- Email: [email protected]
The influence of personality factors on paroxetine response time in patients with major depression
Journal of Affective Disorders 135 (2011) 321–325
Contents lists available at ScienceDirect
Journal of Affective Disorders j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j a d
Preliminary Communication
The influence of personality factors on paroxetine response time in patients with major depression Ayako Kaneda, Norio Yasui-Furukori ⁎, Taku Nakagami, Yasushi Sato, Sunao Kaneko Department of Neuropsychiatry, Graduate School of Medicine, Hirosaki University, Hirosaki 036–8562, Japan
a r t i c l e
i n f o
Article history: Received 21 June 2011 Accepted 29 June 2011 Available online 23 July 2011 Keywords: Personality factors Paroxetine response time Major depression Early response
a b s t r a c t Background: Determining the factors that predict antidepressant response and offering suitable treatments to people who suffer from major depressive disorder (MDD) is important. We investigated the personality factors that influence paroxetine treatment response by dividing antidepressant responders into two groups. Methods: We treated 93 patients with MDD using 40 mg/day of paroxetine for six weeks. We used the Cloninger's Temperament and Character Inventory (TCI) to evaluate each participant's personality before the treatment. Of the 93 patients, 75 completed the protocol. The Montgomery Asberg Depression Rating Scale (MADRS) was used to evaluate depressive symptoms before the treatment and at one-, two-, four-, and six-week intervals. We divided the patients into four groups: later responders (LRs), early responders (ERs), nonresponders (NRs), and dropouts (DOs). Results: Compared with 91 normal control participants, patients with MDD had less novelty seeking and self-directedness and greater harm avoidance. ERs showed less harm avoidance and more self-directedness than the other groups. LRs’ TCI scores did not differ from the other groups. Conclusions: These results suggest that ERs’ personality characteristics are different from those of other patients with MDD and that evaluating patients’ personality using the TCI at baseline may predict their antidepressant response. Limitations: Our sample of patients with MDD was small. Some of the patients with severe MDD had difficulty completing the TCI. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Major depressive disorder (MDD) is one of the most prevalent psychiatric conditions. Selective serotonin re-uptake inhibitors (SSRIs) are frequently used to treat MDD. Although SSRIs are highly effective (Steffens et al., 1997), SSRI response is difficult to predict because response patterns vary between patients. Therefore, determining the factors that predict antidepressant response and offering suitable treatments for people who suffer from MDD is important. Researchers have been investigating the association between personality and antidepressant response for more than two decades. Cloninger (1987) devised a biosocial model of ⁎ Corresponding author. Tel.: + 81 172 39 5066; fax: + 81 172 39 5067. E-mail address: [email protected] (N. Yasui-Furukori). 0165-0327/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.jad.2011.06.050
personality based on 3 independent temperament dimensions: novelty seeking (NS), harm avoidance (HA), and reward dependence (RD). Cloninger postulated that these dimensions are genetically determined and related to the specific activity of neurotransmitter systems such as dopamine, serotonin, or norepinephrine. The Tridimensional Personality Questionnaire (TPQ) was developed to measure NS, HA, and RD. Subsequently, the Temperament and Character Inventory (TCI), a 240-item self-rating questionnaire, was developed to measure 7 dimensions of personality (Cloninger et al., 1993). The TCI consists of four temperament dimensions and the three character dimensions, and it is based on a synthesis of social and cognitive research with personality research from humanistic and transpersonal psychology. Persistence (P), originally thought to be a component of RD, emerged as a distinct fourth temperament factor based on the factor structure of the TPQ.
322
A. Kaneda et al. / Journal of Affective Disorders 135 (2011) 321–325
Furthermore, the TCI added three character dimensions that are susceptible to self-concept change through social learning: selfdirectedness (SD), cooperativeness (C), and self-transcendence (ST). Kampman and Poutanen (in press) reported that an indisputable association existed between TCI scores (particularly HA) and treatment response of patients with MDD in the studies reviewed. Furthermore, several studies have tried to predict clinical response to antidepressants using TCI-measured personality factors (e.g., Hruby et al., 2010; Tome et al., 1997). Tome et al. (1997) showed that a combination of low HA and high RD on the short version of TCI (TCI-125) at baseline predicted better treatment responses in 48 patients with MDD after 6 weeks of a double-blind treatment with paroxetine, an SSRI. Hruby et al. (2010) showed similar results using the TCI with SSRIs or serotonin and noradrenaline reuptake inhibitors (SNRIs). Using the TCI, Sato et al. (1999) showed that high C and SD scores at baseline predicted better responses to a heterocyclic agent, maplotiline, after 8 weeks. Abrams et al. (2004) reported that HA scores at baseline reliably predicted response to sertraline treatment in patients with MDD and dysthymic disorder. However, these previous studies defined “responders” as patients who reacted positively to antidepressants after 6–8 weeks of treatment regardless of their response time. We sought to determine whether there are personality differences between patients who show immediate improvements to antidepressants and patients who show eventual improvement (e.g., five weeks later). Little research has divided responders into two groups and compared results. Therefore, this study investigated the influence of personality factors in patients with MDD on their paroxetine response, especially with regard to their antidepressant response time. 2. Methods 2.1. Participants Between December 2004 and September 2008, male and female 18- to 70-year-old patients with an MDD DSM-IV diagnosis from the Hospital of Hirosaki University School of Medicine, Hirosaki-Aiseikai Hospital, Kuroishi-Akebono Hospital, and Ohdate-city Hospital were identified as eligible for participation in this study. In addition, we required that patients score more than 20 points on the Montgomery Asberg Depression Rating Scale (MADRS) (Montgomery and Asberg, 1979). The MADRS consists of 10 items that are scored from 0 to 6. We
excluded patients who had taken medications, including psychotropic agents, at least one month before the start of the study as well as those with clinically significant abnormal laboratory or electrocardiography findings, a history of mental illness other than depression (i.e., mania, schizophrenia, epilepsy, alcohol or drug abuse) or clinically significant organic or neurological disease. Out of 93 patients, 18 did not complete the questionnaire. We also recruited 91 19- to 64-year-old healthy people as a control group. Control group participants had no psychiatric history or current psychiatric complaints and scored fewer than 15 points on The Center for Epidemiologic Studies Depression Scale (CES-D) (Radloff, 1977). The majority of healthy controls were medical staff such as nurses, nursing assistants, dieticians, occupational therapists, and psychiatric social workers of Hirosaki-Aiseikai Hospital, Kuroishi-Akebono Hospital, and Huyoukai Hospital. Table 1 shows the characteristics of the two groups. There was no significant difference in age and sex between patients with MDD and the control group. We conducted the present study after obtaining approval from the Ethics Committee of the Hirosaki University School of Medicine. Participants provided written informed consent after receiving a full description of the study. 2.2. Procedure We administered a dose of 20 mg/day of paroxetine (Paxil, GlaxoSmithKline, Tokyo, Japan) to patients at 20.00 h during the first week; thereafter, we increased the dose to 40 mg/day. We did not increase the dosage when we observed mild side effects (point 1 in the Udvalg for Kliniske Unders gelser Side Effects Rating Scale (UKU) (Chiba and Takahashi, 2005) or (Lingjaerde et al., 1987); furthermore, we decreased the paroxetine dosage when we observed moderate side effects (point 2 in the UKU) and discontinued the drug in cases of severe side effects (point 3 in the UKU). We also provided diazepam (2–5 mg/day, n = 19) for anxiety, brotizolam (0.25 mg/day, n = 20 and 0.5 mg/day, n = 17) for insomnia, [ME46] and sennoside (12–48 mg/day, n = 12) for constipation. We took blood samples from participants after 1, 2, and 6 weeks of paroxetine treatment. We used the MADRS to evaluate clinical symptoms and used the UKU rating scale to evaluate side effects at 1, 2, 4, and 6 weeks. All participants completed the Japanese version of the TCI personality evaluation (Kijima et al., 1996) before the treatment. The Japanese version of the MADRS evaluated the
Table 1 Demographic and MADRS scores of healthy controls and participants with major depressive disorders.
Age Age range Sex male / female Age of onset (years) Duration of illness (months) Paroxetine (mg/day) MADRS 0 weeks MADRS 6 weeks Data show mean ± SD. a t test was used. b chi square was used.
Healthy (n = 91)
Major Depressive Disorders (n = 75)
value
df
p
43.3 ± 11.3 18 - 69 29 / 62 -
45.7 ± 14.3 18 - 70 28 / 47 44.6 ± 14.2 14.5 ± 25.6 34.8 ± 9.6 39.8 ± 8.7 12.6 ± 12.3
t = -1.17
138.861
0.244a
Χ2 = 0.545
1
0.461b
A. Kaneda et al. / Journal of Affective Disorders 135 (2011) 321–325
depressive symptoms of all patients at baseline and 1, 2, 4, and 6 weeks into the treatment. We divided patients with MDD into four subgroups according to their paroxetine treatment response. We defined patients with less than a 50% improvement rate in MADRS scores at week 6 as nonresponders (NRs). We defined patients with a greater than 50% improvement rate and an MADRS score less than 10 at week 6 as later responders (LRs). We defined patients with a greater than 50% improvement rate in MADRS scores at week 1 as early responders (ERs). Finally, we defined patients who dropped out of this study as dropouts (DOs). Table 2 shows the characteristics of these groups. There were no significant differences between these groups; however, ERs’ MADRS scores at baseline tended to be lower than those of other groups. 2.3. Statistical analyses We used t-tests to analyze the differences between patients and healthy participants regarding participant characteristics and TCI scores. An ANOVA, followed by Tukey's post-hoc test, was used to analyze differences between the groups of patients with MDD in terms of participant characteristics and TCI scores. We considered p-values less than 0.05 to be significant. We performed all analyses using SPSS 17.0 J for Windows (SPSS Japan Inc., Tokyo, Japan). 3. Results 3.1. Comparisons between patients with MDD and controls Table 3 compares with the MDD group and the control group on the seven dimensions of the TCI scores. MDD patients’ NS (t= 4.61; degrees of freedom (df)= 164, p b 0.001) and SD (t= 6.31; df = 164, p b 0.001) scores were significantly lower than the controls. MDD patients’ HA scores (t = −19.68; df = 161.58, p b 0.001) were significantly higher than the controls. There were no differences between MDD patients and controls with regard to RD (t= 1.27; df = 164, p = 0.21), P (t = 0.564; df = 164, p = 0.573), C (t = 1.81; df = 164, p = 0.073), and ST (t= −0.62; df = 164, p = 0.531) scores. 3.2. Within group comparisons Table 4 shows the seven dimensions of the TCI for the four patient groups. There were no significant differences between
323
these groups with regard to NS (F = 1.474; df = 3,71; p = 0.229), RD (F = 1.415; df = 3,71; p = 0.246), P (F = 0.086; df = 3,71; p = 0.968), C (F = 2.522; df = 3,71; p = 0.065), or ST (F = 1.907; df = 3,71; p = 0.136) scores; however, we did find differences in HA (F = 6.884; df = 3,71; p b 0.001) and SD (F = 3.049; df = 3,71; p b 0.05) scores. Posthoc analyses revealed that ERs’ HA scores were significantly lower than NRs’ scores (p = 0.001) and LRs’ scores (p b 0.001). In addition, ERs’ SD scores were significantly higher than those of NRs (p b 0.05). 4. Discussion This study found that patients with MDD had significantly lower NS and SD TCI scores and significantly higher HA TCI scores than healthy control group participants. ERs had significantly different HA and SD scores when compared with other groups, whereas there were no significant differences between LRs and other groups in any TCI scores. These findings suggest that patients with MDD have different personality characteristics than healthy controls and that ERs have different personality characteristics than other patients with MDD. In addition, ERs and healthy controls had roughly similar HA and SD scores. This study is the first to indicate that response time to antidepressants is associated with personality factors. The healthy controls in the present study were Japanese medical staff. Hansenne et al. (1999) and Celikel et al. (2009) recruited medical staff members as healthy controls. Hansenne et al. (1999) reported that depressed patients exhibited higher HA and lower SD and C scores than healthy controls, and Celikel et al. (2009) reported that depressed patients exhibited higher HA scores and lower SD scores than healthy controls. These results are similar to those presented herein. It is possible that the temperament of medical staff members may be different than that of the general population; however, the results of the present study are similar to those reported by Kampman and Poutanen (in press). They suggested that HA is indisputably associated with risk-taking and treatment response in depression and reported that high HA was associated with both current depressive symptoms and depressive trait in a meta-analysis of twelve studies that focused on the occurrence of current or future depressive symptoms and temperament in a nonclinical adult sample. We compared personality factors among the four groups of patients with MDD (i.e., NRs, LRs, ERs, and DOs). ERs (i.e.,
Table 2 Demographic and MADRS score of the five groups of major depressive disorders subjects grouped by treatment response.
Age Sex male / female Age of onset (years) Duration of illness (months) Paroxetine (mg/day) MADRS 0 weeks MADRS 6 weeks Data show mean ± SD. a Compared by ANOVA. b chi square was used.
Non-Responders (n = 17)
Late Responders (n = 32)
Early Responders (n = 13)
Drop Outs (n = 13)
value
df
p
43.5 ± 15.0 7 / 10 42.5 ± 14.5 13.3 ± 14.7 34.1 ± 9.1 39.1 ± 9.7 29.1 ± 9.6
48.8 ± 13.5 9 / 23 48.2 ± 14.2 11.6 ± 17.9 32.1 ± 12.1 42.0 ± 8.6 6.35 ± 5.5
45.1 ± 12.4 8/5 43.6 ± 10.8 21.0 ± 42.3 33.8 ± 9.2 34.7 ± 7.3 5.2 ± 4.4
41.5 ± 15.6 4/9 40.4 ± 15.0 16.2 ± 27.5 40.8 ± 5.2 -
F = 0.989 x2 = 4.762 F = 1.131 F = 0.496
3,71 3 3,71 3,71
0.403a 0.190b 0.343a 0.686a
F = 2.332
3.67
0.082a
324
A. Kaneda et al. / Journal of Affective Disorders 135 (2011) 321–325
Table 3 Comparison of the TCI scores in healthy and major depressive disorder participants.
Novelty Seeking Harm Avoidance Reward Dependence Persistence Self Directedness Cooperativeness Self Transcendence
Healthy
Major Depressive Disorders
t
df
pa
20.6 ± 4.4 19.1 ± 5.6 14.7 ± 3.3 4.0 ± 1.6 27.3 ± 6.1 28.0 ± 4.2 10.0 ± 4.4
17.4 ± 4.5 27.1 ± 5.4 14.1 ± 3.2 3.9 ± 1.7 21.0 ± 6.3 26.8 ± 5.3 10.5 ± 4.7
4.61 -10.68 1.27 0.56 6.31 1.81 -0.62
164 161.58 164 164 164 164 164
pb0.001 pb0.001 0.21 0.573 pb0.001 0.073 0.531
Data shows mean ± SD. a t test was used.
more than 50% improvement at Week 1) showed significantly lower HA scores than other groups and significantly higher SD scores compared with NRs. The personality characteristics of ERs were similar to those of a “responder” in previous reports (Hruby et al., 2010; Tome et al., 1997). However, the personality characteristics of LRs (i.e., more than 50% improvement and a score of less than 10 on the MADRS at week 6) differed from those of a “responder” in previous reports. This result was also inconsistent with the report of Kampman and Poutanen (in press); the HA temperament dimension is obviously associated with treatment response in major depression. The difference between the Kampman and Poutanen's study and our study may be explained by dividing “responder” into two groups by response time to antidepressants. Accordingly, we suppose that not only LRs in this study but also many ERs might be included in the “responder” group in previous studies. Furthermore, it is possible that the personality characteristics of “responders” in previous studies primarily reflected the characteristics of the ERs. Therefore, future study should focus on the response time to antidepressants and the personality factors of patients with major depression. The MADRS interviewer must follow a structured method when asking patients questions. Almost all of the questions include the phrase, “In the past week, have you…?” Therefore, the first-week questions are almost the same as the symptom and condition questions directed toward patients after they begin taking medication. Accordingly, ERs may not actually report antidepressant drug effects; rather, they may show a placebo effect because the clinical effects of antidepressant drugs do not occur until at least 2 to
3 weeks after starting treatment (National Collaborating Center for Mental Health, 2004). Farmer and Goldberg (2008) reported that HA and SD factors were not clearly distinguishable in a confirmatory and exploratory factor analysis of the English-language Cloninger's revised Temperament and Character Inventory (TCI-R) (Cloninger, 1999) and a shortened inventory proposed by Cloninger (1999), the TCI-140. This finding suggests that high HA and low SD are expected to co-occur and to change in the same manner when comparing the treatment response of depressed patients and healthy controls. Accordingly, we recommend that future practitioners provide an environmental control before initiating antidepressant treatment in MDD patients who present low HA and high SD scores at baseline. This study has several limitations. First, the sample size of patients with MDD was small. Therefore, future studies should recruit more patients. Second, the TCI may burden MDD patients. Patients with severe depressive symptoms have difficulty completing the TCI, which consists of 240 questions. In fact, 18 MDD patients in our sample could not complete the TCI. However, future studies could use the short version of the TCI, which consists of 125 questions (Cloninger et al., 1994). Furthermore, the severity of depressive symptoms tends to be milder in ERs than in other depressed patients; thus, URs should be able to complete the TCI. In conclusion, this study shows that the personality characteristics of ERs are different from those of other MDD patients, and the TCI, when used to evaluate personality at baseline, can predict patients’ response to antidepressants. However, additional studies with more participants are needed to confirm these preliminary findings.
Table 4 Demographic and MADRS score of the five groups of major depressive disorders subjects grouped by treatment response.
Novelty Seeking Harm Avoidance Reward Dependence Persistence Self Directedness Cooperativeness Self Transcendence
Non-Responders
Late Responders
Early Responders
Drop Outs
F
df
p
16.2 ± 4.8 28.4 ± 4.2 12.8 ± 3.3 3.9 ± 2.2 18.9 ± 6.0 24.6 ± 4.2 9.5 ± 4.9
17.3 ± 3.6 28.0 ± 3.0 14.1 ± 3.4 3.8 ± 1.6 20.4 ± 5.8 27.5 ± 3.2 10.7 ± 4.6
19.6 ± 4.0 23.0 ± 3.8 14.7 ± 3.6 3.7 ± 1.8 25.8 ± 7.9 28.5 ± 4.2 13.1 ± 5.9
16.9 ± 6.0 27.3 ± 3.6 15.2 ± 2.8 4.0 ± 1.2 19.9 ± 6.9 25.9 ± 6.5 8.8 ± 4.2
1.474 6.884 1.415 0.086 3.049 2.522 1.907
3,71 3,71 3,71 3,71 3,71 3,71 3,71
0.229 p b 0.001a, p b 0.05b 0.246 0.968 p b 0.05c 0.065 0.136
Data shows mean ± SD. a Compared Early responders to Non responders and Late Responders by ANOVA followed by tukey-test was used. b Compared Early responders to Drop outs by ANOVA followed by tukey-test was used. c Compared Early responders to Non responders by ANOVA followed by tukey-test was used.
A. Kaneda et al. / Journal of Affective Disorders 135 (2011) 321–325 Role of the funding source This study was supported by a research fund from Hirosaki University School of Medicine and a grant from the Hirosaki Research Institute for Neurosciences. Conflicts of interest None. Acknowledgment All authors declare that there is no conflict of interest that could be perceived to bias the findings reported in this work. This study was supported by a research fund from Hirosaki University School of Medicine and a grant from the Hirosaki Research Institute for Neurosciences.
References Abrams, K.Y., Yune, S.K., Kim, S.J., Jeon, H.J., Han, S.J., Hwang, J., Sung, Y.H., Lee, K.J., Lyoo, A.I.K., 2004. Trait and state aspects of harm avoidance and its implication for treatment in major depressive disorder, dysthymic disorder, and depressive personality disorder. Psychiatry Clin. Neurosci. 58, 240–248. Celikel, F.C., Kose, S., Cumurcu, B.E., Erkorkmaz, U., Sayar, K., Borckardt, J.J., Cloninger, C.R., 2009. Cloninger's temperament and character dimensions of personality in patients with major depressive disorder. Compr. Psychiatry 50, 556–561. Chiba, S., Takahashi, M., 2005. The Japanese version of UKU side effect rating scale and the translation process. Jpn. J. Clin. Psychopharmacol. 8, 1939–1962. Cloninger, C.R., 1987. A systematic method for clinical description and classification of personality variants. Arch. Gen. Psychiatry 44, 573–588. Cloninger, C.R., Svrakic, D.M., Przybeck, T.R., 1993. A psychobiological model of temperament and character. Arch. Gen. Psychiatry 50, 975–990. Cloninger, C.R., Przybeck, T.R., Svrakic, D.M., Welzel, R.D., 1994. The Temperament and Character Inventory (TCI): A Guide to its Development and Use. Center for Psychobiology of Personality, Washington University, St. Louis, MO. Cloninger, C.R., 1999. Temperament and Character Inventory-Revised. Washington University Center for Psychobiology of Personality, St. Louis, MO.
325
Farmer, R.F., Goldberg, L.R., 2008. A psychometric evaluation of the Revised Temperament and Character Inventory (TCI-R) and the TCI-140. Psychol. Assess. 20 (3), 281–291. Hansenne, M., Reggers, J., Pinto, E., Kjiri, K., Ajamier, A., Ansseau, M., 1999. Temperament and character inventory (TCI) and depression. J. Psychiatr. Res. 33, 31–36. Hruby, R., Nosalova, G., Hruba, S., 2010. Predictive significance of TCI-R for antidepressant treatment. Med. Sci. Monit. 16 (8), 383–388. Kampman, O., Poutanen, O., in press. Can onset and recovery in depression be predicted by temperament? A systematic review and meta-analysis. J. Affect. Disord. Kijima, N., Saito, R., Takeuchi, M., Yoshino, A., Ono, Y., Kato, M., Kitamura, T., 1996. Cloninger's seven-factor model of temperament and character and the Japanese version of theTemperament and Character Inventory. Arch. Psychiatr. Diag. Clin. Eval. 7, 379–399. Lingjaerde, O., Ahlfors, U.G., Bech, P., Dencker, S.J., Elgen, K., 1987. The UKU side effect rating scale. A new comprehensive rating scale for psychotropic drugs and a cross-sectional study of side effects in neuroleptic-treated patients. Acta Psychiatr. Scand. Suppl. 334, 1–100. Montgomery, S.A., Asberg, M., 1979. A new depression scale designed to be sensitive to change. Br. J. Psychiatry. 134, 382–389. National Collaborating Center for Mental Health, 2004. Depression: management of depression in primary and secondary care. Clinical. Guideline. National Institute for Clinical Excellence, London, England, p. 23. Radloff, L.S., 1977. The CES-D scale: a self-report depression scale for research in the general population. Appl. Psychol. Meas. 3, 385–401. Sato, T., Hirano, S., Narita, T., Kusunoki, K., Kato, J., Goto, M., Sakado, K., Uehara, T., 1999. Temperament and character inventory dimensions as a predictor of response to antidepressant treatment in major depression. J. Affect. Disord. 56, 153–161. Steffens, D.C., Krishnan, K.R., Helms, M., et al., 1997. Are SSRIs better than TCAs? Comparison of SSRIs and TCAs: A meta-analysis. Depress. Anxiety 6 (1), 10–18. Tome, M.B., Cloninger, C.R., Watoson, J.P., Isaac, M.T., 1997. Serotonergic autoreceptor blockade in the reduction of antidepressant latency: personality variables and response to paroxetine and pindrol. J. Affect. Disord. 44, 101–109.
Contents lists available at ScienceDirect
Journal of Affective Disorders j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j a d
Preliminary Communication
The influence of personality factors on paroxetine response time in patients with major depression Ayako Kaneda, Norio Yasui-Furukori ⁎, Taku Nakagami, Yasushi Sato, Sunao Kaneko Department of Neuropsychiatry, Graduate School of Medicine, Hirosaki University, Hirosaki 036–8562, Japan
a r t i c l e
i n f o
Article history: Received 21 June 2011 Accepted 29 June 2011 Available online 23 July 2011 Keywords: Personality factors Paroxetine response time Major depression Early response
a b s t r a c t Background: Determining the factors that predict antidepressant response and offering suitable treatments to people who suffer from major depressive disorder (MDD) is important. We investigated the personality factors that influence paroxetine treatment response by dividing antidepressant responders into two groups. Methods: We treated 93 patients with MDD using 40 mg/day of paroxetine for six weeks. We used the Cloninger's Temperament and Character Inventory (TCI) to evaluate each participant's personality before the treatment. Of the 93 patients, 75 completed the protocol. The Montgomery Asberg Depression Rating Scale (MADRS) was used to evaluate depressive symptoms before the treatment and at one-, two-, four-, and six-week intervals. We divided the patients into four groups: later responders (LRs), early responders (ERs), nonresponders (NRs), and dropouts (DOs). Results: Compared with 91 normal control participants, patients with MDD had less novelty seeking and self-directedness and greater harm avoidance. ERs showed less harm avoidance and more self-directedness than the other groups. LRs’ TCI scores did not differ from the other groups. Conclusions: These results suggest that ERs’ personality characteristics are different from those of other patients with MDD and that evaluating patients’ personality using the TCI at baseline may predict their antidepressant response. Limitations: Our sample of patients with MDD was small. Some of the patients with severe MDD had difficulty completing the TCI. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Major depressive disorder (MDD) is one of the most prevalent psychiatric conditions. Selective serotonin re-uptake inhibitors (SSRIs) are frequently used to treat MDD. Although SSRIs are highly effective (Steffens et al., 1997), SSRI response is difficult to predict because response patterns vary between patients. Therefore, determining the factors that predict antidepressant response and offering suitable treatments for people who suffer from MDD is important. Researchers have been investigating the association between personality and antidepressant response for more than two decades. Cloninger (1987) devised a biosocial model of ⁎ Corresponding author. Tel.: + 81 172 39 5066; fax: + 81 172 39 5067. E-mail address: [email protected] (N. Yasui-Furukori). 0165-0327/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.jad.2011.06.050
personality based on 3 independent temperament dimensions: novelty seeking (NS), harm avoidance (HA), and reward dependence (RD). Cloninger postulated that these dimensions are genetically determined and related to the specific activity of neurotransmitter systems such as dopamine, serotonin, or norepinephrine. The Tridimensional Personality Questionnaire (TPQ) was developed to measure NS, HA, and RD. Subsequently, the Temperament and Character Inventory (TCI), a 240-item self-rating questionnaire, was developed to measure 7 dimensions of personality (Cloninger et al., 1993). The TCI consists of four temperament dimensions and the three character dimensions, and it is based on a synthesis of social and cognitive research with personality research from humanistic and transpersonal psychology. Persistence (P), originally thought to be a component of RD, emerged as a distinct fourth temperament factor based on the factor structure of the TPQ.
322
A. Kaneda et al. / Journal of Affective Disorders 135 (2011) 321–325
Furthermore, the TCI added three character dimensions that are susceptible to self-concept change through social learning: selfdirectedness (SD), cooperativeness (C), and self-transcendence (ST). Kampman and Poutanen (in press) reported that an indisputable association existed between TCI scores (particularly HA) and treatment response of patients with MDD in the studies reviewed. Furthermore, several studies have tried to predict clinical response to antidepressants using TCI-measured personality factors (e.g., Hruby et al., 2010; Tome et al., 1997). Tome et al. (1997) showed that a combination of low HA and high RD on the short version of TCI (TCI-125) at baseline predicted better treatment responses in 48 patients with MDD after 6 weeks of a double-blind treatment with paroxetine, an SSRI. Hruby et al. (2010) showed similar results using the TCI with SSRIs or serotonin and noradrenaline reuptake inhibitors (SNRIs). Using the TCI, Sato et al. (1999) showed that high C and SD scores at baseline predicted better responses to a heterocyclic agent, maplotiline, after 8 weeks. Abrams et al. (2004) reported that HA scores at baseline reliably predicted response to sertraline treatment in patients with MDD and dysthymic disorder. However, these previous studies defined “responders” as patients who reacted positively to antidepressants after 6–8 weeks of treatment regardless of their response time. We sought to determine whether there are personality differences between patients who show immediate improvements to antidepressants and patients who show eventual improvement (e.g., five weeks later). Little research has divided responders into two groups and compared results. Therefore, this study investigated the influence of personality factors in patients with MDD on their paroxetine response, especially with regard to their antidepressant response time. 2. Methods 2.1. Participants Between December 2004 and September 2008, male and female 18- to 70-year-old patients with an MDD DSM-IV diagnosis from the Hospital of Hirosaki University School of Medicine, Hirosaki-Aiseikai Hospital, Kuroishi-Akebono Hospital, and Ohdate-city Hospital were identified as eligible for participation in this study. In addition, we required that patients score more than 20 points on the Montgomery Asberg Depression Rating Scale (MADRS) (Montgomery and Asberg, 1979). The MADRS consists of 10 items that are scored from 0 to 6. We
excluded patients who had taken medications, including psychotropic agents, at least one month before the start of the study as well as those with clinically significant abnormal laboratory or electrocardiography findings, a history of mental illness other than depression (i.e., mania, schizophrenia, epilepsy, alcohol or drug abuse) or clinically significant organic or neurological disease. Out of 93 patients, 18 did not complete the questionnaire. We also recruited 91 19- to 64-year-old healthy people as a control group. Control group participants had no psychiatric history or current psychiatric complaints and scored fewer than 15 points on The Center for Epidemiologic Studies Depression Scale (CES-D) (Radloff, 1977). The majority of healthy controls were medical staff such as nurses, nursing assistants, dieticians, occupational therapists, and psychiatric social workers of Hirosaki-Aiseikai Hospital, Kuroishi-Akebono Hospital, and Huyoukai Hospital. Table 1 shows the characteristics of the two groups. There was no significant difference in age and sex between patients with MDD and the control group. We conducted the present study after obtaining approval from the Ethics Committee of the Hirosaki University School of Medicine. Participants provided written informed consent after receiving a full description of the study. 2.2. Procedure We administered a dose of 20 mg/day of paroxetine (Paxil, GlaxoSmithKline, Tokyo, Japan) to patients at 20.00 h during the first week; thereafter, we increased the dose to 40 mg/day. We did not increase the dosage when we observed mild side effects (point 1 in the Udvalg for Kliniske Unders gelser Side Effects Rating Scale (UKU) (Chiba and Takahashi, 2005) or (Lingjaerde et al., 1987); furthermore, we decreased the paroxetine dosage when we observed moderate side effects (point 2 in the UKU) and discontinued the drug in cases of severe side effects (point 3 in the UKU). We also provided diazepam (2–5 mg/day, n = 19) for anxiety, brotizolam (0.25 mg/day, n = 20 and 0.5 mg/day, n = 17) for insomnia, [ME46] and sennoside (12–48 mg/day, n = 12) for constipation. We took blood samples from participants after 1, 2, and 6 weeks of paroxetine treatment. We used the MADRS to evaluate clinical symptoms and used the UKU rating scale to evaluate side effects at 1, 2, 4, and 6 weeks. All participants completed the Japanese version of the TCI personality evaluation (Kijima et al., 1996) before the treatment. The Japanese version of the MADRS evaluated the
Table 1 Demographic and MADRS scores of healthy controls and participants with major depressive disorders.
Age Age range Sex male / female Age of onset (years) Duration of illness (months) Paroxetine (mg/day) MADRS 0 weeks MADRS 6 weeks Data show mean ± SD. a t test was used. b chi square was used.
Healthy (n = 91)
Major Depressive Disorders (n = 75)
value
df
p
43.3 ± 11.3 18 - 69 29 / 62 -
45.7 ± 14.3 18 - 70 28 / 47 44.6 ± 14.2 14.5 ± 25.6 34.8 ± 9.6 39.8 ± 8.7 12.6 ± 12.3
t = -1.17
138.861
0.244a
Χ2 = 0.545
1
0.461b
A. Kaneda et al. / Journal of Affective Disorders 135 (2011) 321–325
depressive symptoms of all patients at baseline and 1, 2, 4, and 6 weeks into the treatment. We divided patients with MDD into four subgroups according to their paroxetine treatment response. We defined patients with less than a 50% improvement rate in MADRS scores at week 6 as nonresponders (NRs). We defined patients with a greater than 50% improvement rate and an MADRS score less than 10 at week 6 as later responders (LRs). We defined patients with a greater than 50% improvement rate in MADRS scores at week 1 as early responders (ERs). Finally, we defined patients who dropped out of this study as dropouts (DOs). Table 2 shows the characteristics of these groups. There were no significant differences between these groups; however, ERs’ MADRS scores at baseline tended to be lower than those of other groups. 2.3. Statistical analyses We used t-tests to analyze the differences between patients and healthy participants regarding participant characteristics and TCI scores. An ANOVA, followed by Tukey's post-hoc test, was used to analyze differences between the groups of patients with MDD in terms of participant characteristics and TCI scores. We considered p-values less than 0.05 to be significant. We performed all analyses using SPSS 17.0 J for Windows (SPSS Japan Inc., Tokyo, Japan). 3. Results 3.1. Comparisons between patients with MDD and controls Table 3 compares with the MDD group and the control group on the seven dimensions of the TCI scores. MDD patients’ NS (t= 4.61; degrees of freedom (df)= 164, p b 0.001) and SD (t= 6.31; df = 164, p b 0.001) scores were significantly lower than the controls. MDD patients’ HA scores (t = −19.68; df = 161.58, p b 0.001) were significantly higher than the controls. There were no differences between MDD patients and controls with regard to RD (t= 1.27; df = 164, p = 0.21), P (t = 0.564; df = 164, p = 0.573), C (t = 1.81; df = 164, p = 0.073), and ST (t= −0.62; df = 164, p = 0.531) scores. 3.2. Within group comparisons Table 4 shows the seven dimensions of the TCI for the four patient groups. There were no significant differences between
323
these groups with regard to NS (F = 1.474; df = 3,71; p = 0.229), RD (F = 1.415; df = 3,71; p = 0.246), P (F = 0.086; df = 3,71; p = 0.968), C (F = 2.522; df = 3,71; p = 0.065), or ST (F = 1.907; df = 3,71; p = 0.136) scores; however, we did find differences in HA (F = 6.884; df = 3,71; p b 0.001) and SD (F = 3.049; df = 3,71; p b 0.05) scores. Posthoc analyses revealed that ERs’ HA scores were significantly lower than NRs’ scores (p = 0.001) and LRs’ scores (p b 0.001). In addition, ERs’ SD scores were significantly higher than those of NRs (p b 0.05). 4. Discussion This study found that patients with MDD had significantly lower NS and SD TCI scores and significantly higher HA TCI scores than healthy control group participants. ERs had significantly different HA and SD scores when compared with other groups, whereas there were no significant differences between LRs and other groups in any TCI scores. These findings suggest that patients with MDD have different personality characteristics than healthy controls and that ERs have different personality characteristics than other patients with MDD. In addition, ERs and healthy controls had roughly similar HA and SD scores. This study is the first to indicate that response time to antidepressants is associated with personality factors. The healthy controls in the present study were Japanese medical staff. Hansenne et al. (1999) and Celikel et al. (2009) recruited medical staff members as healthy controls. Hansenne et al. (1999) reported that depressed patients exhibited higher HA and lower SD and C scores than healthy controls, and Celikel et al. (2009) reported that depressed patients exhibited higher HA scores and lower SD scores than healthy controls. These results are similar to those presented herein. It is possible that the temperament of medical staff members may be different than that of the general population; however, the results of the present study are similar to those reported by Kampman and Poutanen (in press). They suggested that HA is indisputably associated with risk-taking and treatment response in depression and reported that high HA was associated with both current depressive symptoms and depressive trait in a meta-analysis of twelve studies that focused on the occurrence of current or future depressive symptoms and temperament in a nonclinical adult sample. We compared personality factors among the four groups of patients with MDD (i.e., NRs, LRs, ERs, and DOs). ERs (i.e.,
Table 2 Demographic and MADRS score of the five groups of major depressive disorders subjects grouped by treatment response.
Age Sex male / female Age of onset (years) Duration of illness (months) Paroxetine (mg/day) MADRS 0 weeks MADRS 6 weeks Data show mean ± SD. a Compared by ANOVA. b chi square was used.
Non-Responders (n = 17)
Late Responders (n = 32)
Early Responders (n = 13)
Drop Outs (n = 13)
value
df
p
43.5 ± 15.0 7 / 10 42.5 ± 14.5 13.3 ± 14.7 34.1 ± 9.1 39.1 ± 9.7 29.1 ± 9.6
48.8 ± 13.5 9 / 23 48.2 ± 14.2 11.6 ± 17.9 32.1 ± 12.1 42.0 ± 8.6 6.35 ± 5.5
45.1 ± 12.4 8/5 43.6 ± 10.8 21.0 ± 42.3 33.8 ± 9.2 34.7 ± 7.3 5.2 ± 4.4
41.5 ± 15.6 4/9 40.4 ± 15.0 16.2 ± 27.5 40.8 ± 5.2 -
F = 0.989 x2 = 4.762 F = 1.131 F = 0.496
3,71 3 3,71 3,71
0.403a 0.190b 0.343a 0.686a
F = 2.332
3.67
0.082a
324
A. Kaneda et al. / Journal of Affective Disorders 135 (2011) 321–325
Table 3 Comparison of the TCI scores in healthy and major depressive disorder participants.
Novelty Seeking Harm Avoidance Reward Dependence Persistence Self Directedness Cooperativeness Self Transcendence
Healthy
Major Depressive Disorders
t
df
pa
20.6 ± 4.4 19.1 ± 5.6 14.7 ± 3.3 4.0 ± 1.6 27.3 ± 6.1 28.0 ± 4.2 10.0 ± 4.4
17.4 ± 4.5 27.1 ± 5.4 14.1 ± 3.2 3.9 ± 1.7 21.0 ± 6.3 26.8 ± 5.3 10.5 ± 4.7
4.61 -10.68 1.27 0.56 6.31 1.81 -0.62
164 161.58 164 164 164 164 164
pb0.001 pb0.001 0.21 0.573 pb0.001 0.073 0.531
Data shows mean ± SD. a t test was used.
more than 50% improvement at Week 1) showed significantly lower HA scores than other groups and significantly higher SD scores compared with NRs. The personality characteristics of ERs were similar to those of a “responder” in previous reports (Hruby et al., 2010; Tome et al., 1997). However, the personality characteristics of LRs (i.e., more than 50% improvement and a score of less than 10 on the MADRS at week 6) differed from those of a “responder” in previous reports. This result was also inconsistent with the report of Kampman and Poutanen (in press); the HA temperament dimension is obviously associated with treatment response in major depression. The difference between the Kampman and Poutanen's study and our study may be explained by dividing “responder” into two groups by response time to antidepressants. Accordingly, we suppose that not only LRs in this study but also many ERs might be included in the “responder” group in previous studies. Furthermore, it is possible that the personality characteristics of “responders” in previous studies primarily reflected the characteristics of the ERs. Therefore, future study should focus on the response time to antidepressants and the personality factors of patients with major depression. The MADRS interviewer must follow a structured method when asking patients questions. Almost all of the questions include the phrase, “In the past week, have you…?” Therefore, the first-week questions are almost the same as the symptom and condition questions directed toward patients after they begin taking medication. Accordingly, ERs may not actually report antidepressant drug effects; rather, they may show a placebo effect because the clinical effects of antidepressant drugs do not occur until at least 2 to
3 weeks after starting treatment (National Collaborating Center for Mental Health, 2004). Farmer and Goldberg (2008) reported that HA and SD factors were not clearly distinguishable in a confirmatory and exploratory factor analysis of the English-language Cloninger's revised Temperament and Character Inventory (TCI-R) (Cloninger, 1999) and a shortened inventory proposed by Cloninger (1999), the TCI-140. This finding suggests that high HA and low SD are expected to co-occur and to change in the same manner when comparing the treatment response of depressed patients and healthy controls. Accordingly, we recommend that future practitioners provide an environmental control before initiating antidepressant treatment in MDD patients who present low HA and high SD scores at baseline. This study has several limitations. First, the sample size of patients with MDD was small. Therefore, future studies should recruit more patients. Second, the TCI may burden MDD patients. Patients with severe depressive symptoms have difficulty completing the TCI, which consists of 240 questions. In fact, 18 MDD patients in our sample could not complete the TCI. However, future studies could use the short version of the TCI, which consists of 125 questions (Cloninger et al., 1994). Furthermore, the severity of depressive symptoms tends to be milder in ERs than in other depressed patients; thus, URs should be able to complete the TCI. In conclusion, this study shows that the personality characteristics of ERs are different from those of other MDD patients, and the TCI, when used to evaluate personality at baseline, can predict patients’ response to antidepressants. However, additional studies with more participants are needed to confirm these preliminary findings.
Table 4 Demographic and MADRS score of the five groups of major depressive disorders subjects grouped by treatment response.
Novelty Seeking Harm Avoidance Reward Dependence Persistence Self Directedness Cooperativeness Self Transcendence
Non-Responders
Late Responders
Early Responders
Drop Outs
F
df
p
16.2 ± 4.8 28.4 ± 4.2 12.8 ± 3.3 3.9 ± 2.2 18.9 ± 6.0 24.6 ± 4.2 9.5 ± 4.9
17.3 ± 3.6 28.0 ± 3.0 14.1 ± 3.4 3.8 ± 1.6 20.4 ± 5.8 27.5 ± 3.2 10.7 ± 4.6
19.6 ± 4.0 23.0 ± 3.8 14.7 ± 3.6 3.7 ± 1.8 25.8 ± 7.9 28.5 ± 4.2 13.1 ± 5.9
16.9 ± 6.0 27.3 ± 3.6 15.2 ± 2.8 4.0 ± 1.2 19.9 ± 6.9 25.9 ± 6.5 8.8 ± 4.2
1.474 6.884 1.415 0.086 3.049 2.522 1.907
3,71 3,71 3,71 3,71 3,71 3,71 3,71
0.229 p b 0.001a, p b 0.05b 0.246 0.968 p b 0.05c 0.065 0.136
Data shows mean ± SD. a Compared Early responders to Non responders and Late Responders by ANOVA followed by tukey-test was used. b Compared Early responders to Drop outs by ANOVA followed by tukey-test was used. c Compared Early responders to Non responders by ANOVA followed by tukey-test was used.
A. Kaneda et al. / Journal of Affective Disorders 135 (2011) 321–325 Role of the funding source This study was supported by a research fund from Hirosaki University School of Medicine and a grant from the Hirosaki Research Institute for Neurosciences. Conflicts of interest None. Acknowledgment All authors declare that there is no conflict of interest that could be perceived to bias the findings reported in this work. This study was supported by a research fund from Hirosaki University School of Medicine and a grant from the Hirosaki Research Institute for Neurosciences.
References Abrams, K.Y., Yune, S.K., Kim, S.J., Jeon, H.J., Han, S.J., Hwang, J., Sung, Y.H., Lee, K.J., Lyoo, A.I.K., 2004. Trait and state aspects of harm avoidance and its implication for treatment in major depressive disorder, dysthymic disorder, and depressive personality disorder. Psychiatry Clin. Neurosci. 58, 240–248. Celikel, F.C., Kose, S., Cumurcu, B.E., Erkorkmaz, U., Sayar, K., Borckardt, J.J., Cloninger, C.R., 2009. Cloninger's temperament and character dimensions of personality in patients with major depressive disorder. Compr. Psychiatry 50, 556–561. Chiba, S., Takahashi, M., 2005. The Japanese version of UKU side effect rating scale and the translation process. Jpn. J. Clin. Psychopharmacol. 8, 1939–1962. Cloninger, C.R., 1987. A systematic method for clinical description and classification of personality variants. Arch. Gen. Psychiatry 44, 573–588. Cloninger, C.R., Svrakic, D.M., Przybeck, T.R., 1993. A psychobiological model of temperament and character. Arch. Gen. Psychiatry 50, 975–990. Cloninger, C.R., Przybeck, T.R., Svrakic, D.M., Welzel, R.D., 1994. The Temperament and Character Inventory (TCI): A Guide to its Development and Use. Center for Psychobiology of Personality, Washington University, St. Louis, MO. Cloninger, C.R., 1999. Temperament and Character Inventory-Revised. Washington University Center for Psychobiology of Personality, St. Louis, MO.
325
Farmer, R.F., Goldberg, L.R., 2008. A psychometric evaluation of the Revised Temperament and Character Inventory (TCI-R) and the TCI-140. Psychol. Assess. 20 (3), 281–291. Hansenne, M., Reggers, J., Pinto, E., Kjiri, K., Ajamier, A., Ansseau, M., 1999. Temperament and character inventory (TCI) and depression. J. Psychiatr. Res. 33, 31–36. Hruby, R., Nosalova, G., Hruba, S., 2010. Predictive significance of TCI-R for antidepressant treatment. Med. Sci. Monit. 16 (8), 383–388. Kampman, O., Poutanen, O., in press. Can onset and recovery in depression be predicted by temperament? A systematic review and meta-analysis. J. Affect. Disord. Kijima, N., Saito, R., Takeuchi, M., Yoshino, A., Ono, Y., Kato, M., Kitamura, T., 1996. Cloninger's seven-factor model of temperament and character and the Japanese version of theTemperament and Character Inventory. Arch. Psychiatr. Diag. Clin. Eval. 7, 379–399. Lingjaerde, O., Ahlfors, U.G., Bech, P., Dencker, S.J., Elgen, K., 1987. The UKU side effect rating scale. A new comprehensive rating scale for psychotropic drugs and a cross-sectional study of side effects in neuroleptic-treated patients. Acta Psychiatr. Scand. Suppl. 334, 1–100. Montgomery, S.A., Asberg, M., 1979. A new depression scale designed to be sensitive to change. Br. J. Psychiatry. 134, 382–389. National Collaborating Center for Mental Health, 2004. Depression: management of depression in primary and secondary care. Clinical. Guideline. National Institute for Clinical Excellence, London, England, p. 23. Radloff, L.S., 1977. The CES-D scale: a self-report depression scale for research in the general population. Appl. Psychol. Meas. 3, 385–401. Sato, T., Hirano, S., Narita, T., Kusunoki, K., Kato, J., Goto, M., Sakado, K., Uehara, T., 1999. Temperament and character inventory dimensions as a predictor of response to antidepressant treatment in major depression. J. Affect. Disord. 56, 153–161. Steffens, D.C., Krishnan, K.R., Helms, M., et al., 1997. Are SSRIs better than TCAs? Comparison of SSRIs and TCAs: A meta-analysis. Depress. Anxiety 6 (1), 10–18. Tome, M.B., Cloninger, C.R., Watoson, J.P., Isaac, M.T., 1997. Serotonergic autoreceptor blockade in the reduction of antidepressant latency: personality variables and response to paroxetine and pindrol. J. Affect. Disord. 44, 101–109.