The influence of affective temperaments and psychopathological traits on the definition of bipolar disorder subtypes: A study on Bipolar I Italian National sample

Abstract

Affective temperament and psychopathological traits such as separation anxiety (SA) and interpersonal sensitivity (IPS) are supposed to impact on the clinical manifestation and on the course of Bipolar Disorder (BD); in the present study we investigated their influence on the definition of BD subtypes.

Method:

Among 106 BD-I patients with DSM-IV depressive, manic or mixed episode included in a multi-centric Italian study and treated according to the routine clinical practice, 89 (84.0%) were in remission after a follow-up period ranging from 3 to 6 months (Clinical Global Impression-BP [CGI-BP] < 2). Remitting patients underwent a comprehensive evaluation including self-report questionnaires such as the Temperament Evaluation of Memphis, Pisa, Paris and San Diego (TEMPS-A) scale, Separation Anxiety Symptom Inventory (SASI), Interpersonal Sensitivity Measure (IPSM) and the Semi-structured interview for Mood Disorder (SIMD-R) administered by experienced clinicians. Correlation and factorial analyses were conducted on temperamental and psychopathological measures. Comparative analyses were conducted on different temperamental subtypes based on the TEMPS-A, SASI and IPSM profile.

Results:

Depressive, cyclothymic and irritable TEMPS-A score and SASI and IPSM total scores were positively and statistically correlated with each other. On the contrary, hyperthymic temperament score was negatively correlated with depressive temperament and not significantly correlated with the other temperamental and psychopathological dimensions. The factorial analysis of the TEMPS-A subscales and SASI and IPSM total scores allowed the extraction of 2 factors: the cyclothymic-sensitive (explaining 46% of the variance) that included, as positive components, depressive, cyclothymic, irritable temperaments and SASI and IPSM scores; the hyperthymic (explaining the 19% of the variance) included hyperthymic temperament as the only positive component and depressive temperament and IPSM, as negative components. Dominant cyclothymic-sensitive patients (n = 49) were more frequently females and reported higher number of depressive, hypomanic and suicide attempts when compared to the dominant hyperthymic patients (n = 40). On the contrary, these latter showed a higher number of manic episodes and hospitalizations than cyclothymic-sensitive patients. The rates of first-degree family history for both mood and anxiety disorders were higher in cyclothymic-sensitive than in hyperthymic patients. Cyclothymic sensitive patients also reported more axis I lifetime co-morbidities with Panic Disorder/Agoraphobia and Social Anxiety Disorder in comparison with hyperthymics. As concerns axis II co-morbidity the cyclothymic-sensitive patients met more frequently DSM-IV criteria 1, 5 and 7 for borderline personality disorder than the hyperthymics. On the contrary, antisocial personality disorder was more represented among hyperthymic than cyclothymic patients, in particular for DSM-IV criteria 1 and 6.

Limitation:

No blind evaluation and uncertain validity of personality inventory.

Conclusion:

Our results support the view that affective temperaments influence the clinical features of BD in terms of both clinical and course characteristics, family history and axis I and II co-morbidities. Hypothetical temperamental subtypes as measured by TEMPS-A presented important interrelationships that permit to reliably isolate two fundamental temperamental disposition: the first characterized by rapid fluctuations of mood and emotional instability, and the second by hyperactivity, high level of energy and emotional intensity. Dominant cyclothymic and hyperthymic bipolar I patients reported important differences in terms of gender distribution, number and polarity of previous episodes, hospitalizations, suicidality, rates of co-morbid anxiety and personality traits and disorders. Our data are consistent with the hypothesis that affective temperaments, and in particular cyclothymia, could be utilized as quantitative, intermediate phenotypes in order to identify BD susceptibility genes.

Introduction

Affective temperaments are supposed to represent behavioral endophenotypes genetically related to mood disorders (Akiskal et al., 2006b, Evans et al., 2005, Evans et al., 2008, Gonda et al., 2009, Gonda et al., 2006, Rihmer et al., 2007) and possibly increasing the predisposition for developing affective states (Evans et al., 2005, Rottig et al., 2007). Yet, the concept of fundamental, constitutional, biologically determined, temperamental traits as background of mood states and major affective illness, is not a novel acquisition. In fact, Kraepelin [1921] already observed that the cyclothymic, irritable, anxious, depressive “fundamental states” preceded the onset of several manic-depressive episodes, and persisted also during the “lucid” intervals; those observations led him to consider affective temperaments as attenuate expressions of manic-depressive illness as well the background for overt clinical forms (Kraepelin, 1921).

Affective temperament and psychopathological traits are also supposed to have an impact on the clinical manifestation and on the course of Bipolar Disorder (BD) (Akiskal, 2000, Hantouche et al., 1998). However their predictive value on outcome still requires further research. This is particularly true for putative temperamental (e.g. cyclothymia and hyperthymia) and psychopathological traits (e.g. separation anxiety and interpersonal sensitivity), which might play a predisposing or pathoplastic role in the antidepressant-induced (hypo)manic vulnerability (Akiskal et al., 2003b) and/or in several co-morbid syndromes—for review see (Perugi and Akiskal, 2002).

Most of the available literature on the pathoplastic role of temperament is focused on Bipolar II (Akiskal et al., 2003a, Benazzi, 2007, Cassano et al., 1992) or Unipolar samples (Maina et al., 2009). For example, arising from a cyclothymic temperament, the bipolar II patient could pursue an unstable course (Akiskal et al., 2003a), present multiple co-morbidities with anxiety, eating and alcohol/substance use disorders (Perugi et al., 2006), an increased risk for suicidal behavior (Kochman et al., 2005) and could be more likely to be misdiagnosed as an axis II cluster B (Hantouche et al., 2003). In the same line, unipolar patients with cyclothymia or hyperthymia are reported to have earlier age at onset, higher rates of cluster B personality disorder and first-degree family history for BD in comparison with unipolar patients without such a temperamental profile (Maina et al., 2009). The only reports in Bipolar I patients focused on the relationship between temperament and mixed states (Akiskal et al., 1998a, Rottig et al., 2007), supporting the hypothesis that mixed episodes are more frequent in subjects with inverse temperaments (Akiskal and Benazzi, 2003).

Separation anxiety (SA) is included in DSM-IV among childhood and adolescence disorders and it has been typically described in juvenile populations and/or as a childhood antecedent of panic/agoraphobic spectrum disorders (APA, 1994). An adult form of excessive and often disabling distress in the face of actual or perceived separation from major attachment figures has been recently described (Silove et al., 2007). Epidemiological (Bruckl et al., 2007, Shear et al., 2006) and clinical (Costa et al., 2009) data supported the hypothesis that early SA tends to persist in adulthood and operates as a general vulnerability factor for adult anxiety or mood disorders. In particular, juvenile and adult SA have been associated with the development of bipolar spectrum disorders (Pini et al., 2005, Toni et al., 2008).

Interpersonal sensitivity (IPS) is a construct referring to an individual's hypersensitivity to perceived self-deficiencies in relation to others. Partially related to weak self-esteem, it embraces sensitivity to rejection and criticism on the part of others (Davidson et al., 1989). IPS can be defined and rated with good reliability (Stafford, 2007) and has good construct validity, acceptable factorial invariance and differential sensitivity to the effect of treatment (Davidson et al., 1989). A long-standing pattern of rejection sensitivity has been included in DSM-IV as a diagnostic criterion for Atypical Depression. The inclusion of a stable psychopathological trait for axis I syndromes has no precedent and automatically increases the severity and continuity of the syndrome. Increasing evidence for an overlap between traits and states symptomatology in mood disorders suggests that systematic consideration of combined features may be worthwhile (Clayton et al., 1994). In a study conducted by our group in a large sample of patients with atypical depression, mood instability of cyclothymic type resulted strongly related to IPS and to mood reactivity, suggesting the existence of a common background (Perugi et al., 2003). In BD II patients with atypical depression, IPS and mood reactivity seem to be two strictly related characteristics, representing two different features, cognitive and affective, of the same psycho(patho)logical dimension (Perugi et al., 2003).

The aim of this study is to explore, in a national sample of BD I patients, the influence of affective temperaments and psychopathological traits such as separation anxiety and interpersonal sensitivity, prospectively evaluated during the euthymic phase, on the clinical features and the course of BD.