.bbagrm.2014.06.015. Preacher, K.J., Rucker, D.D., MacCallum, R.C., Nicewander, W.A., 2005. Use of the extreme groups approach: a critical reexamination and new recommendations. Psychol. Methods 10, 178–192. Roca, R.P., Blackman, M.R., Ackerley, M.B., Harman, S.M., Gregerman, R.I., 1990. Thyroid hormone elevations during acute psychiatric illness: relationship to severity and distinction from hyperthyroidism. Endocr. Res. 16 (4), 415–447. Sánchez-Huerta, K., García-Martínez, Y., Vergara, P., Segovia, J., Pacheco-Rosado, J., 2016. Thyroid hormones are essential to preserve non-proliferative cells of adult neurogenesis of the dentate gyrus. Mol. Cell. Neurosci. 76, 1–10. http://dx.doi.org/ 10.1016/j.mcn.2016.08.001. Schroeder, A.C., Privalsky, M.L., 2014. Thyroid hormones, t3 and t4, in the brain. Front. Endocrinol. (Lausanne) 5, 40. http://dx.doi.org/10.3389/fendo.2014.00040. Sierra, P., Cámara, R., Tobella, H., Livianos, L., 2014. What is the real significance and management of major thyroid disorders in bipolar patients? Rev. Psiquiatr. Salud Ment. 7 (2), 88–95. http://dx.doi.org/10.1016/j.rpsm.2013.07.005. Sim, K., Chong, S.A., Chan, Y.H., Lum, W.M., 2002. Thyroid dysfunction in chronic schizophrenia within a state psychiatric hospital. Ann. Acad. Med. Singapore 31 (5), 641–644. Sinai, C., Hirvikoski, T., Dencker, E., Nordströmm, A.L., Linder, J., Nordströmm, P., Jokinen, J., 2009. Thyroid hormones and personality traits in attempted suicide. Psychoneuroendocrinology 34, 1526–1532. http://dx.doi.org/10.1016/j.psyneuen. 2009.05.009. Spittlehouse, J.K., Vierck, E., Pearson, J.F., Joyce, P.R., 2014. Temperament and character as determinants of well-being. Compr. Psychiatry 55 (7), 1679–1687. http://dx. doi.org/10.1016/j.comppsych.2014.06.011. Steiblienė, V., Mickuvienė, N., Prange Jr, A.J., Bunevičius, R., 2012. Concentrations of thyroid axis hormones in psychotic patients on hospital admission: the effects of prior drug use. Medicina (Kaunas) 48 (5), 229–234. Svrakic, D.M., Cloninger, R.C., 2010. Epigenetic perspective on behaviour development, personality, and personality disorders. Psychiatr. Danub. 22 (2), 153–166. Svrakic, D.M., Draganic, S., Hill, K., Bayon, C., Przybeck, T.R., Cloninger, R.C., 2002. Temperament, character, and personality disorders: etiologic diagnostic, treatment issues. Acta Psychiatr. Scand. 106, 189–195. Williams, M.D., Harris, R., Dayan, C.M., Evans, J., Gallacher, J., Ben-Shlomo, Y., 2009. Thyroid function and the natural history of depression: findings from the Caerphilly Prospective Study (CaPS) and a meta-analysis. Clin. Endocrinol. (Oxf.) 70 (3), 484–492. http://dx.doi.org/10.1111/j.1365-2265.2008.03352.x. Ylikahri, R.H., Huttunen, M.O., Härkönen, M., 1980. Hormonal changes during alcohol intoxication and withdrawal. Pharmacol. Biochem. Behav. 13 (Suppl 1), 131–137. Zuckerman, M., 1997. The psychobiological basis of personality. In: Nyborg, H. (Ed.), The Scientific Study of Human Nature. Oxford, Pergamon, pp. 3–16.
5, 169. http://dx.doi.org/10.1186/17560500-5-169. American Psychiatric Association, 2013. Diagnostic and Statistical Manual of Mental Disorders, 5th ed. (Washington, D.C.). Arqué, J.M., Segura, R., Torrubia, R., 1987. Correlation of thyroxine and thyroid-stimulation hormone with personality measurements: a study in psychosomatic patients and healthy subjects. Neuropsychobiology 18 (3), 127–133. Balada, F., Torrubia, R., Arqué, J.M., 1992. Thyroid hormone correlates of sensation seeking and anxiety in healthy human females. Neuropsychobiology 25 (4), 208–213. Banki, C.M., Bissette, G., Arato, M., Nemeroff, C.B., 1988. Elevation of immunoreactive CSF TRH in depressed patients. Am. J. Psychiatry 145, 1526–1531. Baturina, N.A., Aidman, B.V., 2010. Metodoki psikhologicheskoy diagnostiki I izmereniya: Ezhegodnik proffesionalnyh recenziy i obzorov. SUSU, Chelyabinsk, pp. 49–55. Bauer, M., Glenn, T., Pilhatsch, M., Pfennig, A., Whybrow, P.C., 2014. Gender differences in thyroid system function: relevance to bipolar disorder and its treatment. Bipolar. Disord. 16 (1), 58–71. http://dx.doi.org/10.1111/bdi.12150. Bremner, A.P., Feddema, P., Leedman, P.J., Brown, S.J., Beilby, J.P., Lim, E.M., Wilson, S.G., O'Leary, P.C., Walsh, J.P., 2012. Age-related changes in thyroid function: a longitudinal study of a community-based cohort. J. Clin. Endocrinol. Metab. 97, 1554–1562. Cloninger, C.R., Svrakic, D.M., 1997. Integrative psychobiological approach to psychiatric assessment and treatment. Psychiatry 60 (2), 120–141. Cloninger, C.R., 1999. The Temperament and Character Inventory. Center for Psychobiology of Personality, Washington University, Revised. St. Louis, MO (Available from C. R. Cloninger, Washington University School of Medicine, Department of Psychiatry, PO Box 8134, St. Louis, MO, 63110). Cloninger, C.R., 2000. Biology of personality dimensions. Curr. Opin. Psychiatry 13, 611–616. Costa, P.T., McCrae, R.R., 1992. Revised NEO Personality Inventory (NEO-PI-R) and NEO Five-Factor Inventory (NEO-FFI) Manual. Psychological Assessment Resources, Odessa, FL. Forman-Hoffman, V., Philibert, R.A., 2006. Lower TSH and higher T4 levels are associated with current depressive syndrome in young adults. Acta Psychiatr. Scand. 114, 132–139. Fox, J., 2000. Nonparametric Simple Regression: Smoothing Scatterplots. Thousand Oaks, CA. Frey, A., Lampert, A., Dietz, K., Striebich, S., Locher, C., Fedorenko, O., Möhle, R., Gallinat, J., Lang, F., Lang, U.E., 2007. Neuropsychobiology 56 (2–3), 123–126. http://dx.doi.org/10.1159/000112954. Frye, M.A., Gary, K.A., Marangell, L.B., George, M.S., Callahan, A.M., Little, J.T., Huggins, T., Corá Locatelli, G., Osuch, E.A., Winokur, A., Post, R.M., 1999. CSF thyrotropin releasing hormone gender difference: implications for neurobiology and treatment of depression. J. Neuropsych. Clin. Neurosci. 11, 349–353. Gitlin, M., Altshuler, L.L., Frye, M.A., Suri, R., Huynh, E.L., Fairbanks, L., Bauer, M., Korenman, S., 2004. Peripheral thyroid hormones and response to selective serotonin reuptake inhibitors. J. Psychiatry Neurosci. 29 (5), 383–386. Gray, J.A., 1993. The Psychology of Fear and Stress. Cambridge University Press, London. Hadlow, N.C., Rothacker, K.M., Wardrop, R., Brown, S.J., Lim, E.M., Walsh, J.P., 2013. The relationship between TSH and free T4 in a large population is complex and nonlinear and differs by age and sex. J. Clin. Endocrinol. Metab. 98, 2936–2943. Heal, D.J., Smith, S.L., 1988. The effects of acute and repeated administration of T3 to mice on 5-HT1 and 5-HT2 function in the brain and its influence on the actions of repeated electroconvulsive shock. Neuropharmacology 27, 1239–1248. Hein, M.D., Jackson, I.M., 1990. Review: thyroid function in psychiatric illness. Gen. Hosp. Psychiatry 12, 232–244. Jackson, I.M., 1998. The thyroid axis and depression. Thyroid 8, 951–956. Jacobsen, J.P., Medvedev, I.O., Caron, M.G., 2012. The 5-HT deficiency theory of depression: perspectives from a naturalistic 5-HT deficiency model, the tryptophan hydroxylase 2Arg439His knockin mouse. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 367 (1601), 2444–2459. http://dx.doi.org/10.1098/rstb.2012.0109. Johnson, A.M., Vernon, P.A., Feiler, A.R., 2008. Behavioural Genetic Studies of Personality: An Introduction and Review of the Results of 5 Years of Research. Handbook of Personality Theory and Assessment. Sage, London, pp. 145–173. Kessler, L., Nedeltceva, A., Imperial, J., Penev, P.D., 2010. Changes in serum TSH and free T4 during human sleep restriction. Sleep 33, 1115–1118. Kirkegaard, C., Faber, J., 1998. The role of thyroid hormones in depression. Eur. J. Endocrinol. 138, 1–9.
187
Contents lists available at ScienceDirect
Psychoneuroendocrinology journal homepage: www.elsevier.com/locate/psyneuen
The hypothalamic-pituitary-thyroid axis and personality in a sample of healthy subjects
T
⁎
Aleksei Piskunova, , Adela Fustéa, Nadezhda Teryaevab, Aleksei Moshkinb, José Ruiza a b
Department of Personality, Assessment and Psychological Treatment, Faculty of Psychology, University of Barcelona, Spain N.N. Burdenko Neurosurgical Institute, Moscow, Russia
A R T I C L E I N F O
A B S T R A C T
Keywords: TSH Thyroxin Triiodothyronine Personality Health Neuroticism Depression
Thyroid hormones influence various brain pathologies, including psychiatric disorders. However, the relationship between these hormones and the psychological state in the normal, non-clinical population is poorly understood. We aimed to test whether serum levels of thyroid hormones are associated with personality in the healthy population. Thyrotropin (TSH), free T3 (FT4), total and free T4 (TT4 and FT4) concentrations were measured in the blood of 104 healthy participants (44% men) aged 18–59 (M = 35 ± 9). Personality traits were assessed using the revised short versions of the Temperament and Character Inventory (TCI-140) and the NEO Five-Factor Inventory (NEO-FFI). The data were analysed by correlational, regression, extreme groups and graphical techniques, which showed significant correlations between inter-individual variations in serum thyroid hormone levels and specific aspects of personality. In particular, high serum TSH was strongly associated with higher Persistence and Self-Directedness, and negatively correlated with Harm Avoidance on the TCI-140 and Neuroticism on the NEO-FFI, thus representing a more adaptive personality profile. Furthermore, increased FT4 was associated with lower Reward-Dependence, and increased TT4 was associated with lower Cooperativeness, representing a deficit in social attachment. Our data indicate that the relationship between thyroid hormones and personality in the healthy population might be rather more complex than the results obtained in clinical samples suggest.
1. Introduction The study of the relationship between hormones and personality dates back to Antiquity, when Hippocrates and Galen proposed the first notions of humoral typologies based on the deficit or excess of body fluids. Today, according to Netter (2004), the hormones most frequently studied in relation to personality are the ones associated with the hypothalamo-pituitary-adrenal (HPA) axis such as cortisol, the ones in the gonadal-pituitary-adrenal (HPG) axis such as testosterone, and the ones in the adrenomedullary axis such as the catecholamines noradrenaline, adrenaline and dopamine. The most significant results of the study of HPA axis hormones and personality show a positive relationship between cortisol and Neuroticism (anxiety and depression), and a negative relationship with Novelty/Sensation seeking and Aggressiveness. For its part, testosterone has also been associated with Dominance, Aggressiveness and Novelty/sensation seeking. Likewise, higher levels of noradrenaline have been associated with Anger and Aggressiveness, while adrenaline levels are more related to Fear and serotonin levels to Anxiety (Gray, 1993). In contrast, the dopaminergic
⁎
system has been fundamentally associated with Novelty/Sensation seeking, and the serotoninergic system with Harm Avoidance (Cloninger, 2000). In addition, the adrenomedullary axis (the catecholaminergic system) in interaction with the HPG axis has been considered as the regulatory mechanism of emotions, social behavior and clinical conditions (Zuckerman, 1997). Thyroid hormones appear to be less studied in the relation to personality. However, there is a lot of background data. A variety of mechanisms of thyroid hormone uptake into brain tissues and hormone activation and their influences on neurotransmitter generation have been described (for a review, see Schroeder and Privalsky, 2014). Thyroid hormones are also involved in neurogenesis and neuron-astrocyte communication (Morte and Bernal, 2014; Préau et al., 2014; Sánchez-Huerta et al., 2016). Thyroid-stimulating hormone (TSH) serum level predicts response to selective serotonin reuptake inhibitors in depressive patients, suggesting a relationship between thyrotrophic hormone and serotoninergic system susceptibility (Gitlin et al., 2004). TSH secretion is exerted by classical negative feedback loop, thus elevated TSH is believed to indicate low peripheral thyroid function. An
Corresponding author. E-mail address: [email protected] (A. Piskunov).
http://dx.doi.org/10.1016/j.psyneuen.2017.10.023 Received 6 June 2017; Received in revised form 6 September 2017; Accepted 29 October 2017 0306-4530/ © 2017 Elsevier Ltd. All rights reserved.
Psychoneuroendocrinology 87 (2018) 181–187
A. Piskunov et al.
«did not sleep at all») and how they felt at the present moment. Seven participants were excluded from the data analysis: one who reported feeling unwell, four who had drunk alcohol the day before, and two who had slept very poorly or not at all the previous night.
association between hypothyroidism and depression has been long accepted by scientists and clinical specialists. Indeed, many studies suggest that the hypothalamic–pituitary–thyroid axis (HPTA) is involved in pathogenesis of psychiatric disorders, especially mood disorders such as depression (Hein and Jackson, 1990) and bipolar disorder (Sierra et al., 2014; Bauer et al., 2014). Hypothyroidism and depression share some symptoms (Jackson, 1998) and many patients with depression have abnormal thyroid hormone levels (Kirkegaard and Faber, 1998). However, larger studies on the relationship of thyroid hormones and depression conducted over the last 15 years in normal populations, including studies with prospective and community approaches, have unexpectedly shown serum TSH to be inversely associated with anxiety and depression (Forman-Hoffman and Philibert, 2006; Williams et al., 2009; Medici et al., 2014). Furthermore, a negative correlation has been found between serum TSH and Neuroticism (Frey et al., 2007). Regarding peripheral thyroid hormones, several studies (Roca et al., 1990; Sim et al., 2002; Steiblienė et al., 2012; Akiibinu et al., 2012) have reported elevated T4 in patients with cluster A personality disorders (e.g., Schizoid and Schizotypal). Stalenheim et al. (1998; cited by Sinai et al., 2009) found that elevated T3 levels were associated with psychopathy and antisocial disorder, whereas serum levels of FT4 were negatively related to these disorders. Sinai et al. (2009) also mention other studies (Arqué et al., 1987; Balada et al., 1992) suggesting a negative correlation between TSH and T4 with sensation seeking, and a positive correlation between T4 and scores on depression-anxiety scales in healthy females with no psychiatric history. As can be seen from the above, despite the long history of investigation, the relationship between thyroid hormones and psychological and psychiatric functions is not clearly defined. Most of the studies in the area dealt with thyroid pathology or psychiatric illness. However, we believe that better and more fundamental understanding of the problem might be obtained by studying the role of HPTA in health, i.e. by investigating the role of thyroid hormones in normal regulation of psychological functions, such as personality, in subjects with no thyroid disease. In the present study we aimed to explore in detail whether personality dimensions in mentally healthy individuals might be related to physiological concentrations of serum thyroid hormones.
2.2. Psychological measures Two models for assessing personality were used: the NEO Five Factor Inventory (NEO-FFI; Costa and McCrae, 1992) and the short version (TCI-140) of the revised Cloninger’s Temperament and Character Inventory (Cloninger, 1999). The NEO-FFI is based on a lexical analysis of trait adjectives and the TCI-140 is based on a psychobiological model that accounts for individual differences in personality traits by integrating neurochemical systems, learning, and social influences. Personality measures from these models correlate to problem behaviors and psychiatric diagnoses. Personality dimensions assessed with both the NEO-FFI and the TCI-R demonstrate broad-sense heritability in excess of 30% (Johnson et al., 2008). All participants completed the short version of the revised 5-point Likert scale version of TCI-140, which allows valid personality assessment over four dimensions of temperament (NS: Novelty Seeking, HA: Harm Avoidance, RD: Reward Dependence, PS: Persistence) and three dimensions of character (SD: Self-Directedness, С: Cooperativeness, ST: Self-Transcendence). The NEO-FFI was performed in 71 subjects (70%). The participants completed the short 60-item version of the NEO-FFI, which allows reliable and valid assessment of personality along the dimensions Neuroticism (N), Extraversion (E), Openness to experiences (O), Agreeableness (A), and Conscientiousness (C) (Kudryashev, 1992; Baturina and Aidman, 2010). If not indicated as a raw score, each dimensional score is presented as a T-score with a mean of 50 and a standard deviation of 10. Table 1 shows descriptive statistics and reliability data for TCI-140, and Table 2 shows descriptive statistics for NEO-FFI. Since we used a commercial version of the questionnaire that does not provide scoring keys or raw items data, we were unable to calculate Cronbach’s alphas and so reliability data on NEO-FFI are not presented here. 2.3. Thyroid hormone measurements
2. Methods All analyses were performed in accordance with accredited routines at the Laboratory for Clinical Biochemistry of the Burdenko Neurosurgical Institute. Blood was collected in the morning between 8:00 a.m. and 10:30 a.m. after an overnight fast using separating gel tubes. Serum was separated by centrifugation at 3000g for 15 min. The levels of thyrotropin (TSH), total and free thyroxin (TT4 and FT4) and free triiodothyronine (FT3) were determined using automated chemiluminescence immunoassay (“Immulite2000”, Siemens, USA).
2.1. Participants A total of 104 healthy volunteers (46 male, 58 female, aged 18–55 years, M35 ± 12) from the Moscow region in Russia were recruited for this study via a social network advertisement containing a brief description of the study, which was posted in the online communities of students at the biology and psychology faculties and on the personal webpages of researchers. The advertisement invited students and researchers to participate in the study and asked them to share the invitation with their peers. As an incentive to take part, prospective participants were told that they would be given information on their blood hormone levels and psychological measures. Before enrollment, participants completed an online medical history in order to assess the presence of any clinical syndrome (DSM, Axis I; American Psychiatric Association, 2013), medical condition or any thyroid or psychotropic drug intake, all of which were considered as exclusion criteria. After a complete description of the study to the subjects, written informed consent was obtained. The study was approved by the local ethics committee of the N.N. Burdenko Neurosurgical Institute. Participants were asked not to come for blood sampling if they had slept poorly the night before, and not to drink alcohol in the preceding days before in order to mitigate the influence of these factors on thyroid hormone plasma levels (Ylikahri et al., 1980; Kessler et al., 2010). On the day of blood sampling, participants were asked to state when they had last drunk alcohol, how they had slept the previous night (recorded on a 1–5 Likert scale with «5» corresponding to «slept perfectly» and «1»
2.4. Statistical analysis We used correlations, extreme groups, graphical and regression techniques to investigate the relationships between serum thyroid hormones and personality traits. Each type of analysis complemented Table 1 TCI-140 descriptive statistics and alpha internal consistency.
Novelty Seeking Harm Avoidance Reward Dependence Persistence Self-Directedness Cooperativeness Self-Transcendence
Mean
SD
S
K
alpha
62.3 55.2 62.3 68.1 65.7 72.3 46.7
9.08 12.7 10.9 10.0 10.4 11.2 8.2
0.19 0.13 −0.24 0.08 −0.55 −0.70 −0.22
0.16 −0.47 0.25 0.72 0.44 0.34 0.04
0.71 0.87 0.81 0.84 0.80 0.84 0.71
K − kurtosis, S − skewness, n = 104.
182
Psychoneuroendocrinology 87 (2018) 181–187
A. Piskunov et al.
Table 2 NEO-FFI descriptive statistics.
Table 4 Partial correlations between thyroid indices and NEO-FFI scores controlled for age.
Neuroticism Extraversion Openness Agreeableness Conscientiousness
Mean
SD
26.1 28.2 32.2 29.1 29.8
7.61 6.77 5.81 7.22 9.91
Neuroticism Extraversion Openness Agreeableness Conscientiousness
TSH
FT3
FT4
TT4
−0,30** 0,22 0,01 0,19 0,13
0,11 −0,01 −0,07 −0,19 0,00
0,00 −0,21 −0,16 −0,10 0,10
0,06 −0,11 −0,10 −0,14 −0,15
n = 71.
** − p < 0.01. n = 71.
the other. The distributions of variables were initially plotted to identify pathological cases. Cases with TSH values beyond normal range (0.4–4 μIU/mL) were excluded (four cases in total) in order to form a clinically euthyroid sample. Gender effects were tested using the T-test for independent samples, and age effects were determined with Pearson's correlation test. Thyroid indices were correlated with NEOFFI and TCI-140 scores using a partial correlation test. The results are presented as M ± 1 SD. The p-values are two-tailed. The variance inflation factor (VIF) and tolerance statistic indicated no problem with multicollinearity. In all analyses, p values less than 0.05 were considered significant. All calculations were carried out using the IBM SPSS (version 22). Nonlinear relationships were analysed through the LOcally WEighted Scatter-plot Smoother (LOESS), non-parametric, local area, polynomial regression procedure (Fox, 2000) to produce data points for the TCI-140 and NEO-FFI scores. This method involves a series of local regression analyses that allows the shape of a curve to vary across the variable continua. The resulting curves are the best unbiased depictions of the patterns in the data. The curves are much more accurate than the lines (linear or quadratic) that are imposed on the data in familiar parametric analyses (Fox, 2000).
(r = −0.24, p < 0.05) and positively correlated to Persistence (r = 0.25, p < 0.05) and Self-Directedness (r = 0.27, p < 0.01). FT3 showed no significant correlations with TCI-140 scores. FT4 was negatively correlated to Reward Dependence (r = −0.19, p < 0.05). TT4 was negatively correlated to Cooperativeness (r = −0.22, p < 0.05). For the NEO-FFI dimensions, only the correlation between TSH and Neuroticism was significant (r = −0.30, p < 0.01). No further significant correlations were found for the other personality dimensions and thyroid hormones. 3.3. Regression To further examine the relationships between thyroid hormones and personality factors, we performed series of multivariate stepwise regression with thyroid hormones as dependent variable and personality scores as predictors. Age and sex were included as covariates. First, we tested the relationship between thyroid hormones and TCI-140 dimensions, separately for Temperament and Character scales. In the case of Temperament, only the model for TSH reached statistical significance (R = 0.36, R2 = 0.12, F = 7.69, p < 0.001) with standardized regression coefficient for Persistence (βPS) equal to 0.22. For Character dimensions, the TSH serum level was positively related to Self-Directedness with βSD = 0.22 (R = 0.38, R2 = 0.14, F = 7.83, p < 0.001), and TT4 negatively related to Cooperativeness with βC = −0.25 (R = 0.31, R2 = 0.10, F = 6.40, p < 0.01). Using stepwise regression, the NEO-FFI scales could not predict thyroid hormone levels with β statistically different from zero. However, the regression model with forced entry of predictors showed Neuroticism along with Age to be significant predictors of TSH with βN = −0.32 and βAge = 0.23 (R = 0.42, R2 = 0.17, F = 4.5, p < 0.01). The regression model with forced entry of Extraversion and Age as TSH level predictors was also significant, with βE = 0.25 and βAge = 0.23 (R = 0.34, R2 = 0.12, F = 2.8, p < 0.05).
3. Results 3.1. Descriptive data on thyroid hormones All subjects were clinically euthyroid: TSH range 0.42–3.98 M 1.74 ± 0.71, median 1.67 μlU/L, (normal reference range 0.4–4 μIU/ mL); FT3 range 1.24–6.14, M 4.84 ± 0.63, median 4.87 pmol/L (normal reference range 2.76–6.45 pmol/L); FT4 range 11.6–18.1, M 13.30 ± 1.35 pmol/L, median 13.39 (normal reference range 11.5–22.7 pmol/L); TT4 range 52–159, M 96 ± 13 nmol/L, median 94 (normal reference range 50–150 nmol/L). TSH was correlated to age (r = 0.28, p = 0.005), but not to other indices (p > 0.5), as is generally the case in people with no obvious thyroid disease (Bremner et al., 2012; Hadlow et al., 2013).
3.4. Extreme groups analysis Next, we compared personality profiles between extreme groups with relatively high and low levels of thyroid hormones. Classical cutoff values for extreme group analyses include M ± 1 SD, M ± 0.5 SD or use centiles such as deciles and quartiles (Preacher et al., 2005). More extreme cut-off values provide greater differences in dependent variable scores, but their statistical validity is reduced due to the decrease in the number of observations in each group. As our sample size was in the medium range (n = 104), we chose the moderate option for cut-off values, M ± 0.5 SD, which places almost a third of the observations (about 30.9%) in each extreme group. High TSH subjects scored higher in Persistence (56.3 ± 2.3 vs 47.8 ± 2.3, p = 0.02). With regard to the NEO-FFI, low TSH subjects scored higher on Neuroticism (52.9 ± 1.5 vs 46.7 ± 1.7, p = 0.016). Low and high FT3 groups showed no differences in TCI-140 or in NEOFFI scores. High FT4 subjects scored lower on Reward Dependence (46.1 ± 2. vs 52.4 ± 1.8, p = 0.041). High and low TT4 subjects showed no difference in TCI-140 scores. Neither FT4 nor TT4 extreme groups showed statistically significant differences in NEO-FFI scores.
3.2. Correlations Partial correlations between thyroid indices and personality scores controlling for age are shown in Table 3 for TCI-140 and in Table 4 for NEO-FFI. TSH was negatively correlated to Harm Avoidance Table 3 Partial correlations between thyroid indices and TCI-140 scores controlled for age.
Novelty Seeking Harm Avoidance Reward Dependence Persistence Self-Directedness Cooperativeness Self-Transcendence
TSH
FT3
FT4
TT4
−0,12 −0,24* 0,06 0,25** 0,27** 0,16 0,14
−0,04 0,11 −0,16 −0,01 −0,01 −0,12 0,08
0,11 −0,07 −0,19* −0,02 0,08 −0,05 0,10
0,03 0,04 −0,06 −0,18 −0,08 −0,22* −0,12
* − p < 0.05, ** − p < 0.01. n = 104.
183
Psychoneuroendocrinology 87 (2018) 181–187
A. Piskunov et al.
Fig. 1. LOESS graphs for TCI Character scores and TSH (a), FT3 (b), FT4 (b), TT4 (d). NS − Novelty Seeking, HA − Harm Avoidance, RD − Reward Dependence, PS − Persistence. The information to the publisher: graphs should be printed in color.
significant predictor of TSH level in regression analysis with a forced entry of predictors. FT3 was negatively related to Reward Dependence and positively related to Novelty Seeking, but only for values above the mean (Fig. 1b). These tendencies were also shown by correlation analysis but did not reach statistical significance. FT3 showed no expressed trends for TCI-140 Character and NEO-FFI dimensions, except for a relatively noticeable negative relationship between Conscientiousness (Fig. 3b) in agreement with the results of the other analyses. As in the correlational and extreme groups analyses, LOESS showed a negative relationship between FT4 and Reward Dependence (Fig. 1c), but only for subjects with FT4 around and above mean values. In addition, there were positive relationships between FT4 and both Cooperativeness and Self-Directedness for values below the mean (Fig. 2c) which were not revealed by the other analyses. FT4 was not associated with the NEO-FFI dimensions, except perhaps for a slight decrease in Extraversion with increased FT4 (Fig. 3c); this was also seen in the correlation analysis, although it was not statistically significant. TT4 was positively related with Novelty Seeking and negatively related with Persistence (for values below mean) (Fig. 1d), with both tendencies being insignificant in the other analyses. TT4 was also negatively related to all TCI-140 Character dimensions (Fig. 2d), especially to Cooperativeness, confirming the results of both correlational and regression analyses. TT4 according to the NEO-FFI dimensions plot (Fig. 3d) showed moderate negative relationships between the hormone level and Agreeableness, Extraversion and Conscientiousness which were also seen in correlation analysis, but were not statistically significant. Graphical analysis mostly confirmed the results of the other analyses on data processing, with a greater slope corresponding to a greater r-value in correlational and regression analyses. It also identified a number of relationships between personality and thyroid hormones that were not revealed by linear types of analyses, such as a positive relation between TSH and Cooperativeness for subjects with the hormone level above mean values, or a negative relationship between Persistence and
3.5. Graphical analysis The graphical analysis procedure using LOESS plots was performed to illustrate the data from other sections, and also to explore possible non-linear relationships between thyroid hormone levels and personality dimensions. In contrast to linear types of analysis, LOESS plots do not provide easily comparable statistical measures and do not exclude the influence of covariates such as age and sex. For this reason, this section is mostly illustrative and explorative. In order to standardize the graphs, personality raw scores were transformed into T-scores and thyroid hormone levels into z-scores. For the hormone axis, mean ± 2SD intervals were plotted since they contained the highest density of observations. According to the graphical analysis for TCI-140 temperamental scores (Fig. 1a), TSH was positively and almost linearly related to Persistence, confirming the results of the correlation, stepwise regression and extreme groups analyses. Subjects with TSH levels below the mean exhibited a negative relationship between hormone level and Novelty Seeking that was not evident during other types of analysis. Subjects with TSH serum levels above mean value showed a negative relationship between Harm Avoidance and hormone level that coincides with the results of the correlational analysis. Subjects with TSH above the mean value presented a positive relationship between the hormone level and Self-Directedness (Fig. 2a) which was previously revealed by both correlational and stepwise regression analyses. These subjects also showed a strong tendency towards above mean Cooperativeness and Self-Transcendence (Fig. 2a), which was observed during correlational analysis but did not reach statistical significance. In respect to NEO-FFI, TSH was negatively related to Neuroticism in a linear manner, i.e., with the same slope throughout the graph (Fig. 3a), this confirming the results of all the other types of analysis performed. Extraversion, Agreeableness and Conscientiousness were positively related to TSH level in a non-linear manner, with the slopes changing significantly along the graphs. These trends did not reach statistical significance in the other analyses except for Extraversion, which was a 184
Psychoneuroendocrinology 87 (2018) 181–187
A. Piskunov et al.
Fig. 2. LOESS graphs for TCI Temperament scores and TSH (a), FT3 (b), FT4 (b), TT4 (d). SDR − SelfDirectedness, C − Cooperativeness, ST − SelfTranscendence. The information to the publisher: graphs should be printed in color.
Fig. 3. LOESS graphs for NEO FFI scores and TSH (a), FT3 (b), FT4 (b), TT4 (d). O − Openness, C − Conscientiousness, E − Extraversion, A − Agreeableness, N − Neuroticism. The information to the publisher: graphs should be printed in color.
185
Psychoneuroendocrinology 87 (2018) 181–187
A. Piskunov et al.
Probably, these observations are related to a greater incidence of hyperthyroidism in psychiatric illness. However, as mentioned above, these data for TT4 were obtained from graphical analysis only and should be interpreted with caution. As the present study is based on correlational analysis techniques, it is difficult to investigate the cause-and-effect relationships. However, drawing on previously published data from animal and clinical studies we can propose some mechanisms for the action of the thyroid hormones on the brain. Both depression and personality dimensions of Neuroticism and Harm Avoidance have been associated with lower serotoninergic system activity (Cloninger, 2000; Jacobsen et al., 2012). Interestingly, in some studies the basal TSH level predicted susceptibility to selective serotonin reuptake inhibitors in depressive patients (Gitlin et al., 2004) with low TSH values associated with more clinical improvement. 5-HT activity in the brain may be reduced by the thyrotropin-releasing hormone (TRH) (Heal and Smith, 1988); this action might also explain the relationship observed between depression-related personality traits and low TSH due to a compensatory increase in TRH recorded in depressive patients (Banki et al., 1988; Frye et al., 1999). Indeed, we cannot rule out blunted TSH diurnal rhythm in depression (Peteranderl et al., 2002) as one of the possible reasons for the relationships observed in this study.
TT4 in subjects with relatively low TT4. However, these data need additional validation using other statistical methods, and, probably, a larger sample as well. 4. Discussion In our sample of normal human subjects, we found a relation between individual variation in serum thyroid hormone concentrations and specific aspects of personality functioning. TSH was the hormone with the strongest relationship with personality scores. TSH serum concentrations were positively associated with Persistence (in all types of analyses) and Self-Directedness (in partial correlations and regression analyses), and negatively associated with Harm Avoidance (by partial correlations and graphical analyses of TCI-140) and Neuroticism on the NEO-FFI (in all types of analysis). These results corroborate those obtained by Forman-Hoffman and Philibert (2006), Williams et al. (2009), and Medici et al. (2014) who showed that serum TSH is inversely associated with anxiety and depression (Neuroticism). Furthermore, our data are in full agreement with those of Frey et al. (2007) who revealed a negative correlation between Neuroticism and serum TSH in a sample of 121 healthy participants using the same psychometrical tool as the present article (NEO-FFI). These relationships suggest that higher TSH is associated with a more adaptive personality profile: that is, it facilitates optimal coping with daily problems, and promotes an adequate sense of self-efficacy and satisfaction with one’s life and one’s relations with others. According to Cloninger’s psychobiological model of temperament and character (Cloninger and Svrakic 1997; Svrakic et al., 2002; Svrakic and Cloninger, 2010), the adaptive personality profile is basically characterized by high scores on Self-Directedness and Cooperativeness dimensions, and low scores on Harm Avoidance (Spittlehouse et al., 2014). These findings may appear to challenge the common belief that low thyroid function is associated with affective disorders such as depression. In our study, however, we tested a normal population with no evidence of mental or thyroid illness. Moreover, the relationships observed between TSH level and personality traits were independent of the levels of peripheral thyroid hormones (FT3, FT4, TT4). Therefore, the correlations between TSH and personality obtained here are more likely to be associated with central than with peripheral thyroid function. One of the strengths of the present study is its use of the TCI-140, which provides much more detailed psychobiological information on the relationship between depressive symptoms and thyroid hormones. For instance, the present data suggest that the relationship between depression and low Self-Directedness, low Persistence, or low Cooperativeness observed in a number of studies (Matsudaira and Kitamura, 2006; Spittlehouse et al., 2014) may be partially mediated by low TSH. This study also showed a negative association between Reward Dependence and serum FT4 (in the partial correlation, extreme groups and graphical analyses) which was independent of age and TSH. Low Reward Dependence is known to be associated with cluster A personality disorders of DSM-5 (American Psychiatric Association, 2013) such as Schizoid and Schizotypal Personality Disorders (Ohi et al., 2012). Interestingly, several studies (Roca et al., 1990; Sim et al., 2002; Steiblienė et al., 2012; Akiibinu et al., 2012) have reported elevated T4 in these patients. Finally, by partial correlation, regression and graphical analyses we found a negative association between total T4 and Cooperativeness (by partial correlation, regression and graphical analysis), a personality dimension which has been shown to be decreased in almost all mental disorders and is indicative of problems in the interpersonal domain. Graphical analysis also revealed a relation between high TT4 and lower scores for Persistence and for all character dimensions (SelfDirectedness, Cooperativeness and Self-Transcendence), thus representing a maladaptive personality profile (Spittlehouse et al., 2014).
5. Conclusions The present study shows that the relationship between personality and thyroid hormones may be more complex than previously supposed. The most significant findings are the positive association of serum TSH with Persistence and Self-Directedness (maturity and self-regulation) and its negative association with Harm Avoidance and NEO-FFI Neuroticism (emotional dysregulation). Thus, our study supports the hypothesis that reduced serum TSH may represent the endophenotype associated with a maladaptive personality profile. Increased FT4 was associated with lower Reward-Dependence and increased TT4 with lower Cooperativeness (both suggesting a deficit in social attachment). Studies with larger sample sizes and facet level analyses may broaden our understanding of the relationship between thyroid hormones and personality. 6. Limitations The unavailability of Cronbach's alphas for NEO-FFI was one of the limitations of the study. Other limitations were the lack of information on TSH diurnal rhythm and the inability to fully control all the factors that might influence TSH level (in particular, individual sleep patterns). Finally, because of the sample size used we were unable to study the relationship between specific personality profiles (resulting of the combination of main dimensions) and the dependent variables studied. Contributors AP, NT, AF & JR designed the study, wrote the protocol, and conducted literature searches. AP & NT conducted the study. AP & AF analysed the data. AP, NT, AF, AM & JR interpreted the data. AP & AF wrote the first draft of the manuscript. All authors contributed to the manuscript and have approved the final version. Conflict of interest The authors have no conflict of interest to declare regarding this manuscript. References Akiibinu, M.O., Ogundahunsi, O.A., Ogunyemi, E.O., 2012. Inter-relationship of plasma markers of oxidative stress and thyroid hormones in schizophrenics. BMC Res. Notes
186
Psychoneuroendocrinology 87 (2018) 181–187
A. Piskunov et al.
Kudryashev, A.F. (Ed.), 1992. Luchshie psichologicheskiye testi dlya profotbora i proforientacii [Best Psychological Tests for Vocational Assessment and Orientation]. Petrozavodsk University, Petroza-vodsk, Russia. Matsudaira, T., Kitamura, T., 2006. Personality traits as risk factors of depression and anxiety among Japanese students. J. Clin. Psychol. 62 (1), 97–109. Medici, M., Direk, N., Visser, W.E., Korevaar, T.I., Hofman, A., Visser, T.J., Tiemeier, H., Peeters, R.P., 2014. Thyroid function within the normal range and the risk of depression: a population-based cohort study. J. Clin. Endocrinol. Metab. 99 (4), 1213–1219. http://dx.doi.org/10.1210/jc.2013-3589. Morte, B., Bernal, J., 2014. Thyroid hormone action: astrocyte-neuron communication. Front. Endocrinol. (Lausanne) 5, 82. http://dx.doi.org/10.3389/fendo.2014.00082. Netter, P., 2004. Personality and hormones. In: Stelmack, R.M. (Ed.), On the Psychobiology of Personality: Essays in Honor of Marvin Zuckerman. Elsevier, Ltd., Oxford, pp. 353–377. Ohi, K., Hashimoto, R., Yasuda, Y., Fukumoto, M., Yamamori, H., Iwase, M., Kazui, H., Takeda, M., 2012. Personality traits and schizophrenia: evidence from a case-control study and meta-analysis. Psychiatry Res. 198 (1), 7–11. http://dx.doi.org/10.1016/j. psychres.2011.12.018. Peteranderl, C., Antonijevic, I.A., Steiger, A., Murck, H., Held, K., Frieboes, R.M., Uhr, M., Schaaf, L., 2002. Nocturnal secretion of TSH and ACTH in male patients with depression and healthy controls. J. Psychiatr. Res. 36, 189–196. Préau, L., Fini, J.B., Morvan-Dubois, G., Demeneix, B., 2014. Thyroid hormone signaling during early neurogenesis and its significance as a vulnerable window for endocrine disruption. Biochim. Biophys. 1874–9399 (14), 00167–00169. http://dx.doi.org/10. 1016/j.bbagrm.2014.06.015. Preacher, K.J., Rucker, D.D., MacCallum, R.C., Nicewander, W.A., 2005. Use of the extreme groups approach: a critical reexamination and new recommendations. Psychol. Methods 10, 178–192. Roca, R.P., Blackman, M.R., Ackerley, M.B., Harman, S.M., Gregerman, R.I., 1990. Thyroid hormone elevations during acute psychiatric illness: relationship to severity and distinction from hyperthyroidism. Endocr. Res. 16 (4), 415–447. Sánchez-Huerta, K., García-Martínez, Y., Vergara, P., Segovia, J., Pacheco-Rosado, J., 2016. Thyroid hormones are essential to preserve non-proliferative cells of adult neurogenesis of the dentate gyrus. Mol. Cell. Neurosci. 76, 1–10. http://dx.doi.org/ 10.1016/j.mcn.2016.08.001. Schroeder, A.C., Privalsky, M.L., 2014. Thyroid hormones, t3 and t4, in the brain. Front. Endocrinol. (Lausanne) 5, 40. http://dx.doi.org/10.3389/fendo.2014.00040. Sierra, P., Cámara, R., Tobella, H., Livianos, L., 2014. What is the real significance and management of major thyroid disorders in bipolar patients? Rev. Psiquiatr. Salud Ment. 7 (2), 88–95. http://dx.doi.org/10.1016/j.rpsm.2013.07.005. Sim, K., Chong, S.A., Chan, Y.H., Lum, W.M., 2002. Thyroid dysfunction in chronic schizophrenia within a state psychiatric hospital. Ann. Acad. Med. Singapore 31 (5), 641–644. Sinai, C., Hirvikoski, T., Dencker, E., Nordströmm, A.L., Linder, J., Nordströmm, P., Jokinen, J., 2009. Thyroid hormones and personality traits in attempted suicide. Psychoneuroendocrinology 34, 1526–1532. http://dx.doi.org/10.1016/j.psyneuen. 2009.05.009. Spittlehouse, J.K., Vierck, E., Pearson, J.F., Joyce, P.R., 2014. Temperament and character as determinants of well-being. Compr. Psychiatry 55 (7), 1679–1687. http://dx. doi.org/10.1016/j.comppsych.2014.06.011. Steiblienė, V., Mickuvienė, N., Prange Jr, A.J., Bunevičius, R., 2012. Concentrations of thyroid axis hormones in psychotic patients on hospital admission: the effects of prior drug use. Medicina (Kaunas) 48 (5), 229–234. Svrakic, D.M., Cloninger, R.C., 2010. Epigenetic perspective on behaviour development, personality, and personality disorders. Psychiatr. Danub. 22 (2), 153–166. Svrakic, D.M., Draganic, S., Hill, K., Bayon, C., Przybeck, T.R., Cloninger, R.C., 2002. Temperament, character, and personality disorders: etiologic diagnostic, treatment issues. Acta Psychiatr. Scand. 106, 189–195. Williams, M.D., Harris, R., Dayan, C.M., Evans, J., Gallacher, J., Ben-Shlomo, Y., 2009. Thyroid function and the natural history of depression: findings from the Caerphilly Prospective Study (CaPS) and a meta-analysis. Clin. Endocrinol. (Oxf.) 70 (3), 484–492. http://dx.doi.org/10.1111/j.1365-2265.2008.03352.x. Ylikahri, R.H., Huttunen, M.O., Härkönen, M., 1980. Hormonal changes during alcohol intoxication and withdrawal. Pharmacol. Biochem. Behav. 13 (Suppl 1), 131–137. Zuckerman, M., 1997. The psychobiological basis of personality. In: Nyborg, H. (Ed.), The Scientific Study of Human Nature. Oxford, Pergamon, pp. 3–16.
5, 169. http://dx.doi.org/10.1186/17560500-5-169. American Psychiatric Association, 2013. Diagnostic and Statistical Manual of Mental Disorders, 5th ed. (Washington, D.C.). Arqué, J.M., Segura, R., Torrubia, R., 1987. Correlation of thyroxine and thyroid-stimulation hormone with personality measurements: a study in psychosomatic patients and healthy subjects. Neuropsychobiology 18 (3), 127–133. Balada, F., Torrubia, R., Arqué, J.M., 1992. Thyroid hormone correlates of sensation seeking and anxiety in healthy human females. Neuropsychobiology 25 (4), 208–213. Banki, C.M., Bissette, G., Arato, M., Nemeroff, C.B., 1988. Elevation of immunoreactive CSF TRH in depressed patients. Am. J. Psychiatry 145, 1526–1531. Baturina, N.A., Aidman, B.V., 2010. Metodoki psikhologicheskoy diagnostiki I izmereniya: Ezhegodnik proffesionalnyh recenziy i obzorov. SUSU, Chelyabinsk, pp. 49–55. Bauer, M., Glenn, T., Pilhatsch, M., Pfennig, A., Whybrow, P.C., 2014. Gender differences in thyroid system function: relevance to bipolar disorder and its treatment. Bipolar. Disord. 16 (1), 58–71. http://dx.doi.org/10.1111/bdi.12150. Bremner, A.P., Feddema, P., Leedman, P.J., Brown, S.J., Beilby, J.P., Lim, E.M., Wilson, S.G., O'Leary, P.C., Walsh, J.P., 2012. Age-related changes in thyroid function: a longitudinal study of a community-based cohort. J. Clin. Endocrinol. Metab. 97, 1554–1562. Cloninger, C.R., Svrakic, D.M., 1997. Integrative psychobiological approach to psychiatric assessment and treatment. Psychiatry 60 (2), 120–141. Cloninger, C.R., 1999. The Temperament and Character Inventory. Center for Psychobiology of Personality, Washington University, Revised. St. Louis, MO (Available from C. R. Cloninger, Washington University School of Medicine, Department of Psychiatry, PO Box 8134, St. Louis, MO, 63110). Cloninger, C.R., 2000. Biology of personality dimensions. Curr. Opin. Psychiatry 13, 611–616. Costa, P.T., McCrae, R.R., 1992. Revised NEO Personality Inventory (NEO-PI-R) and NEO Five-Factor Inventory (NEO-FFI) Manual. Psychological Assessment Resources, Odessa, FL. Forman-Hoffman, V., Philibert, R.A., 2006. Lower TSH and higher T4 levels are associated with current depressive syndrome in young adults. Acta Psychiatr. Scand. 114, 132–139. Fox, J., 2000. Nonparametric Simple Regression: Smoothing Scatterplots. Thousand Oaks, CA. Frey, A., Lampert, A., Dietz, K., Striebich, S., Locher, C., Fedorenko, O., Möhle, R., Gallinat, J., Lang, F., Lang, U.E., 2007. Neuropsychobiology 56 (2–3), 123–126. http://dx.doi.org/10.1159/000112954. Frye, M.A., Gary, K.A., Marangell, L.B., George, M.S., Callahan, A.M., Little, J.T., Huggins, T., Corá Locatelli, G., Osuch, E.A., Winokur, A., Post, R.M., 1999. CSF thyrotropin releasing hormone gender difference: implications for neurobiology and treatment of depression. J. Neuropsych. Clin. Neurosci. 11, 349–353. Gitlin, M., Altshuler, L.L., Frye, M.A., Suri, R., Huynh, E.L., Fairbanks, L., Bauer, M., Korenman, S., 2004. Peripheral thyroid hormones and response to selective serotonin reuptake inhibitors. J. Psychiatry Neurosci. 29 (5), 383–386. Gray, J.A., 1993. The Psychology of Fear and Stress. Cambridge University Press, London. Hadlow, N.C., Rothacker, K.M., Wardrop, R., Brown, S.J., Lim, E.M., Walsh, J.P., 2013. The relationship between TSH and free T4 in a large population is complex and nonlinear and differs by age and sex. J. Clin. Endocrinol. Metab. 98, 2936–2943. Heal, D.J., Smith, S.L., 1988. The effects of acute and repeated administration of T3 to mice on 5-HT1 and 5-HT2 function in the brain and its influence on the actions of repeated electroconvulsive shock. Neuropharmacology 27, 1239–1248. Hein, M.D., Jackson, I.M., 1990. Review: thyroid function in psychiatric illness. Gen. Hosp. Psychiatry 12, 232–244. Jackson, I.M., 1998. The thyroid axis and depression. Thyroid 8, 951–956. Jacobsen, J.P., Medvedev, I.O., Caron, M.G., 2012. The 5-HT deficiency theory of depression: perspectives from a naturalistic 5-HT deficiency model, the tryptophan hydroxylase 2Arg439His knockin mouse. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 367 (1601), 2444–2459. http://dx.doi.org/10.1098/rstb.2012.0109. Johnson, A.M., Vernon, P.A., Feiler, A.R., 2008. Behavioural Genetic Studies of Personality: An Introduction and Review of the Results of 5 Years of Research. Handbook of Personality Theory and Assessment. Sage, London, pp. 145–173. Kessler, L., Nedeltceva, A., Imperial, J., Penev, P.D., 2010. Changes in serum TSH and free T4 during human sleep restriction. Sleep 33, 1115–1118. Kirkegaard, C., Faber, J., 1998. The role of thyroid hormones in depression. Eur. J. Endocrinol. 138, 1–9.
187