- Email: [email protected]
Affective temperaments and antidepressant response in the clinical management of mood disorders
Journal of Affective Disorders 155 (2014) 138–141
Contents lists available at ScienceDirect
Journal of Affective Disorders journal homepage: www.elsevier.com/locate/jad
Research report
Affective temperaments and antidepressant response in the clinical management of mood disorders Alexandre de Aguiar Ferreira a,b,c, Alina Gomide Vasconcelos a, Fernando Silva Neves a,d, Humberto Correa a,d,n a
Neuroscience Program, Federal University, Minas Gerais (UFMG), Belo Horizonte, MG, Brazil Raul Soares Institute – FHEMIG, Belo Horizonte, MG, Brazil c Faculty of Medical Sciences of Minas Gerais – FCMMG, Belo Horizonte, MG, Brazil d Department of Mental Health, Faculty of Medicine, Federal University, Minas Gerais, Av. Alfredo Balena, 190, Belo Horizonte, MG 30130-100, Brazil b
art ic l e i nf o
a b s t r a c t
Article history: Received 25 September 2013 Accepted 22 October 2013 Available online 28 October 2013
Background: The aim of this study was to investigate the presence of a relationship between affective temperament and antidepressant treatment response in mood disorder patients. Methods: The lifetime history of antidepressant response of 90 bipolar disorder patients and 88 major depressive disorder patients were retrospectively evaluated and then assigned to one of four subgroups: complete response (CR), partial response (PR), no response (NR), and antidepressant associated mania response (AAMR). Using TEMPS-Rio de Janeiro – the brief Brazilian version of TEMPS-A – we compared affective temperament subscale scores across these groups. Results: We observed a statistically significant relationship between depressive and anxious affective temperaments and no antidepressant response. In bipolar disorder patients, cyclothymic temperament (po 0.01) and hyperthymic temperament (po 0.05) were associated with antidepressant-associated mania. Hyperthymic temperament was associated with complete antidepressant responses in major depressive disorder patients. Limitations: The evaluation of antidepressant response was retrospective. Conclusions: Our data are consistent with the theory that affective temperament traits are factors that can influence the antidepressant response and the recovery from depressive episodes, but more longitudinal studies are needed to confirm this theory and our findings. & 2013 Published by Elsevier B.V.
Keywords: Affective temperament TEMPS-A TEMPS-Rio de Janeiro Antidepressant response Mood disorders
1. Introduction In the general approach to patients with mood disorders the characterization of affective temperament may be an important parameter. It may be useful in the differential diagnosis of mood disorders as studies have consistently shown that cyclothymic temperament is significantly more prominent in bipolar versus unipolar depressive patients (Mendlowicz et al., 2005a) and that cyclothymic traits may represent vulnerability markers found in clinically healthy relatives of bipolar disorder (BD) patients (Mendlowicz et al., 2005b; Aguiar Ferreira et al., 2013). So, when assessing depressive patients, we should consider the hypothesis of bipolar disorder, especially the patient has first degree relatives with BD and high cyclothymic temperament scores. A hyperthymic temperament has also been associated with bipolarity (Goto et al., n Corresponding author at: Department of Mental Health, Faculty of Medicine, Federal University, Minas Gerais, Av. Alfredo Balena, 190, Belo Horizonte, MG 30130-100, Brazil. Tel.: þ 55 31 3248 9785. E-mail addresses: [email protected], [email protected] (H. Correa).
0165-0327/$ - see front matter & 2013 Published by Elsevier B.V. http://dx.doi.org/10.1016/j.jad.2013.10.038
2011), but might actually constitute a “protective factor” in subjects without susceptibility to bipolar disorder (Evans et al., 2005; Mendlowicz et al., 2005b). Henry et al. (1999) assessed both depressive temperament (DT) and hyperthymic temperament (HT) in a dimensional approach (Akiskal and Mallya, 1987). They found statistically significant associations between a higher DT score or a lower HT score and a greater number of mood episodes in bipolar disorder patients. Furthermore, a higher DT score was strongly associated with a higher percentage of major depressive episodes and with a history of suicide attempts, while a higher HT score was associated with a trend to manic rather than depressive episodes (Henry et al., 1999). Additionally, there is evidence suggesting that among patients with Major Depressive Disorder (MDD), mood lability predicts switching to Bipolar II Disorder (Akiskal et al., 1995), and that in recurrent depressive patients cyclothymic temperament is associated with several clinical factors which are predictive of bipolarity (Mechri et al., 2011). Rihmer et al. (2013) found a statistically significant association between depressive and cyclothymic affective temperament and a personal history of suicide
A. de Aguiar Ferreira et al. / Journal of Affective Disorders 155 (2014) 138–141
attempts, and between cyclothymic and anxious temperament and a family history of completed suicide in first and second degree relatives. These data are consistent with the theory that affective temperament traits are familial and may constitute markers which predict vulnerability and may help to identify those at risk of developing a specific type of mood disorder and or a propensity to suicidal behavior. Another important issue in the clinical management of mood disorders is the ability to predict a response to antidepressant pharmacotherapy. Antidepressants treatments have beneficial and adverse effects; they can be efficacious, ineffective, or even harmful. Henry et al. (2001) found that bipolar disorder patients with a hyperthymic temperament have a greater risk of experiencing antidepressant-associated mania. Kaneda et al. (2011), using the Japanese version of the Cloninger's temperament and character inventory (TCI) (Kijima et al., 1996), suggested that personality characteristics of patients with MDD may influence the antidepressant response time. In this study, the early responders showed less harm avoidance (HA) and more self-directedness than later responders and non-responders groups. In their systematic review and meta-analysis, Kampman and Poutanen (2011) reported that an indisputable association existed between TCI scores (particularly HA) and the antidepressant treatment response experienced by patients with MDD. In studies of MDD patients, a consistent negative change in HA was found during treatment and this change was even more clearly associated with treatment response (Kampman and Poutanen, 2011). Building on these evidences, we used the Temperament Evaluation of Memphis, Pisa, Paris, and San Diego (TEMPS), an instrument designed for measuring affective temperaments (Akiskal et al., 2005), to investigate the influence of personality factors on antidepressant response. The self-administered questionnaire version (TEMPS-A) is a Yes-or-No type instrument which contains 110 items. It assesses dysthymic (items 1–22), cyclothymic (items 23–42), hyperthymic (items 43–63), irritable (items 64–84) and anxious (items 85–110) temperaments. Over the past ten years, the TEMPS-A has been translated into more than 25 languages, including a brief Brazilian version (TEMPS-Rio de Janeiro, we which abbreviate TEMPS-RJ), a validated compact scale with a total of 45 items. TEMPS-RJ consists of eight items assigned to each of the five original subscales and five items to the “worrying” subscale, which corresponds to a “general distress factor” (Woodruff et al., 2011). In the present study, using the TEMPS-RJ in a Brazilian sample, we evaluated the temperament profiles of bipolar and unipolar depressive patients against the antidepressant response in these groups. We hypothesized that affective temperament is a moderator of antidepressant response in mood disorder patients.
2. Methods The patients for the study were recruited from the Mental Health Treatment Unit of the Medical Sciences Faculty of Minas Gerais (FCMMG), the Mood Disorders Treatment Units of UFMG (Universidade Federal, Minas Gerais) and Raul Soares Institute/ FHEMIG. Diagnosis was made by a trained psychiatrist using a structured interview, MINI-PLUS, following DSM-IV criteria (Amorim, 2000). A complete review of medical records and an interview with the patient and at least one close relative was made to determine the lifetime history of the first antidepressant treatment. Severity of mood symptoms during the interview was assessed using the 17 item version of the Hamilton Depression Rating Scale (HDRS-17) (Hamilton, 1960) and the Young Mania Rating Scale (YMRS) (Young et al., 1978). To be eligible subjects had to score less than eight on the HDRS and YMRS, and had to have
139
had at least one depressive episode treated with a selective serotonin reuptake inhibitor (SSRI) for eight weeks. We enrolled 178 subjects: (1) Bipolar I/II Disorder patients [n¼ 90; 35.5% men; mean age¼38.4 (712.0) yr]; (2) major depressive disorder patients [n¼ 88; 25% men; mean age¼ 46.8 (711.9) yr]. The research protocol and consent forms and procedures were approved by each institution's Ethics Committee, in accordance with the Helsinki Declaration. Written informed consent was obtained from all subjects. Using the TEMPS-RJ instrument, affective temperament profiles were generated for all subjects. BD and MDD groups were each divided by treatment response into four subgroups. BD patients who had a total remission of depressive symptoms after eight weeks of antidepressant treatment were considered complete responders (CR). Partial responders (PR) had an antidepressant response, but it was incomplete. Non-responders (NR) had no improvement of depressive symptoms, i.e., no antidepressant response. Patients who experienced a manic or hypomanic episode during the eight weeks of treatment were assigned to the antidepressant-associated mania response (AAMR) group. MDD patients were divided following the same criteria. The treatment responses for the 90 BD patients were classified as 13 CR, 26 PR, 19 NR and 32 AAMR. The treatment responses of the 88 MDD patients were classified as 31 CR, 35 PR and 22 NR. We used t-tests and chi-square to analyze the differences between patients regarding demographic (sex, age) characteristics. Differences in TEMPS-Rio de Janeiro dimensions between antidepressant response groups in each BD and MDD patients were performed by one-way ANOVA. Tukey's HSD post-hoc test was used to investigate how treatment response groups differed from each other. In BD patients analysis, the non-cycling individuals (CR þPR þNR) were aggregated to permit specific comparison with AAMR group. All analyses were performed using SPSS for Windows version 19.0.
3. Results Among BD patients, there were no statistically significant differences between the CR, PR, NR and AAMR subgroups with regard to sex [λ2(3)¼ 1.17, p 40.05] or to age [F(3) ¼0.99, p 40.05]. Similarly, no difference was found for these variables when we considered the comparison between non-cycling subgroups combined (CR þPR þNR) and the AAMR subgroup [sex λ2 (1)¼ 0.45, p4 0.05; age t (88) ¼ 0.37, p 40.05]. Table 1 presents the BD patients' mean temperament subscales scores as measured by the TEMPS-RJ for each of the antidepressant response subgroups. One-way ANOVA found statistically significant differences were across the four BD subgroups on TEMPS-RJ subscale scores. Specific post-hoc comparisons using the Tukey HSD are summarized in Table 2. Depressive (po0.001) and anxious (po0.05) temperament subscale scores were higher for the NR subgroup as compared to the CR subgroup and the differences were statistically significant. Affective temperament scores in the CR subgroup did not differ from patients who developed a manic or hypomanic episode while taking an antidepressant (the AAMR subgroup). Finally, cyclothymic and hyperthymic temperaments were associated with antidepressantassociated mania in the specific comparison between non-cycling patients and AAMR subgroup. In the analysis of MDD patients no statistically significant differences were found with regard to sex or age when we considered comparisons across the three antidepressant response subgroups [sex λ2 (2)¼0.80, p 40.05; age F(2)¼ 0.18, p4 0.05].
140
A. de Aguiar Ferreira et al. / Journal of Affective Disorders 155 (2014) 138–141
Table 1 Temperament subscale scores (mean 7 SD) on the TEMPS-RJ for the BD subgroups. TEMPS-RJ subscales
NR subgroup (n¼19)
PR subgroup (n ¼26)
CR subgroup (n¼ 13)
[NR þPR þ CR] (n¼58)
AAMR subgroup (n¼ 32)
Depressive Cyclothymic Irritable Hyperthymic Anxious Worrying
64.47 46.71 40.13 34.21 59.87 83.16
50.00 49.04 45.67 42.31 50.96 86.92
29.81 54.81 39.42 50.96 33.65 63.08
50.22 49.57 42.46 41.59 50.00 80.34
31.64 67.19 42.19 58.59 42.19 75.00
(30.69) (28.82) (19.80) (14.93) (32.43) (31.46)
(16.96) (23.43) (22.06) (20.64) (16.93) (22.59)
(18.78) (24.23) (24.39) (26.74) (19.35) (24.28)
(25.70) (25.21) (21.71) (21.12) (25.11) (27.40)
(19.57) (24.75) (22.62) (22.10) (23.38) (29.62)
Note: CR ¼complete responders, PR ¼partial responders, NR ¼non-responders, AAMR ¼antidepressant-associated mania response; data presented are means ( 7 standard deviation).
Table 2 Summary of post-hoc tests in Bipolar Disorder patients. Subgroup comparisons
Depressive
Cyclothymic
Irritable
Hyperthymic
Anxious
Worrying
NR PR NR CR NR AAMR PR CR PR AAMR CR AAMR [RC þ RPþ AR] AAMRa
þþþ þþþ þ þ þþþ
þ þ þþ
þ þ
þ
Note: CR ¼complete responders, PR ¼partial responders, NR ¼non-responders, AAMR ¼ antidepressant-associated mania response; a ( þ ) means that a statistically significant difference of p o 0.05 was found for a particular comparison, a ( þ þ ) means that a p o 0.01 was found, a ( þ þ þ ) means that a p o 0.001 was found, while a ( ) means a statistically significant difference was not found. a
Compared by Student t test.
Table 3 Temperament subscale scores (mean7SD) on the TEMPS-RJ for the MDD subgroups. TEMPS-RJ subscales
NR subgroup (n¼ 22)
PR subgroup (n¼ 35)
CR subgroup (n¼ 31)
Depressive Cyclothymic Irritable Hyperthymic Anxious Worrying
72.73 48.30 53.41 23.86 69.89 89.09
53.93 40.36 39.64 32.50 57.14 80.57
33.06 27.02 32.66 41.13 45.97 70.32
(24.29) (22.26) (27.33) (20.01) (18.36) (16.01)
(23.83) (18.71) (19.53) (20.83) (18.01) (17.81)
(18.97) (20.18) (19.28) (25.25) (21.74) (16.22)
Note: CR ¼complete responders, PR ¼partial responders, NR ¼non-responders; data presented are mean (7 standard deviation).
The Table 3 shows the MDD patients mean temperament subscale scores on the TEMPS-RJ by antidepressant response subgroups. One-way ANOVA results indicated that in six of the TEMPS-RJ subscales, there was at least one statistically significant difference between subgroups. Specific post-hoc comparisons in MDD group using the Tukey HSD are summarized in Table 4. Temperament subscale scores were higher in the NR subgroup as compared to the CR subgroup for five temperament types and the differences were statistically significant. The exception was the hyperthymic temperament, which was associated with complete antidepressant responses. In the comparison between NR and PR groups, once again the depressive and anxious temperaments were associated with a poor antidepressant response.
4. Discussion In the present study, affective temperament was assessed in mood disorder patients grouped according to their historical antidepressant response. Comparisons across these groups indicated that the mean subscale scores as measured by TEMPS-RJ for
the depressive temperament and anxious temperament were higher in non-responders; these differences were statistically significant. In bipolar disorder patients high cyclothymic and hyperthymic temperament subscale scores were related with antidepressantassociated mania. Curiously, in MDD patients the highest mean hyperthymic temperament subscale score was found in patients who had a complete remission of symptoms after using antidepressant for eight weeks. These findings suggest that personality factors could influence the antidepressant response in mood disorder patients, consisted with other studies (Kaneda et al., 2011; Kampman and Poutanen, 2011). Depressive and anxious temperaments were associated with a poor antidepressant response. In a sample of patients with type 2 diabetes, individuals with excessive depressive and anxious temperaments had more depressive symptoms, worse psychological adjustment to diabetes and worse metabolic control (Gois et al., 2011). A depressive temperament may constitute a vulnerability factor to behavioral or biological type 2 diabetes outcomes (Gois et al., 2012). These findings may support the predictive value of markedly depressive and anxious temperaments on disease outcomes, such as treatment response. However, these conclusions should await validation by a prospective design study. Our data showed higher cyclothymic and hyperthymic temperament subscale scores in bipolar disorder patients who developed manic symptoms during the first eight weeks of antidepressant treatment. This finding confirms the results founded by Henry et al. (2001) that showed hyperthymic temperament was associated with a greater risk of developing antidepressant-associated mania. An interesting finding was that depressive temperament subscale scores were higher in non-cycling patients, which suggests that bipolar disorder patients with markedly depressive temperaments have a lower risk of antidepressant-associated mania. The controversial issue regarding antidepressant use in bipolar depression warrants further research incorporating the affective temperament profiles of patients. Hyperthymic temperament is characterized by lifelong exuberant, upbeat, overenergetic and overconfident traits (Akiskal and Akiskal, 2005). In some studies, the hyperthymic temperament
A. de Aguiar Ferreira et al. / Journal of Affective Disorders 155 (2014) 138–141
141
Table 4 Summary of post-hoc tests in MDD patients. Subgroup comparisons
Depressive
Cyclothymic
Irritable
Hyperthymic
Anxious
Worrying
NR PR NR CR PR CR
þþ þþþ þþþ
þþþ þ
þþ
þ
þ þþþ
þþþ þ
Note: CR¼ complete responders, PR¼ partial responders, NR ¼non-responders; a ( þ ) means that a statistically significant difference of p o0.05 was found for a particular comparison. a (þ þ ) means that a p o 0.01 was found, a ( þ þ þ ) means that a p o0.001 was found, while a ( ) means a statistically significant difference was not found.
subscale score did not distinguish patients from normal controls (Evans et al., 2005; Mendlowicz et al., 2005b), but in our MMD group it was associated with better antidepressant response. Kesebir et al. (2013) investigated the presence of a relationship between affective temperament and resilience in patients with a MDD diagnosis and observed a strong relationship between hyperthymic temperament and psychological resilience. Studies to investigate if these hyperthymic traits could be associated with coping strategies that improve antidepressant action are needed, but we have to keep in mind that this potentially positive factor might be lost if a subject happens to have the predisposition to develop manic symptoms. The present study has several limitations. The evaluation of antidepressant response was retrospective, with all the inherent limitations of such analysis. An ideal study that aims to characterize antidepressant responses should be longitudinal and use drugnaive patients randomized to receive an antidepressant or a placebo in a double blind fashion. We grouped together bipolar I and II patients even though there are some differences in the clinical courses of these two types of mood disorder (Vieta et al., 1997). Despite the known clinical differences, however, many have hypothesized that bipolar II disorder is intermediate to bipolar I disorder and major depressive disorder on a continuum of affective disorders (Evans et al., 2005). Although we tried to decrease the influence of the mood state in patient answers by accepting only subjects who scored less than eight on the HDRS and YMRS, some effect of mood symptoms among patients has to be considered a limitation. Finally, the lack of standardized treatment intervention is another limitation. In conclusion, our data are consistent with the theory of affective temperament traits being factors that may influence the antidepressant response and recovery from depressive episodes, but more longitudinal studies are needed to test the theory and corroborate our findings. This theory suggests that the evaluation of affective temperaments profiles in antidepressant treatments highlights the individual variability in therapeutic response among individuals with mood disorders. Role of funding source Funding for this study was provided by CNPq and FAPEMIG. The CNPq and FAPEMIG had no further role in study design, in the collection, analysis and interpretation of data, in the writing of the report, and in the decision to submit the paper for publication.
Conflict of interest All authors declare that they have no conflict of interest.
Acknowledgments This work was supported by CNPq and FAPEMIG.
References Aguiar Ferreira, A. de, Vasconcelos, A.G., Neves, F.S., Laks, J., Correa, H., 2013. Affective temperaments: familiality and clinical use in mood disorders. J. Affect. Disord. 148 (1), 53–56.
Akiskal, H.S., Akiskal, K.K., Haykal, R.F., Manning, S., Connor, P., 2005. TEMPS-A: progress towards validation of a self-rated clinical version of the Temperament Evaluation of the Memphis, Pisa, Paris, and San Diego Autoquestionnaire. J. Affect. Disord. 85, 3–16. Akiskal, H.S., Mallya, G., 1987. Criteria for the “soft” bipolar spectrum: treatment implications. Psychopharmacol. Bull. 23, 68–73. Akiskal, H.S., Maser, J.D., Zeller, P.J., Endicott, J., Coryell, W., Keller, M., Warshaw, M., Clayton, P., Goodwin, F., 1995. Switching from ‘unipolar’ to bipolar II. An 11-year prospective study of clinical and temperamental predictors in 559 patients. Arch. Gen. Psychiatry 52, 114–123. Akiskal, K.K., Akiskal, H.S., 2005. The theoretical underpinnings of affective temperaments: implications for evolutionary foundations of bipolar disorder and human nature. J. Affect. Disord. 85 (1–2), 231–239. Amorim, P., 2000. Mini international neuropsychiatric interview (MINI): validation of a short structured diagnostic psychiatric interview. Rev. Bras. Psiquiatr. 22, 106–115. Evans, L., Akiskal, H.S., Keck Jr., P.E., McElroy, S.L., Sadovnick, A.D., Remick, R.A., Kelsoe, J.R., 2005. Familiality of temperament in bipolar disorder: support for a genetic spectrum. J. Affect. Disord. 85, 153–168. Gois, C., Akiskal, H., Akiskal, K., Figueira, M.L., 2012. Depressive temperament, distress, psychological adjustment and depressive symptoms in type 2 diabetes. J. Affect. Disord. 143, 1–4. Gois, C., Barbosa, A., Ferro, A., Santos, A.L., Sousa, F., Akiskal, H., Akiskal, K., Figueira, M.L., 2011. The role of affective temperaments in metabolic control in patients with type 2 diabetes. J. Affect. Disord. 134 (1–3), 52–58. Goto, S., Terao, T., Hoaki, N., Wang, Y., 2011. Cyclothymic and hyperthymic temperaments may predict bipolarity in major depressive disorder: a supportive evidence for bipolar II1/2 and IV. J. Affect. Disord. 129, 34–38. Hamilton, M., 1960. A rating scale for depression. J. Neurol., Neurosurg. Psychiatry 23, 56–62. Henry, C., Lacoste, J., Bellivier, F., Verdoux, H., Bourgeois, M.L., Leboyer, M., 1999. Temperament in bipolar illness: impact on prognosis. J. Affect. Disord. 56 (2–3), 103–108. Henry, C., Sorbara, F., Lacoste, J., Gindre, C., Leboyer, M., 2001. Antidepressantinduced mania in bipolar patients: identification of risk factors. J. Clin. Psychiatry 62, 249–255. Kampman, O., Poutanen, O., 2011. Can onset and recovery in depression be predicted by temperament? A systematic review and meta-analysis. J. Affect. Disord. 135, 20–27. Kaneda, A., Yasui-Furukori, N., Nakagami, T., Sato, Y., Kaneko, S., 2011. The influence of personality factors on paroxetine response time in patients with major depression. J. Affect. Disord. 135, 321–325. Kesebir, S., Gündoğar, D., Küçüksubaşı, Y., Tatlıdil Yaylacı, E., 2013. The relation between affective temperament and resilience in depression: a controlled study. J. Affect. Disord. 148 (2–3), 352–356. 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 the Temperament and Character Inventory. Arch. Psychiatr. Diagnosis Clin. Eval. 7, 379–399. Mechri, A., Kerkeni, N., Touati, I., Bacha, M., Gassab, L., 2011. Association between cyclothymic temperament and clinical predictors of bipolarity in recurrent depressive patients. J. Affect. Disord. 132 (1–2), 285–288. Mendlowicz, M.V., Akiskal, H.S., Kelsoe, J.R., Rapaport, M.H., Jean-Louis, G., Gillin, J. C., 2005a. Temperament in the clinical differentiation of depressed bipolar and unipolar major depressive patients. J. Affect. Disord. 84, 219–223. Mendlowicz, M.V., Jean-Louis, G., Kelsoe, J.R., Akiskal, H.S., 2005b. A comparison of recovered bipolar patients, healthy relatives of bipolar probands, and normal controls using the short TEMPS-A. J. Affect. Disord. 85, 147–151. Rihmer, Z., Gonda, X., Torzsa, P., Kalabay, L., Akiskal, H.S., Eory, A., 2013. Affective temperament, history of suicide attempt and family history of suicide in general practice patients. J. Affect. Disord. 149 (1–3), 350–354. Vieta, E., Gasto, C., Otero, A., Nieto, E., Vallejo, J., 1997. Differential features between bipolar I and bipolar II disorder. Compr. Psychiatry 38, 98–101. Woodruff, E., Genaro, L.T., Landeira-Fernandez, J., Cheniaux, E., Laks, J., Jean-Loius, G., Nardi, A.E., Versiani, M.C., Akiskal, H.S., Mendlowicz, M.V., 2011. Validation of the Brazilian brief version of the temperament auto-questionnaire TEMPS-A: the brief TEMPS-Rio de Janeiro. J. Affect. Disord. 134, 65–76. Young, R.C., Biggs, J.T., Ziegler, V.E., Meyer, D.A., 1978. A rating scale for mania: reliability, validity and sensitivity. Br. J. Psychiatry 133, 429–435.
Contents lists available at ScienceDirect
Journal of Affective Disorders journal homepage: www.elsevier.com/locate/jad
Research report
Affective temperaments and antidepressant response in the clinical management of mood disorders Alexandre de Aguiar Ferreira a,b,c, Alina Gomide Vasconcelos a, Fernando Silva Neves a,d, Humberto Correa a,d,n a
Neuroscience Program, Federal University, Minas Gerais (UFMG), Belo Horizonte, MG, Brazil Raul Soares Institute – FHEMIG, Belo Horizonte, MG, Brazil c Faculty of Medical Sciences of Minas Gerais – FCMMG, Belo Horizonte, MG, Brazil d Department of Mental Health, Faculty of Medicine, Federal University, Minas Gerais, Av. Alfredo Balena, 190, Belo Horizonte, MG 30130-100, Brazil b
art ic l e i nf o
a b s t r a c t
Article history: Received 25 September 2013 Accepted 22 October 2013 Available online 28 October 2013
Background: The aim of this study was to investigate the presence of a relationship between affective temperament and antidepressant treatment response in mood disorder patients. Methods: The lifetime history of antidepressant response of 90 bipolar disorder patients and 88 major depressive disorder patients were retrospectively evaluated and then assigned to one of four subgroups: complete response (CR), partial response (PR), no response (NR), and antidepressant associated mania response (AAMR). Using TEMPS-Rio de Janeiro – the brief Brazilian version of TEMPS-A – we compared affective temperament subscale scores across these groups. Results: We observed a statistically significant relationship between depressive and anxious affective temperaments and no antidepressant response. In bipolar disorder patients, cyclothymic temperament (po 0.01) and hyperthymic temperament (po 0.05) were associated with antidepressant-associated mania. Hyperthymic temperament was associated with complete antidepressant responses in major depressive disorder patients. Limitations: The evaluation of antidepressant response was retrospective. Conclusions: Our data are consistent with the theory that affective temperament traits are factors that can influence the antidepressant response and the recovery from depressive episodes, but more longitudinal studies are needed to confirm this theory and our findings. & 2013 Published by Elsevier B.V.
Keywords: Affective temperament TEMPS-A TEMPS-Rio de Janeiro Antidepressant response Mood disorders
1. Introduction In the general approach to patients with mood disorders the characterization of affective temperament may be an important parameter. It may be useful in the differential diagnosis of mood disorders as studies have consistently shown that cyclothymic temperament is significantly more prominent in bipolar versus unipolar depressive patients (Mendlowicz et al., 2005a) and that cyclothymic traits may represent vulnerability markers found in clinically healthy relatives of bipolar disorder (BD) patients (Mendlowicz et al., 2005b; Aguiar Ferreira et al., 2013). So, when assessing depressive patients, we should consider the hypothesis of bipolar disorder, especially the patient has first degree relatives with BD and high cyclothymic temperament scores. A hyperthymic temperament has also been associated with bipolarity (Goto et al., n Corresponding author at: Department of Mental Health, Faculty of Medicine, Federal University, Minas Gerais, Av. Alfredo Balena, 190, Belo Horizonte, MG 30130-100, Brazil. Tel.: þ 55 31 3248 9785. E-mail addresses: [email protected], [email protected] (H. Correa).
0165-0327/$ - see front matter & 2013 Published by Elsevier B.V. http://dx.doi.org/10.1016/j.jad.2013.10.038
2011), but might actually constitute a “protective factor” in subjects without susceptibility to bipolar disorder (Evans et al., 2005; Mendlowicz et al., 2005b). Henry et al. (1999) assessed both depressive temperament (DT) and hyperthymic temperament (HT) in a dimensional approach (Akiskal and Mallya, 1987). They found statistically significant associations between a higher DT score or a lower HT score and a greater number of mood episodes in bipolar disorder patients. Furthermore, a higher DT score was strongly associated with a higher percentage of major depressive episodes and with a history of suicide attempts, while a higher HT score was associated with a trend to manic rather than depressive episodes (Henry et al., 1999). Additionally, there is evidence suggesting that among patients with Major Depressive Disorder (MDD), mood lability predicts switching to Bipolar II Disorder (Akiskal et al., 1995), and that in recurrent depressive patients cyclothymic temperament is associated with several clinical factors which are predictive of bipolarity (Mechri et al., 2011). Rihmer et al. (2013) found a statistically significant association between depressive and cyclothymic affective temperament and a personal history of suicide
A. de Aguiar Ferreira et al. / Journal of Affective Disorders 155 (2014) 138–141
attempts, and between cyclothymic and anxious temperament and a family history of completed suicide in first and second degree relatives. These data are consistent with the theory that affective temperament traits are familial and may constitute markers which predict vulnerability and may help to identify those at risk of developing a specific type of mood disorder and or a propensity to suicidal behavior. Another important issue in the clinical management of mood disorders is the ability to predict a response to antidepressant pharmacotherapy. Antidepressants treatments have beneficial and adverse effects; they can be efficacious, ineffective, or even harmful. Henry et al. (2001) found that bipolar disorder patients with a hyperthymic temperament have a greater risk of experiencing antidepressant-associated mania. Kaneda et al. (2011), using the Japanese version of the Cloninger's temperament and character inventory (TCI) (Kijima et al., 1996), suggested that personality characteristics of patients with MDD may influence the antidepressant response time. In this study, the early responders showed less harm avoidance (HA) and more self-directedness than later responders and non-responders groups. In their systematic review and meta-analysis, Kampman and Poutanen (2011) reported that an indisputable association existed between TCI scores (particularly HA) and the antidepressant treatment response experienced by patients with MDD. In studies of MDD patients, a consistent negative change in HA was found during treatment and this change was even more clearly associated with treatment response (Kampman and Poutanen, 2011). Building on these evidences, we used the Temperament Evaluation of Memphis, Pisa, Paris, and San Diego (TEMPS), an instrument designed for measuring affective temperaments (Akiskal et al., 2005), to investigate the influence of personality factors on antidepressant response. The self-administered questionnaire version (TEMPS-A) is a Yes-or-No type instrument which contains 110 items. It assesses dysthymic (items 1–22), cyclothymic (items 23–42), hyperthymic (items 43–63), irritable (items 64–84) and anxious (items 85–110) temperaments. Over the past ten years, the TEMPS-A has been translated into more than 25 languages, including a brief Brazilian version (TEMPS-Rio de Janeiro, we which abbreviate TEMPS-RJ), a validated compact scale with a total of 45 items. TEMPS-RJ consists of eight items assigned to each of the five original subscales and five items to the “worrying” subscale, which corresponds to a “general distress factor” (Woodruff et al., 2011). In the present study, using the TEMPS-RJ in a Brazilian sample, we evaluated the temperament profiles of bipolar and unipolar depressive patients against the antidepressant response in these groups. We hypothesized that affective temperament is a moderator of antidepressant response in mood disorder patients.
2. Methods The patients for the study were recruited from the Mental Health Treatment Unit of the Medical Sciences Faculty of Minas Gerais (FCMMG), the Mood Disorders Treatment Units of UFMG (Universidade Federal, Minas Gerais) and Raul Soares Institute/ FHEMIG. Diagnosis was made by a trained psychiatrist using a structured interview, MINI-PLUS, following DSM-IV criteria (Amorim, 2000). A complete review of medical records and an interview with the patient and at least one close relative was made to determine the lifetime history of the first antidepressant treatment. Severity of mood symptoms during the interview was assessed using the 17 item version of the Hamilton Depression Rating Scale (HDRS-17) (Hamilton, 1960) and the Young Mania Rating Scale (YMRS) (Young et al., 1978). To be eligible subjects had to score less than eight on the HDRS and YMRS, and had to have
139
had at least one depressive episode treated with a selective serotonin reuptake inhibitor (SSRI) for eight weeks. We enrolled 178 subjects: (1) Bipolar I/II Disorder patients [n¼ 90; 35.5% men; mean age¼38.4 (712.0) yr]; (2) major depressive disorder patients [n¼ 88; 25% men; mean age¼ 46.8 (711.9) yr]. The research protocol and consent forms and procedures were approved by each institution's Ethics Committee, in accordance with the Helsinki Declaration. Written informed consent was obtained from all subjects. Using the TEMPS-RJ instrument, affective temperament profiles were generated for all subjects. BD and MDD groups were each divided by treatment response into four subgroups. BD patients who had a total remission of depressive symptoms after eight weeks of antidepressant treatment were considered complete responders (CR). Partial responders (PR) had an antidepressant response, but it was incomplete. Non-responders (NR) had no improvement of depressive symptoms, i.e., no antidepressant response. Patients who experienced a manic or hypomanic episode during the eight weeks of treatment were assigned to the antidepressant-associated mania response (AAMR) group. MDD patients were divided following the same criteria. The treatment responses for the 90 BD patients were classified as 13 CR, 26 PR, 19 NR and 32 AAMR. The treatment responses of the 88 MDD patients were classified as 31 CR, 35 PR and 22 NR. We used t-tests and chi-square to analyze the differences between patients regarding demographic (sex, age) characteristics. Differences in TEMPS-Rio de Janeiro dimensions between antidepressant response groups in each BD and MDD patients were performed by one-way ANOVA. Tukey's HSD post-hoc test was used to investigate how treatment response groups differed from each other. In BD patients analysis, the non-cycling individuals (CR þPR þNR) were aggregated to permit specific comparison with AAMR group. All analyses were performed using SPSS for Windows version 19.0.
3. Results Among BD patients, there were no statistically significant differences between the CR, PR, NR and AAMR subgroups with regard to sex [λ2(3)¼ 1.17, p 40.05] or to age [F(3) ¼0.99, p 40.05]. Similarly, no difference was found for these variables when we considered the comparison between non-cycling subgroups combined (CR þPR þNR) and the AAMR subgroup [sex λ2 (1)¼ 0.45, p4 0.05; age t (88) ¼ 0.37, p 40.05]. Table 1 presents the BD patients' mean temperament subscales scores as measured by the TEMPS-RJ for each of the antidepressant response subgroups. One-way ANOVA found statistically significant differences were across the four BD subgroups on TEMPS-RJ subscale scores. Specific post-hoc comparisons using the Tukey HSD are summarized in Table 2. Depressive (po0.001) and anxious (po0.05) temperament subscale scores were higher for the NR subgroup as compared to the CR subgroup and the differences were statistically significant. Affective temperament scores in the CR subgroup did not differ from patients who developed a manic or hypomanic episode while taking an antidepressant (the AAMR subgroup). Finally, cyclothymic and hyperthymic temperaments were associated with antidepressantassociated mania in the specific comparison between non-cycling patients and AAMR subgroup. In the analysis of MDD patients no statistically significant differences were found with regard to sex or age when we considered comparisons across the three antidepressant response subgroups [sex λ2 (2)¼0.80, p 40.05; age F(2)¼ 0.18, p4 0.05].
140
A. de Aguiar Ferreira et al. / Journal of Affective Disorders 155 (2014) 138–141
Table 1 Temperament subscale scores (mean 7 SD) on the TEMPS-RJ for the BD subgroups. TEMPS-RJ subscales
NR subgroup (n¼19)
PR subgroup (n ¼26)
CR subgroup (n¼ 13)
[NR þPR þ CR] (n¼58)
AAMR subgroup (n¼ 32)
Depressive Cyclothymic Irritable Hyperthymic Anxious Worrying
64.47 46.71 40.13 34.21 59.87 83.16
50.00 49.04 45.67 42.31 50.96 86.92
29.81 54.81 39.42 50.96 33.65 63.08
50.22 49.57 42.46 41.59 50.00 80.34
31.64 67.19 42.19 58.59 42.19 75.00
(30.69) (28.82) (19.80) (14.93) (32.43) (31.46)
(16.96) (23.43) (22.06) (20.64) (16.93) (22.59)
(18.78) (24.23) (24.39) (26.74) (19.35) (24.28)
(25.70) (25.21) (21.71) (21.12) (25.11) (27.40)
(19.57) (24.75) (22.62) (22.10) (23.38) (29.62)
Note: CR ¼complete responders, PR ¼partial responders, NR ¼non-responders, AAMR ¼antidepressant-associated mania response; data presented are means ( 7 standard deviation).
Table 2 Summary of post-hoc tests in Bipolar Disorder patients. Subgroup comparisons
Depressive
Cyclothymic
Irritable
Hyperthymic
Anxious
Worrying
NR PR NR CR NR AAMR PR CR PR AAMR CR AAMR [RC þ RPþ AR] AAMRa
þþþ þþþ þ þ þþþ
þ þ þþ
þ þ
þ
Note: CR ¼complete responders, PR ¼partial responders, NR ¼non-responders, AAMR ¼ antidepressant-associated mania response; a ( þ ) means that a statistically significant difference of p o 0.05 was found for a particular comparison, a ( þ þ ) means that a p o 0.01 was found, a ( þ þ þ ) means that a p o 0.001 was found, while a ( ) means a statistically significant difference was not found. a
Compared by Student t test.
Table 3 Temperament subscale scores (mean7SD) on the TEMPS-RJ for the MDD subgroups. TEMPS-RJ subscales
NR subgroup (n¼ 22)
PR subgroup (n¼ 35)
CR subgroup (n¼ 31)
Depressive Cyclothymic Irritable Hyperthymic Anxious Worrying
72.73 48.30 53.41 23.86 69.89 89.09
53.93 40.36 39.64 32.50 57.14 80.57
33.06 27.02 32.66 41.13 45.97 70.32
(24.29) (22.26) (27.33) (20.01) (18.36) (16.01)
(23.83) (18.71) (19.53) (20.83) (18.01) (17.81)
(18.97) (20.18) (19.28) (25.25) (21.74) (16.22)
Note: CR ¼complete responders, PR ¼partial responders, NR ¼non-responders; data presented are mean (7 standard deviation).
The Table 3 shows the MDD patients mean temperament subscale scores on the TEMPS-RJ by antidepressant response subgroups. One-way ANOVA results indicated that in six of the TEMPS-RJ subscales, there was at least one statistically significant difference between subgroups. Specific post-hoc comparisons in MDD group using the Tukey HSD are summarized in Table 4. Temperament subscale scores were higher in the NR subgroup as compared to the CR subgroup for five temperament types and the differences were statistically significant. The exception was the hyperthymic temperament, which was associated with complete antidepressant responses. In the comparison between NR and PR groups, once again the depressive and anxious temperaments were associated with a poor antidepressant response.
4. Discussion In the present study, affective temperament was assessed in mood disorder patients grouped according to their historical antidepressant response. Comparisons across these groups indicated that the mean subscale scores as measured by TEMPS-RJ for
the depressive temperament and anxious temperament were higher in non-responders; these differences were statistically significant. In bipolar disorder patients high cyclothymic and hyperthymic temperament subscale scores were related with antidepressantassociated mania. Curiously, in MDD patients the highest mean hyperthymic temperament subscale score was found in patients who had a complete remission of symptoms after using antidepressant for eight weeks. These findings suggest that personality factors could influence the antidepressant response in mood disorder patients, consisted with other studies (Kaneda et al., 2011; Kampman and Poutanen, 2011). Depressive and anxious temperaments were associated with a poor antidepressant response. In a sample of patients with type 2 diabetes, individuals with excessive depressive and anxious temperaments had more depressive symptoms, worse psychological adjustment to diabetes and worse metabolic control (Gois et al., 2011). A depressive temperament may constitute a vulnerability factor to behavioral or biological type 2 diabetes outcomes (Gois et al., 2012). These findings may support the predictive value of markedly depressive and anxious temperaments on disease outcomes, such as treatment response. However, these conclusions should await validation by a prospective design study. Our data showed higher cyclothymic and hyperthymic temperament subscale scores in bipolar disorder patients who developed manic symptoms during the first eight weeks of antidepressant treatment. This finding confirms the results founded by Henry et al. (2001) that showed hyperthymic temperament was associated with a greater risk of developing antidepressant-associated mania. An interesting finding was that depressive temperament subscale scores were higher in non-cycling patients, which suggests that bipolar disorder patients with markedly depressive temperaments have a lower risk of antidepressant-associated mania. The controversial issue regarding antidepressant use in bipolar depression warrants further research incorporating the affective temperament profiles of patients. Hyperthymic temperament is characterized by lifelong exuberant, upbeat, overenergetic and overconfident traits (Akiskal and Akiskal, 2005). In some studies, the hyperthymic temperament
A. de Aguiar Ferreira et al. / Journal of Affective Disorders 155 (2014) 138–141
141
Table 4 Summary of post-hoc tests in MDD patients. Subgroup comparisons
Depressive
Cyclothymic
Irritable
Hyperthymic
Anxious
Worrying
NR PR NR CR PR CR
þþ þþþ þþþ
þþþ þ
þþ
þ
þ þþþ
þþþ þ
Note: CR¼ complete responders, PR¼ partial responders, NR ¼non-responders; a ( þ ) means that a statistically significant difference of p o0.05 was found for a particular comparison. a (þ þ ) means that a p o 0.01 was found, a ( þ þ þ ) means that a p o0.001 was found, while a ( ) means a statistically significant difference was not found.
subscale score did not distinguish patients from normal controls (Evans et al., 2005; Mendlowicz et al., 2005b), but in our MMD group it was associated with better antidepressant response. Kesebir et al. (2013) investigated the presence of a relationship between affective temperament and resilience in patients with a MDD diagnosis and observed a strong relationship between hyperthymic temperament and psychological resilience. Studies to investigate if these hyperthymic traits could be associated with coping strategies that improve antidepressant action are needed, but we have to keep in mind that this potentially positive factor might be lost if a subject happens to have the predisposition to develop manic symptoms. The present study has several limitations. The evaluation of antidepressant response was retrospective, with all the inherent limitations of such analysis. An ideal study that aims to characterize antidepressant responses should be longitudinal and use drugnaive patients randomized to receive an antidepressant or a placebo in a double blind fashion. We grouped together bipolar I and II patients even though there are some differences in the clinical courses of these two types of mood disorder (Vieta et al., 1997). Despite the known clinical differences, however, many have hypothesized that bipolar II disorder is intermediate to bipolar I disorder and major depressive disorder on a continuum of affective disorders (Evans et al., 2005). Although we tried to decrease the influence of the mood state in patient answers by accepting only subjects who scored less than eight on the HDRS and YMRS, some effect of mood symptoms among patients has to be considered a limitation. Finally, the lack of standardized treatment intervention is another limitation. In conclusion, our data are consistent with the theory of affective temperament traits being factors that may influence the antidepressant response and recovery from depressive episodes, but more longitudinal studies are needed to test the theory and corroborate our findings. This theory suggests that the evaluation of affective temperaments profiles in antidepressant treatments highlights the individual variability in therapeutic response among individuals with mood disorders. Role of funding source Funding for this study was provided by CNPq and FAPEMIG. The CNPq and FAPEMIG had no further role in study design, in the collection, analysis and interpretation of data, in the writing of the report, and in the decision to submit the paper for publication.
Conflict of interest All authors declare that they have no conflict of interest.
Acknowledgments This work was supported by CNPq and FAPEMIG.
References Aguiar Ferreira, A. de, Vasconcelos, A.G., Neves, F.S., Laks, J., Correa, H., 2013. Affective temperaments: familiality and clinical use in mood disorders. J. Affect. Disord. 148 (1), 53–56.
Akiskal, H.S., Akiskal, K.K., Haykal, R.F., Manning, S., Connor, P., 2005. TEMPS-A: progress towards validation of a self-rated clinical version of the Temperament Evaluation of the Memphis, Pisa, Paris, and San Diego Autoquestionnaire. J. Affect. Disord. 85, 3–16. Akiskal, H.S., Mallya, G., 1987. Criteria for the “soft” bipolar spectrum: treatment implications. Psychopharmacol. Bull. 23, 68–73. Akiskal, H.S., Maser, J.D., Zeller, P.J., Endicott, J., Coryell, W., Keller, M., Warshaw, M., Clayton, P., Goodwin, F., 1995. Switching from ‘unipolar’ to bipolar II. An 11-year prospective study of clinical and temperamental predictors in 559 patients. Arch. Gen. Psychiatry 52, 114–123. Akiskal, K.K., Akiskal, H.S., 2005. The theoretical underpinnings of affective temperaments: implications for evolutionary foundations of bipolar disorder and human nature. J. Affect. Disord. 85 (1–2), 231–239. Amorim, P., 2000. Mini international neuropsychiatric interview (MINI): validation of a short structured diagnostic psychiatric interview. Rev. Bras. Psiquiatr. 22, 106–115. Evans, L., Akiskal, H.S., Keck Jr., P.E., McElroy, S.L., Sadovnick, A.D., Remick, R.A., Kelsoe, J.R., 2005. Familiality of temperament in bipolar disorder: support for a genetic spectrum. J. Affect. Disord. 85, 153–168. Gois, C., Akiskal, H., Akiskal, K., Figueira, M.L., 2012. Depressive temperament, distress, psychological adjustment and depressive symptoms in type 2 diabetes. J. Affect. Disord. 143, 1–4. Gois, C., Barbosa, A., Ferro, A., Santos, A.L., Sousa, F., Akiskal, H., Akiskal, K., Figueira, M.L., 2011. The role of affective temperaments in metabolic control in patients with type 2 diabetes. J. Affect. Disord. 134 (1–3), 52–58. Goto, S., Terao, T., Hoaki, N., Wang, Y., 2011. Cyclothymic and hyperthymic temperaments may predict bipolarity in major depressive disorder: a supportive evidence for bipolar II1/2 and IV. J. Affect. Disord. 129, 34–38. Hamilton, M., 1960. A rating scale for depression. J. Neurol., Neurosurg. Psychiatry 23, 56–62. Henry, C., Lacoste, J., Bellivier, F., Verdoux, H., Bourgeois, M.L., Leboyer, M., 1999. Temperament in bipolar illness: impact on prognosis. J. Affect. Disord. 56 (2–3), 103–108. Henry, C., Sorbara, F., Lacoste, J., Gindre, C., Leboyer, M., 2001. Antidepressantinduced mania in bipolar patients: identification of risk factors. J. Clin. Psychiatry 62, 249–255. Kampman, O., Poutanen, O., 2011. Can onset and recovery in depression be predicted by temperament? A systematic review and meta-analysis. J. Affect. Disord. 135, 20–27. Kaneda, A., Yasui-Furukori, N., Nakagami, T., Sato, Y., Kaneko, S., 2011. The influence of personality factors on paroxetine response time in patients with major depression. J. Affect. Disord. 135, 321–325. Kesebir, S., Gündoğar, D., Küçüksubaşı, Y., Tatlıdil Yaylacı, E., 2013. The relation between affective temperament and resilience in depression: a controlled study. J. Affect. Disord. 148 (2–3), 352–356. 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 the Temperament and Character Inventory. Arch. Psychiatr. Diagnosis Clin. Eval. 7, 379–399. Mechri, A., Kerkeni, N., Touati, I., Bacha, M., Gassab, L., 2011. Association between cyclothymic temperament and clinical predictors of bipolarity in recurrent depressive patients. J. Affect. Disord. 132 (1–2), 285–288. Mendlowicz, M.V., Akiskal, H.S., Kelsoe, J.R., Rapaport, M.H., Jean-Louis, G., Gillin, J. C., 2005a. Temperament in the clinical differentiation of depressed bipolar and unipolar major depressive patients. J. Affect. Disord. 84, 219–223. Mendlowicz, M.V., Jean-Louis, G., Kelsoe, J.R., Akiskal, H.S., 2005b. A comparison of recovered bipolar patients, healthy relatives of bipolar probands, and normal controls using the short TEMPS-A. J. Affect. Disord. 85, 147–151. Rihmer, Z., Gonda, X., Torzsa, P., Kalabay, L., Akiskal, H.S., Eory, A., 2013. Affective temperament, history of suicide attempt and family history of suicide in general practice patients. J. Affect. Disord. 149 (1–3), 350–354. Vieta, E., Gasto, C., Otero, A., Nieto, E., Vallejo, J., 1997. Differential features between bipolar I and bipolar II disorder. Compr. Psychiatry 38, 98–101. Woodruff, E., Genaro, L.T., Landeira-Fernandez, J., Cheniaux, E., Laks, J., Jean-Loius, G., Nardi, A.E., Versiani, M.C., Akiskal, H.S., Mendlowicz, M.V., 2011. Validation of the Brazilian brief version of the temperament auto-questionnaire TEMPS-A: the brief TEMPS-Rio de Janeiro. J. Affect. Disord. 134, 65–76. Young, R.C., Biggs, J.T., Ziegler, V.E., Meyer, D.A., 1978. A rating scale for mania: reliability, validity and sensitivity. Br. J. Psychiatry 133, 429–435.