The soft bipolar spectrum redefined: focus on the cyclothymic, anxious-sensitive, impulse-dyscontrol, and binge-eating connection in bipolar II and related conditions

Psychiatr Clin N Am 25 (2002) 713–737

The soft bipolar spectrum redefined: focus on the cyclothymic, anxious-sensitive, impulse-dyscontrol, and binge-eating connection in bipolar II and related conditions Giulio Perugia,b,*, Hagop S. Akiskalc,d a

Institute of Behavioral Sciences G.De Lisio, Viale Monzone 3, 54031 Carrara, Italy b Department of Psychiatry, University of Pisa, Via Roma 67, 56100 Pisa, Italy c International Mood Center, Department of Psychiatry, University of California, San Diego, 3350 La Jolla Village Drive, San Diego, CA 92161, USA d Department of Psychiatry, VA Hospital, 9500 Gilman Drive, La Jolla, CA 92093-0603, USA

The concept of spectrum in psychiatry The concept of spectrum originally was used in physics to indicate an apparent qualitative distinction arising from a quantitative continuum (ie, a series of colors formed when a beam of white light is dispersed by a prism so that their parts are arranged in the order of their wavelengths) [1]. In psychiatry, the concept of spectrum was first used with a slightly different connotation to identify a group of disorders that is qualitatively distinct in appearance (eg, depression and alcoholism) but believed to be related from a pathogenetic point of view [2]. For different investigators, the hypothetical common pathogenetic link has been of a different nature: familial-genetic (ie, schizophrenia spectrum) [3–6]; response to a class of pharmacologic treatments (ie, depressive spectrum) [7]; common underlying psychopathologic traits (ie, dissociative spectrum) [8,9]; the combination of some of the foregoing validating strategies (ie, bipolar spectrum); and obsessivecompulsive spectrum [10–14]. Department of Psychiatry, University of Pisa, Via Roma 67, 56100, Pisa, Italy. * Corresponding author. E-mail address: [email protected] (G. Perugi). 0193-953X/02/$ - see front matter
2002, Elsevier Science (USA). All rights reserved. PII: S 0 1 9 3 - 9 5 3 X ( 0 2 ) 0 0 0 2 3 - 0

714

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

Strictly speaking, none of the foregoing spectrum conditions is part of conventional psychiatric thinking as conceptualized in official nomenclature. For instance, schizotypal personality is classified on axis II rather than as an attenuated precursor of schizophrenia on axis I in the Diagnostic and Statistical Manual of Mental Disorders (DSM)-IV [15]. On the other hand, cyclothymia, which is validated as a spectrum bipolar disorder [16,17], is classified on axis I within bipolar disorders without drawing clear-cut pathogenetic links between the two. Many North American clinicians, following Gunderson and Phillips [18], preferentially diagnose the mood-unstable borderline personality rather than cyclothymia, thereby missing the opportunity to make sense of the temperamental foundations of instability in the lives of their bipolar patients. Fortunately, this nefarious influence of the ‘‘borderline’’ concept is changing [19]. Apropos DSM-IV, one wonders what type of inconsistent nosologic logic was used to dissociate (ie, place them on separate axes) schizotypal from schizophrenia when the evidence strongly supports the link between the two and permit labeling the temperamental pathology of bipolar patients with the terminology of axis II when cyclothymia is already accorded a position in the affective realm of axis I. Other investigators have used the term ‘‘spectrum’’ in a different meaning to refer to ‘‘broad areas of psychiatric phenomena relating to a single mental disorder’’ [20]. This latter kind of psychiatric spectrum should include ‘‘core, atypical and subclinical symptoms of the primary Axis I disorder; signs, isolated symptoms, symptom clusters, and behavioral patterns related to the core symptoms, that may be prodromal, … precursors … or sequelae of a … fullfledged disorder.’’ This conceptualization refers to a broad definition of syndrome or disorder that enlarges upon a set of criteria—essentially based on state symptomatology—proposed by current formal diagnostic manuals. The use of the term ‘‘spectrum’’ in this sense is superposable to that of syndromic continuum [21]. If the syndrome or disorder is poorly defined by a certain set of core criteria, as seems to be the case for mood disorders in DSM-IV or International Classification of Diseases (ICD-10), it would be appropriate to change or broaden the latter without introducing a new term to indicate what is still a more widely defined syndrome or disorder. This article exemplifies the use of the concept of the spectrum in a much broader sense. After a brief review of the concept of bipolar spectrum as developed by one of the authors [10,22], the article reviews and integrates the emerging evidence for a ‘‘soft bipolar spectrum’’ [12] organized around bipolar II, in which we submit that cyclothymic temperamental instability represents the common substrate underlying a ‘‘spectrum’’ of related clinical conditions characterized by a complex combination of mood, anxiety, and impulse-control disorders. Building on previous formulations of this lessthan-manic spectrum [11,23,24], the authors broadly sketch out a pathogenetic perspective; they further submit that eating disorders, especially those that involve bingeing and purging behavior, might belong to this broad spectrum of disorders.

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

715

The bipolar spectrum In the last third of the twentieth century, the unipolar-bipolar distinction of manic-depressive illness, originally conceptualized by Edda Neele [25] and Leonhard [26] and subsequently developed by Angst [27], Perris [28], and Winokur et al [29], has proved to be of great heuristic value for clinical and therapeutic research. This dichotomous approach, however, left undefined many affective conditions in the interface of unipolar and bipolar disorders. Originally, Fieve and Dunner [30] and Dunner et al [31] identified bipolar II patients on the basis of the presence of severe depressive episodes that required hospitalization, alternating with hypomanic periods that did not require hospitalization. Although hospitalization might be considered an artificial criterion for defining the diagnostic threshold for mania, this conceptualization represented an important advance toward the recognition of the large universe of bipolar patients whose excited periods remained at the submanic level. It is of great relevance to the theme of this article that Dunner et al [31] subsequently identified instability of course and suicidality as important features of bipolar II. Klerman [32] extended this distinction to subsume bipolar III patients, persons with hypomanic or manic episodes caused by pharmacotherapy (especially antidepressants). Other terms used to designate these conditions include ‘‘Dm’’ for bipolar disorders with mainly depressive presentation [33], and Unipolar-L [34] or ‘‘pseudounipolar depression’’ [35], which raises the provocative possibility that some unipolar patients could be related to bipolar disorder on the basis of pharmacologic response to lithium carbonate [36] and perhaps other mood stabilizers. A broader conceptualization of the ‘‘soft bipolar spectrum’’ [10,12,22] modifies the foregoing definitions of bipolar II by incorporating depressions with hypomanic episodes, both protracted and brief in duration, cyclothymic and hyperthymic traits, and those with familial bipolarity. More recently, in a more formal proposal of bipolar spectrum [23], bipolarity was categorized into seven different clinical subtypes: type 1/2 (schizobipolar), type I (mania with or without depression), type 11/2 (protracted hypomania with depression), type II (depression with hypomania), type II1/2 (depression with cyclothymia), type III (depression with hypomania associated with antidepressants), type III1/2 (depressive and hypomanic swings associated with stimulants), and type IV (depressions with hyperthymic temperament). With the exception of types 1/2 and I, all the other proposed subtypes in the Akiskal-Pinto scheme [23] can be grouped conveniently around ‘‘bipolar II spectrum disorders,’’ in which the severity of the elated phases never reaches the level of manic or manic mixed states, but remains at clinical or subthreshold hypomanic level. Depressed patients who arise from a hyperthymic temperament (type IV), which represents a more stable trait form of hypomania, largely are excluded from this article because there are few data published on this proposed type of bipolarity.

716

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

The concept of bipolar spectrum is in line with classical descriptions of manic-depressive illness that derive from the work of such authorities as Kraepelin [37] and Kretschmer [38], who wrote about affective states ranging from the severest to the mildest and pass without sharp boundary into the domain of personal predisposition or temperament. These authors described individuals with affective temperaments in whom low-grade affective manifestations of a subdepressive or hypomanic nature—without necessarily reaching a clinical or pathologic level—oscillated over long periods of the lifespan. The concept of a spectrum that extends from psychotic mania to mild depression is also endorsed by the most influential contemporary treatise on the subject [39]. Based on the fact that in bipolar pedigrees defined by mania the most common phenotypic expression of the illness was depressive in nature, it has been suggested [12] that soft bipolarity could account for 4% to 5% of the general population. This spectrum concept of manic-depressive illness recently has been supported by the epidemiologic studies of Angst [40], one of the original researchers whose work was highly influential in promoting the bipolar-unipolar dichotomy. He demonstrated the high prevalence of brief hypomanic episodes below the threshold of the 4 days required in such formal classifications as the DSM-IV. This work persuasively argued for the need to enlarge bipolarity at the subthreshold end of the spectrum. Individuals with these conditions do not seek clinical consultation for hypomania but for their depressive states in association with life disruptions, comorbid anxiety, and alcohol and substance abuse. Angst’s [40] concept is currently the broadest, involving 8.3% of the general population, much of it beyond classic bipolar I. This would redefine the bulk of bipolarity within the realm of the soft bipolar spectrum. Unlike bipolar I, which is defined by discrete and full-blown manic episodes, the spectrum of bipolarity described in this article is characterized by depressive episodes interspersed with varying degrees of affective instability caused by temperamental excesses that manifest in brief hypomania, cyclothymia, and hyperthymia. This article does not cover validation of the entire spectrum of bipolarity, which has been accomplished elsewhere [11,41]. The validity of the general scheme for the bipolar spectrum (see box) is no longer in doubt, because it has received ‘‘consensus’’ approval by a group of US and European psychiatrists from key centers chosen on the basis of their long-standing track record in bipolarity [11]. These authors based their deliberations on a thorough and exhaustive review of the international literature. The reader must note that the validation summarized in the Box 1 goes much beyond DSM-IV and ICD-10 to include, among others, short hypomanias (2 days); mixed manic states with few depressive symptoms (2 depressive symptoms) rather than the full depressive syndrome; recurrent brief hypomanias (1 day); pharmacologically occasioned hypomanias and manias; probably all others that occur during ECT, phototherapy, and sleep deprivation; and most provocatively, preference for the

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

717

1998 International Barcelona ‘‘consensus’’ beyond DSM-IV and ICD-10 on bipolarity • • • • • • •

Mood incongruent bipolar psychosis Bipolar I (mania) Mixed manic state (2 depressive symptoms) Bipolar II (hypomania 2 days) Somatic teatment-induced hypomania or mania (bipolar III) Cyclothymic disorder (borderline personality) Recurrent brief hypomania

Data from Akiskal HS, Bourgeois ML, Angst J, et al. Re-evaluating the prevalence of and diagnostic composition within the broad clinical spectrum of bipolar disorders. J Affect Disord 2000;59(Suppl 1):5s–30s.

diagnosis of cyclothymic and bipolar II disorder in lieu of ‘‘comorbid’’ borderline personality. This article deals with the principal challenge that clinicians encounter in their practice in understanding and diagnosing the varied manifestations of the bipolar II spectrum. What complicates this task is the temperamental instability that underlies the soft bipolar spectrum. The unresolved question is whether the soft forms of the illness with underlying temperamental instability are genetically distinct from lithium-responsive classical mania. What is clearer, as the authors discuss in the remainder of the article, are the distinct features that separate the soft bipolar spectrum from bipolar I. The authors examine these issues along several validating parameters, especially from a temperamental perspective. Is the bipolar II spectrum autonomous? Do patients with bipolar II disorder represent an autonomous type of bipolar illness, or are they a transitory condition between unipolar and full-blown bipolar I disorder? Several lines of evidence concerning phenomenology, pattern of familial illness, natural history, and treatment response support the qualitative distinction between bipolar II and bipolar I patients. In a progress report on bipolar II disorder, William Coryell [42] concluded that bipolar II patients cannot be included with bipolar I without introducing significant heterogeneity. Vieta et al drew the same conclusion [43]. Family studies [44–47] provide data that indicate that (1) bipolar II disorder is most prevalent in the relatives of bipolar II probands, (2) bipolar I disorder is less prevalent in the relatives of bipolar II probands than in the relatives of bipolar I probands, and (3) rates of major affective disorders (bipolar plus unipolar) are higher in relatives of bipolar II probands than in the relatives of bipolar I probands. The last finding tends to contradict the continuum view of unipolar, bipolar II, and bipolar I illness and is

718

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

consistent with the hypothesis that a substantial proportion of bipolar II patients have an illness that is distinct from the familial perspective. Follow-up studies support the same conclusion; in depressed patients with at least a 5-year history of affective illness, bipolar II diagnosis represents a stable condition with a low risk (<5%) to progress to bipolar I over an 11-year period of follow-up [48]. Further evidence of diagnostic stability derives from a treatment study [49] in which none of the bipolar II patients versus more than 50% of bipolar I patients developed manic episodes in the year after lithium discontinuation. Unipolar, bipolar II, and bipolar I disorders have not been discriminated easily on the basis of phenomenology and clinical presentation [41,50]. Propensity for switching is one of the characteristics of bipolar disorder, particularly bipolar II disorder [12,16,22,51]. Recently, instead of hypomanic episodes, mood lability, instability in the clinical picture, and complex temperamental dysregulation have been suggested as the fundamental features of bipolar II disorder [52]. Data from the National Institutes of Mental Health Collaborative Depression Study on ‘‘unipolar’’ patients who switched to bipolar II during 11 years of follow-up are particularly informative in this respect [53]. Early age at onset, recurrent depression, high rates of divorce or separation, high rates of scholastic and job maladjustment, isolated ‘‘antisocial acts,’’ and drug abuse characterized at entry those who proved on prospective follow-up to be bipolar II converters. The index depressive episode also was characterized by such features as phobic anxiety, interpersonal sensitivity, separation anxiety, obsessive-compulsive symptoms, somatization (subpanic symptoms), worsening of symptoms in evening, self-pity, demanding behavior, subjective or overt anger, jealousy, suspiciousness, and ideas of reference. This pattern points to a broad array of ‘‘atypical’’ depressive symptoms with comorbid anxious and impulsive features. Finally, temperamental attributes of ‘‘mood lability,’’ ‘‘energy activity,’’ and ‘‘daydreaming’’ at index interview were highly sensitive (91%) predictors of which depressions prospectively will change to bipolar II; of these, ‘‘mood lability’’ was the most specific (86%). These temperamental attributes—reminiscent of Kretschmer’s [38] description of the ‘‘cycloid temperament’’ [54]—were uncharacteristic of depressed patients who switched to mania. These data indicate considerable distinctness of bipolar II from bipolar I at the trait level. In conclusion, according to clinical experience and an extensive systematic worldwide literature, rather than being defined by major depressive episodes that alternate with clinical or subclinical hypomanic features, bipolar II disorder can be better characterized as a complex affective illness based on affective and biographic instability that derives from an unstable temperamental substrate [53]. The affective dysregulation of bipolar II disorder extends beyond elation and depression to include, among others, such negative affective arousal states as anxiety, panic, irritability, impulsivity, and mood lability [24]. Unfortunately, the formal diagnostic systems (eg, ICD-10 and

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

719

DSM-IV) are symptom oriented and do not consider temperamental dysregulation in clinical evaluation. Regrettably, such patients often are labeled as borderline or other ‘‘erratic’’ personality disorders [55,56], and the link to mood, especially bipolar disorders, is grossly underestimated [18]. This is caused in large part by the DSM-IV [15] decision to make mood lability a pathognomonic feature of borderline personality. The tragedy for the nosology is that Henry et al [57], who demonstrated mood lability to be a common feature of borderline personality and bipolar II disorders, concluded that the two disorders were distinct. This a tragedy, because such a conceptualization robs ‘‘borderline’’ patients from being considered as affectively ill. Diagnostic boundaries of bipolar II spectrum conditions Emerging data from several epidemiologic studies conducted in the United States and abroad have challenged the conservative figures of 1% commonly cited in the literature for bipolar disorder [58–60]. With the inclusion of various hypomanic conditions and submanic symptoms, the rates have increased to 5% to 8% [40,61–63]. Although individuals within the bipolar II spectrum represent the most common bipolar phenotype [47], they are often unrecognized, poorly researched, and typically mismanaged. The population prevalence of the bipolar spectrum is summarized in Table 1 The narrow figures (approximately 1%) comprise primarily bipolar I, and the larger ones (appoximately 5%) comprise the soft bipolar spectrum. Much research has been conducted on the clinical prevalence of bipolar II disorder among patients who present with major depressive disorder to various clinical settings worldwide. What emerges is that from 27% to 62% of all major depressions conform to the features of bipolar II or its variants [11]. It is noteworthy that the data are not limited to academic centers that specialize in mood disorders [12,50] but include at least two large outpatient psychiatric private practice settings [64,65] and one family practice clinic [66]. The Clinical Epidemiology of Depression (EPIDEP) database [67], which derives from a national sample from various hospital and ambulatory Table 1 Lifetime prevalence rates of bipolar and bipolar spectrum disordersa Author (year/country)

Rate (%)

Regier et al (1988/USA) [59] Kessler et al (1994/USA) [58] Lewinsohn et al (1995/USA) [62] Weissman et al (1996/cross national) [60] Szadoszky et al (1998/Hungary) [63] Angst et al (1998/Switzerland) [40] Judd and Akiskal (2002/USA) [61]

1.2 1.6 5.7 0.3–1.5 5.0 8.3 6.3

a The higher prevalence rates reflect more explicit diagnostic approaches to bipolar II and related conditions.

720

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

settings in France, has provided the most compelling data on the high prevalence of bipolar II disorder among major depressive patients. The study indicated that at index interview 22% of nearly 500 major depressive patients could be diagnosed as DSM-IV bipolar II; 1 month later, upon re-interview, 40% of patients were diagnosed within the bipolar II spectrum on the basis of more in-depth evaluation and collateral information from significant others and observed hypomania by the clinician. This finding suggests that the diagnosis of bipolar II disorder cannot be made reliably in cross-section, but the probability of observing the phenomenon is increased by repeated evaluations and interviews of significant others. The identification of hypomania is critical for the definition of bipolar II spectrum conditions. Unfortunately, the criteria for hypomanic episodes as described in DSM-IV are insufficiently distinct from those for mania. The 4-day threshold of DSM-IV is unjustified and unnecessarily narrows down the range of bipolar spectrum disorders diagnosable in clinical and epidemiologic studies. An extensive literature based on clinical and community samples validates the definition of hypomania at a duration threshold of 2 rather than 4 days [11], reporting a modal duration of 1 to 3 days. What further complicates the diagnosis of bipolar II disorder is the fact that hypomania often can serve an adaptive function [68,69] and is not associated—by DSM-IV definition—with major life disruptions and psychotic features. Another important question for clinical practice pertains to hypomania that first manifests upon pharmacologic treatment with antidepressants; DSM-IV and ICD-10 have denied bipolar status to these patients. This is an unfortunate decision, because an extensive literature strongly supports the inclusion of such patients within the rubric of bipolar disorders [70–77]. In prospective observation, nearly all adult patients with antidepressantassociated hypomanic episodes progress months or years later to bipolar states with spontaneous hypomania or mania [71]; this also seems to be true for adolescent depressives [75,78]. Antidepressant-associated hypomania is not limited to major depressions. It can occur in dysthymic patients [79,80] and social phobic [81], obsessive-compulsive [82], and other anxiety states [83]. In brief, the depressive phase of bipolar II patients can, in a significant minority of patients, be replaced by subthreshold depressive, socially anxious, or obsessive inhibitions. Because hypomania in itself rarely leads individuals to seek treatment, the patient usually presents with major depression (often with atypical features), anxiety states, bulimia, substance abuse, and personality dysfunction [52,84], and on further inquiry, history of hypomanic episodes is elicited. The diagnosis of bipolar II disorder depends on how rigorously the clinician pursues questions about past hypomania and, most importantly, whether relatives are interviewed. Otherwise, unless recorded in the patient’s past psychiatric chart, the examining clinician may have little indication from the patient’s index cross-sectional depressive mental state about the bipolar elements in the patient’s history.

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

721

The diagnosis of bipolar II depression is crucial, not only for therapeutic implications but also for prognostic reasons. These patients are at high risk for relapse and multiple episodes [43,85] and show a higher rate of suicide attempts and completed suicides than persons with bipolar I or unipolar disorder [31,86]. Bipolar II depressions typically have an early onset, tend to be chronic, have atypical features, and are interspersed with mild hypomanias and infrequent euthymic periods [84,85,87,88]. Although depressive symptomatology is usually mild to moderate in severity, bipolar II patients function more poorly than expected [85], which seems to be related to a high frequency of coexisting nonaffective psychopathologic conditions, such as panic disorder, phobias, alcohol and drug abuse, eating disorders, and personality disorders [24,52]. Dunner and Tay [89] found that clinicians who are specifically trained to recognize bipolar II disorder outperformed such structured instruments as the Schedule for Affective Disorders and Schizophrenia (SADS) or the Structured Clinical Interview for DSM (SCID) in the diagnosis of bipolar II disorder. This methodologic point supports earlier recommendations made by Akiskal et al [16] that the diagnosis of hypomania among cyclothymic subjects should be based on repeated expert interview. Although this point goes against the grail in the literature on structured interviewing, it is consistent in suggesting that the proper identification of bipolar II disorder requires a more sophisticated clinical approach in interviewing and diagnosis. Recent studies also suggest that intense activation in psychomotor and social and professional spheres may better define clinical hypomania than elated mood [90]. The temperamental foundation of the bipolar II spectrum The concept of ‘‘affective temperament’’ derives from Greco-Roman and continental European psychiatry [37] and refers to specific constitutionally based affective dispositions (ie, melancholic-dysthymic, choleric-irritable, sanguine-hyperthymic, and cyclothymic). It is a dimensional construct, which only in its extremes can be considered to be abnormal in a statistical, and perhaps clinical, sense. Kraepelin [37] described the cyclothymic disposition as one of the constitutional substrates from which manic-depressive illness arose. Kretschmer [38] went one step further and proposed that this constitution represented the core characteristic of the illness. In some cyclothymes, depressive or irritable moodiness predominates; in others, subclinical hypomanic traits (hyperthymic temperament) are more characteristic [68]. In the modern literature, criteria for cyclothymic temperament have been operationalized at Tennessee [68]. The bipolar nature of cyclothymic temperament is confirmed by the propensity to switch to hypomania or mania or both on antidepressants and family history of bipolar disorders [16]. Although largely neglected by contemporary psychiatry and clinical psychology, several large-scale studies actually have examined affective

722

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

temperaments and their variants. One such study on cyclothymia, which was conducted in college students in Albany, New York, reported a prevalence rate of 4% to 6% [17]. Cyclothymic traits were significantly more represented among the offspring of persons with bipolar disorders versus control subjects, and many of these cyclothymic students developed depressive and suicidal states and substance abuse during prospective observation, which clearly pointed to a strong bipolar diathesis [87,91]. The psychometric validation of the interview version of the Tennessee criteria for cyclothymia has come through an Italian student population study [92,93]. These criteria are summarized in the following box. Because cyclothymia is so central to the definition, measurement, and pathogenesis of the soft bipolar II spectrum discussed in this article, the authors provide proposed self-rated criteria for its diagnosis under development through the efforts of the International Mood Center (TEMPS-A, Temperament Assessment of Memphis, Pisa, Paris, and San Diego, auto-questionnaire version) [94] (see Appendix 1). Whether ‘‘extreme temperamental traits’’ could predispose individuals to the development of emotional and behavioral problems during their lifespan has been explored in a nonclinical sample of 1010 students without major psychiatric disorders (collaboration of the authors with Signoretta et al, unpublished data). Cyclothymic subjects reported the highest number of problems. In particular, a cyclothymic disposition was most frequently associated with anxiety-sleep disturbances, sensitivity to separation, eating disturbances in women, and antisocial-aggressive behavior in men. These findings are consistent with the observation that the cyclothymic tempera-

The cyclothymic temperament • Biphasic mood swings—abrupt shifts from one phase to the other, each phase lasting for a few days at a time with infrequent interphase euthymia • At least four of the following features constitute the habitual long-term baseline of the subject: 1. lethargy alternating with eutonia 2. shaky self-esteem alternating between low self-confidence and overconfidence 3. decreased verbal output alternating with talkativeness 4. mental confusion alternating with sharpened and creative thinking 5. unexplained tearfulness alternating with excessive punning and jocularity 6. introverted self-absorption alternating with uninhibited people-seeking Data from Akiskal et al [68] criteria-validated in a population study [92,93].

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

723

ment is associated with lifetime personality dysfunction [16] and, in a high proportion of cases, with anxiety and impulsivity [52]. Affective instability characterized by trait and mood lability is the best predictor of which depressives will switch to bipolar II disorder [53]. When affective oscillations are extreme and associated with significant disruption and interpersonal conflict, many cyclothymic individuals also meet criteria for so-called ‘‘borderline’’ and other erratic personality disorders. Studies that focused on borderline personality cohorts also have reported high rates of cyclothymic [95] and soft bipolar spectrum diagnoses [19]. In a German study [96] that rigorously rated ‘‘subaffective personality disorders,’’ borderline and irritable-cyclothymic conditions overlapped considerably. As expected, borderline personality has been found to be a predictor of pharmacologic hypomania [55,97]. In further exploring the putative temperamental foundations and clinical characteristics of bipolar II disorder, atypical depression represents an intriguing challenge. The distinctive features of atypical depression are the presence of reactive mood, marked asthenia, reverse vegetative symptoms, such as increased appetite, hypersomnia, and diurnal variation with evening worsening, and a trait of exaggerated vulnerability to feeling hurt by criticism or rejection (interpersonal sensitivity). The constructs of bipolar II, cyclothymic, and atypical depressions overlap widely, and it is difficult to separate the shifting affective symptomatology of these patients from their long-term temperamental and characterologic attributes. The literature suggests considerable commonality in the clinical symptoms [98,99], long-term traits [87,100–102], and preferential response to monoamine oxidase inhibitors [103–106] between atypical and cyclothymic-bipolar II depressions. The authors have observed that 72% of 86 major depressive patients with atypical features, as defined by DSM-IV, meet criteria for bipolar II and related soft bipolar disorders [52]. Nearly 60% had antecedent cyclothymic or hyperthymic temperaments. The finding of high rates of bipolarity— defined either by hypomanic episodes or by cyclothymic/hyperthymic temperaments—in atypical depressives is consistent with the prospective observation of unipolar major depressives switching to a bipolar II course [53]. The authors wish to point out that they are not the only research team reporting high rates of bipolar II disorders in atypical depressives [84,88,107,108]. In patients with bipolar II spectrum disorders and atypical depression, lifetime comorbidity with anxiety disorders, particularly panic disorderagoraphobia [52,109], bulimia nervosa [110,111], body dysmorphic disorder [35,112], alcohol and substance abuse disorder [52,113] (Table 2), and cluster C (anxious) and cluster B (dramatic) personality disorders, is the rule rather than the exception. In particular, a large proportion of these patients meets DSM-IV criteria for borderline personality disorder [52,56]. An analysis of the explanatory power of affective temperaments and personality disorders for each of the criteria of borderline personality disorder [24] revealed that

724

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

Table 2 Lifetime axis I ‘‘comorbidity’’ in major depression with atypical features Axis I disorder

Rate (%)

Bipolar II Panic-agoraphobia Social phobia Body dysmorphic disorder Bulimia nervosa Substance and alcohol abuse

72 51 30 42 16 33

Data from Perugi G, Akiskal HS, Lattanzi L, et al. The high prevalence of soft bipolar (II) features in atypical depression. Compr Psychiatry 1998;39:63–71.

the presence of cyclothymic temperament explains much of the relationship between atypical depression and borderline personality disorder. These data suggest that the diagnosis of borderline personality disorder in patients with atypical depression is favored by the coexistence of an affective cyclothymic temperamental dysregulation and anxious comorbidity, which clinically manifests in a mood disorder subtype that seems to be preponderantly in the realm of the bipolar II spectrum. These findings overall support the hypothesis that borderline characterologic features are related to the mood instability of the cyclothymic type that coexists with anxious-dependent traits [16,19,95]. The authors find no reason to separate bipolar II with cyclothymic instability from the stable instability of the borderline type [57], because mood lability is a common characteristic of both sets of disorders [53]. Correlational analysis [24] suggests that in bipolar II disorder, atypical depressives’ mood reactivity and interpersonal sensitivity traits might be related constructs with a cyclothymic temperamental matrix. Interpersonal sensitivity and usual mood reactivity seem to represent two strongly related temperamental features, which represent different aspects (cognitive and affective) of the same psycho(patho)logic dimension. On the other hand, the presence of cyclothymic traits also seems to be related to interpersonal sensitivity and extreme mood reactivity, which suggests a common temperamental background among these constructs. Recent epidemiologic data [114] indicate that the more enduring atypical depressive features are mood reactivity and rejection sensitivity. The inclusion of stable temperamental traits for an axis I diagnosis increases the validity and the continuity of the syndrome. The overlap of the first two axes of the DSM system suggests that to systematically take into account combined trait and state features may be useful to increase the validity of current diagnostic categories. This is particularly true for affective disorders, which usually undergo a complex development during the patients’ entire lifetime, influencing cognitive, emotional, and behavioral functioning even during relatively ‘‘asymptomatic’’ phases. Inadequate attention to this trait-state relationship represents one of the major limitations of the current diagnostic systems. The introduction of a separate axis II for personality disorders [15], based on the assumption of a relative independence between mental disorders and personality traits,

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

725

cannot resolve this problem. It actually seems to have broadened the gulf between affective and personality disorders (and their respective researchers!). In recent years an emerging array of evidence seems to indicate that the affective temperamental dispositions might influence the clinical expression of mixed and manic states [82,115,116]. For instance, the authors recently observed that the depressive temperament is associated with the presence of depressive features during mania [117,118]. Although a full discussion of this topic is beyond the scope of this article, in an analogous fashion the authors would like to entertain a hypothesis for depressive mixed states [12]. Hyperthymic temperament might be the substrate of agitated depression, depressive-anancastic disposition might be related to the presence of melancholic features, and cyclothymic instability could represent the background of bipolar II-atypical depression. More specifically, according to this hypothesis mood reactivity and interpersonal sensitivity might be interpreted as the persistence of the cyclothymic instability during the depressive phases [53]. Cyclothymic-bipolar II-atypical depressive patients display a long-lasting ‘‘stable’’ hyperreactivity to many psychological (ie, rejection, separation) and physical (ie, food, light, drugs) stimuli, which can persist during depressive phases. This marked reactivity of mood also could explain the frequent concomitance of impulse control disorder and substances and alcohol abuse. The presence of a long-lasting cyclothymic-anxious-sensitive disposition seems to constitute the common background of the complex syndromic pattern of anxiety, mood, and impulsive disorders that patients in this realm display during their adolescence and young adulthood [52]. The bipolar II spectrum, impulse control, and eating disorders connection The foregoing discussion has considered what can be categorized within a broad spectrum of bipolar II disorder and extends into the realm of anxiety, impulse control, eating and personality disorders, and substance abuse. This formulation is diagrammed in Fig. 1, in which the light bulb is a schematic representation of the source of the bipolar II spectrum and its temperament: The light goes across the prism of environmental stressors and results in the different expressions of the soft bipolar spectrum disorders. It is beyond the scope of this article to consider the relationship between bipolar II and childhood and adolescent externalizing disorders, such as conduct disorder and attention-deficit hyperactivity disorder, and their adult counterparts, antisocial personality and substance abuse. History for childhood hyperactivity is more common in adults with bipolar disorder than adults with unipolar disorder [119]. We have not considered in requisite depth alcohol, opioid, and stimulant abuse comorbidity [120–122], which should be evaluated carefully for bipolar II spectrum diagnoses. Regrettably, the DSM-IV conventions tend to favor the diagnosis of alcohol- and substanceinduced mood disorders at the expense of bipolar II disorder.

726

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

Fig. 1. The soft bipolar spectrum.

McElroy et al [123] have drawn attention to the possible bipolar nature of some impulse-ridden behaviors, such as those in the realm of sexual and aggression control, paraphilias, and gambling. Impulse control disorders (ICDs), which have been related to bipolar disorder on the basis of phenomenology, comorbidity, abnormalities of serotonergic neurotransmission, and response to thymoleptic agents and mood stabilizers [124–129], have many affinities with bipolar disorder. Both are characterized by harmful, dangerous, or pleasurable ego-syntonic behaviors, impulsivity, impaired insight, and similar affective instability. Patients with comorbid ICDs and bipolar disorder reported a strong relationship and similarities between their ICD, manic, and depressive symptomatology [126,127]. ICDs and bipolar disorder display a similar course of illness in that both conditions often begin in adolescence or early adulthood and subsequently have an episodic or chronic course. Comorbidity data provide further support for a strong relationship between ICDs and bipolar disorder. Elevated rates of bipolar disorder have been found in patients with ICDs and vice versa [127]. Available studies also indicate that ICDs and bipolar disorder have similar comorbidity patterns with other mental disorders, including alcohol and substance use, anxiety, and eating disorders [125,126,129,130]. The proposed bipolar II-ICD relationship suggests important theoretical implications. In the authors’ bipolar II spectrum model, cyclothymic mood reactivity and impulsivity might be related, each characterized by environmentally induced periods of disinhibited or facilitated thinking and behav-

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

727

ior, poor insight or ego-syntonicity, and marked changes in mood state between dysphoric and pleasurable affects. In this sense, eating disorders, especially those related to the presence of bulimic behavior, such as anorexia nervosa binge-eating/purging type, bulimia nervosa, and binge-eating disorder, can be considered as putative variants of ICDs. The frequent association of eating disorders and affective disorders is well documented [131,132]. Most studies on this topic have focused on the concomitance between eating disorders and unipolar depression [133]. The literature on the comorbidity with bipolar disorder is sparser. Several case series [134–137] and family studies [7,138–140] of eating disorder patients have suggested that there may be a relationship between eating disorders and bipolar disorder, in particular bipolar II. This association might be more frequent in bulimic patients who present with a severe chronic type of eating disorder [141]. The relatives of bulimic patients also display an increased risk for bipolar morbidity [138–140,142]. Few data are available concerning the therapeutic implications of the association of bulimia nervosa and bipolar disorder. Only case reports of concomitant remission of both disorders with lithium salts or anticonvulsant agents have been published [136,143]. Greenberg and Harvey [138] reported that the combination of biphasic mood shifts and dietary restraint was a better predictor of severity of binge eating in a nonclinical population of female college students than was the combination of depression and dietary restraint. In another study [139], increased ‘‘hypomania’’ was found in bulimic but not unipolar depressed patients. A well-controlled study [142] investigated the lifetime prevalence of affective disorders in bulimics, recovered bulimics, patients with major depression, and nonpsychiatric controls. Subjects were interviewed with the Diagnostic Interview Schedule: 12% of current bulimics and 11% of recovered bulimics had a history of bipolar affective disorder, compared with 0% of both control groups. The concomitance between eating disorders and bipolar II disorder is probably underestimated in epidemiologic and clinical studies conducted with structured interviews. In this regard, the low diagnostic reliability of the bipolar II category should be considered a major methodologic problem; the low Kappa value for hypomania is caused by low sensitivity rather than low specificity [144], which makes false-negative results common and falsepositive results virtually nonexistent. Although the comorbidity of bulimia and bipolar disorder does not seem coincidental, the precise relationship between these two disorders is not clear. The possible relationships between seasonal affective disorders and bulimia nervosa have been suggested recently by some epidemiologic studies [145], which demonstrates that bulimia may display seasonal variations with winter worsening of bulimic symptoms. Eating disorder symptoms are present in winter depression and bulimia nervosa [146]. Epidemiologic data suggest that winter depression is most frequently part of a bipolar II disorder [110,111]. Considerable nonaffective psychopathologic conditions have

728

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

been reported in patients with eating disorders, especially in persons with bulimia. These conditions include alcohol or substance abuse or dependence, panic disorder, phobias, and obsessive-compulsive symptoms [141]. Such a wide array of comorbidity is seen not only in eating disorder populations but also in bipolar II patients without eating disorders [35,52,147–149]. Many patients with eating disorders report bingeing in an attempt to improve their mood, which may decrease anxiety and depression, but the benefits are short-lived [150]. Patients with bipolar II disorder are unusually reactive to changes in their environment, and they often attempt to heighten their low mood with food, alcohol, drugs, exercise, or sexual activity in a more or less unconscious way. These abnormal behaviors could be seen as developing into persistent addictive states or abnormally motivated behaviors under the influence of the mood disorder. DePaulo et al [147] and Gershon et al [151] have found that only rarely in family studies of affective disorder do the probands or their affectively ill relatives suffer from clinically defined syndromal eating disorders, although many bipolar II patients overeat and gain weight when depressed. The finding of low rates of eating disorders in family studies of bipolar patients suggests that the mood disorder is not the only explanation for the vulnerability to eating disorders. Because the mood disorder often precedes the eating disorders in patients with both disorders, a high prevalence of bipolar II disorders in different populations with eating disorders suggests an important role of the mood disorder in the initiation and maintenance of the disordered eating behavior. It would be relevant to study formally the temperament of eating disorder patients to find whether cyclothymic-sensitive traits represent contributory factors, just as is the case for the bipolar spectrum.

Summary The bipolar II spectrum represents the most common phenotype of bipolarity. Numerous studies indicate that in clinical settings this soft spectrum might be as common—if not more common than—major depressive disorders. The proportion of depressive patients who can be classified as bipolar II further increases if the 4-day threshold for hypomania proposed by the DSM-IV is reconsidered. The modal duration of hypomanic episodes is 2 days; highly recurrent brief hypomania is as short as 1 day, and when complicated by major depression, it should be classified as a variant of bipolar II. Another variant of the bipolar II pattern is represented by major depressive episodes superimposed on cyclothymic or hyperthymic temperamental characteristics. The literature is unanimous in supporting the idea that depressed patients who experience hypomania during antidepressant treatment belong to the bipolar II spectrum. So-called alcohol- or substance-induced mood disorders may have much in common with bipolar II spectrum disorders, in particular when mood swings outlast detoxification.

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

729

Finally, many patients within the bipolar II spectrum, especially when recurrence is high and the interepisodic period is not free of affective manifestations, may meet criteria for personality disorders. This is particularly true for cyclothymic bipolar II patients, who are often misclassified as borderline personality disorder because of their extreme mood instability. Subthreshold mood lability of a cyclothymic nature seems to be the common thread that links the soft bipolar spectrum. The authors submit this to represent the endophenotype likely to be informative in genetic investigations. Mood lability can be considered the core characteristics of the bipolar II spectrum, and it has been validated prospectively as a sensitive and specific predictor of bipolar II outcome in major depressives. In a more hypothetical vein, cyclothymic-anxious-sensitive temperamental disposition might represent the mediating underlying characteristic in the complex pattern of anxiety, mood, and impulsive disorders that bipolar II spectrum patients display throughout much of their lifetimes. The foregoing conclusions, based on clinical experience and the research literature, challenge several conventions in the formal classificatory system (ie, ICD-10 and DSM-IV). The authors submit that the enlargement of classical bipolar II disorders to include a spectrum of conditions subsumed by a cyclothymic-anxious-sensitive disposition, with mood reactivity and interpersonal sensitivity, and ranging from mood, anxiety, impulse control, and eating disorders, will greatly enhance clinical practice and research endeavors. Prospective studies with the requisite methodologic sophistication are needed to clarify further the relationship of the putative temperamental and developmental variables to the complex syndromic patterns described herein. The authors believe that viewing these constructs as related entities with a common temperamental diathesis will make patients in this realm more accessible to pharmacologic and psychological approaches geared to their common temperamental attributes. The authors submit that the use of the term ‘‘spectrum’’ is distinct from a simple continuum of subthreshold and threshold cases. The underlying temperamental dimensions postulated by the authors define the disposition for soft bipolarity and its variations and dysregulation in anxious disorders and dyscontrol in appetitive, mental, and behavioral disorders, much beyond affective disorders in the narrow sense.

References [1] Halliday D, Resnick R, Walker J. Fundamentals of physics. 5th Edition. New York: John Wiley & Sons; 1997. [2] Winokur G. Depression spectrum disease: description and family study. Compr Psychiatry 1972;13:3–8. [3] Baron M, Gruen R, Rainer JD, et al. A family study of schizophrenic and normal control probands: implications for the spectrum concept of schizophrenia. Am J Psychiatry 1985;142:447–55.

730

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

[4] Kendler KS, Neale MC, Walsh D. Evaluating the spectrum concept of schizophrenia in the Roscommon Family Study. Am J Psychiatry 1995;152:749–54. [5] Kety SS, Rosenthal D, Wender PH, et al. The types and prevalence of mental illness in the biological and adoptive families of adopted schizophrenics. J Psychiatr Res 1968; 6:345–62. [6] Siever LJ, Silverman JM, Horvath TB, et al. Increased morbid risk for schizophreniarelated disorders in relatives of schizotypal personality disordered patients. Arch Gen Psychiatry 1990;47:634–40. [7] Hudson JI, Pope HG. Affective spectrum disorder: does antidepressant response identify a family of disorders with a common pathophysiology? Am J Psychiatry 1991;148:548. [8] Ross CA, Anderson G, Fraser GA, et al. Differentiating multiple personality disorder and dissociative disorder not otherwise specified. Dissociation Progress in the Dissociative Disorders 1992;5:87–90. [9] Shoda H. Splitting phenomena from a viewpoint of experiencing time: Spectrum from multiple personality and hysteria to borderline personality disorder. Psychopathology 1993;26:240–54. [10] Akiskal HS. The bipolar spectrum: new concepts in classification and diagnosis. In: Grinspoon L, editor. Psychiatry update: the American Psychiatric Association annual review, Vol. II. Washington, DC: Am Psychiatr Press; 1983. p. 271–92. [11] Akiskal HS, Bourgeois ML, Angst J, et al. Re-evaluating the prevalence of and diagnostic composition within the broad clinical spectrum of bipolar disorders. J Affect Disord 2000;59(Suppl 1):5s–30s. [12] Akiskal HS, Mallya G. Criteria for the ‘‘soft’’ bipolar spectrum: treatment implications. Psychopharmacol Bull 1987;23:68–73. [13] Hollander E. Obsessive compulsive spectrum disorders: an overview. Psychiatric Annals 1993;23:355–8. [14] McElroy SL, Phillips KA, Jecj PE. Obsessive compulsive spectrum disorder. J Clin Psychiatry 1994;55(Suppl 10):33–51. [15] American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4th edition. Washington, DC: American Psychiatric Association Press; 1994. [16] Akiskal HS, Djenderedjian AM, Rosenthal RH, et al. Cyclothymic disorder: validating criteria for inclusion in the bipolar affective group. Am J Psychiatry 1977;134:1227–33. [17] Depue RA, Slater JF, Wolfsetter-Kausch H. A behavioral paradigm for identifying persons at risk for bipolar disorder: a conceptual framework and five validation studies. J Abnorm Psychol 1981;90:381–437. [18] Gunderson JG, Phillips KA. A current view of the interface between borderline personality disorder and depression. Am J Psychiatry 1991;148:967–75. [19] Deltito J, Martin L, Riefkohl J, et al. Do patients with borderline personality disorder belong to the bipolar spectrum? J Affect Disord 2001;67:221–28. [20] Cassano GB, Michelini S, Shear K, et al. The panic-agoraphobic spectrum: a descriptive approach to the assessment and treatment of subtle symptoms. Am J Psychiatry 1997;154(Suppl):27–38. [21] Alarcon RD, Rippetoe PA, Walter-Ryan WG. A spectrum is not a continuum. Am J Psychiatry 1986;143:678. [22] Akiskal HS. The prevalent clinical spectrum of bipolar disorders: beyond DSM-IV. J Clin Psychopharmacol 1996;17(Suppl 3):117s–22s. [23] Akiskal HS, Pinto O. The evolving bipolar spectrum: prototypes I, II, III, and IV. Psychiatr Clin N Am 1999;22:517–34. [24] Perugi G, Toni C, Travierso M, et al. The role of cyclothymia in atypical depression: toward a data-based reconceptualization of the borderline-bipolar II connection. J Affect Disord 2002;(in press). [25] Neele E. Die phasischen psychosen nach ihrem erscheinungs-und erbbild. Leipzig: Barth; 1949.

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

731

[26] Leonhard K. The classification of endogenous psychoses. New York: Irvington Publishers, Inc.; 1979. [27] Angst J. The etiology and nosology of endogenous depressive psychoses. Foreign Psychiatry 1973;217. [28] Perris C. A study of bipolar (manic-depressive) and bipolar recurrent depressive psychoses. Acta Psychiatr Scand 1966;42(Suppl 194):153–94. [29] Winokur G, Clayton PJ, Reich T. Manic depressive illness. St. Louis: CV Mosby; 1969. [30] Fieve RR, Dunner DL. Unipolar and bipolar affective states. In: Flach FF, Draghi SS, editors. The nature and treatment of depression. New York: John Wiley & Sons; 1975. p. 145–60. [31] Dunner DL, Gershon ES, Goodwin FK. Heritable factors in the severity of affective illness. Biol Psychiatry 1976;11:31–42. [32] Klerman GL. The spectrum of mania. Compr Psychiatry 1981;22:11–20. [33] Angst J, Frey R, Lohmeyer B, et al. Bipolar manic-depressive psychoses: results of a genetic investigation. Hum Genet 1980;55:237–54. [34] Kupfer DJ, Pickar D, Himmelhoch JM, et al. Are there two types of unipolar depression? Arch Gen Psychiatry 1975;32:866–71. [35] Perugi G, Giannotti D, Di Vaio S, et al. Fluvoxamine in the treatment of body dysmorphic disorder (dysmorphophobia). Int Clin Psychopharmacol 1996;11:247–54. [36] Bowden CL. Lithium-responsive depression. Compr Psychiatry 1978;19:224–31. [37] Kraepelin E. Manic-depressive insanity and paranoia. Edinburgh: ES Livingstone; 1921. [38] Kretschmer E. Physique and character. London: Kegan, Paul, Trench, Trubner and Co, Ltd; 1936. [39] Goodwin FK, Jamison KR. Manic-depressive illness. New York: Oxford University Press; 1990. [40] Angst J. The emerging epidemiology of hypomania and bipolar II disorder. J Affect Disord 1998;50:143–51. [41] Akiskal HS. Classification, diagnosis and boundaries of bipolar disorders. In: Maj M, Akiskal HS, Lopez-Ibor JJ, Sartorius N, editors. Bipolar disorder. London: John Wiley & Sons; 2002. p. 1–52. [42] Coryell W. Bipolar II disorder: a progress report. J Affect Disord 1996;41:159–62. [43] Vieta E, Gasto` C, Otero A, et al. Differential features between bipolar I and bipolar II disorder. Compr Psychiatry 1997;38:98–101. [44] Coryell W, Endicott J, Andreasen N, et al. A family study of bipolar II disorder. Br J Psychiatry 1984;145:49–54. [45] Gershon ES, Hamovit J, Guroff JJ, et al. A family study of schizoaffective, bipolar I, bipolar II, unipolar and control probands. Arch Gen Psychiatry 1982;39:1157–67. [46] Heun R, Maier W. Bipolar II disorders in six first-degree relatives. Biol Psychiatry 1993;34:274–6. [47] Simpson SG, Folstein SE, Meyers DA, et al. Bipolar II: the most common bipolar phenotype? Am J Psychiatry 1993;150:901–3. [48] Coryell W, Endicott J, Maser JD, et al. Long-term stability of polarity distinctions in the affective disorders. Am J Psychiatry 1995;152:385–90. [49] Faedda GL, Tondo L, Baldessarini RJ, et al. Outcome after rapid versus gradual discontinuation of lithium treatment in bipolar disorders. Arch Gen Psychiatry 1993;50:448–55. [50] Cassano GB, Akiskal HS. Savino, et al. Proposed subtypes of bipolar II and related disorders: with hypomanic episodes (or cyclothymia) and with hyperthymic temperament. J Affect Disord 1992;26:127–40. [51] Bunney WE Jr, Goodwin FK, Murphy DL. The switch process in manic-depressive illness II: relationship to catecholamines, REM sleep, and drugs. Arch Gen Psychiatry 1972;27:304–9. [52] Perugi G, Akiskal HS, Lattanzi L, et al. The high prevalence of soft bipolar (II) features in atypical depression. Compr Psychiatry 1998;39:63–71.

732

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

[53] Akiskal HS, Maser JD, Zeller P, et al. Switching from ‘‘unipolar’’ to ‘‘bipolar II’’: an 11-year prospective study of clinical and temperamental predictors in 559 patients. Arch Gen Psychiatry 1995;52:114–23. [54] Brieger P, Marneros A. Dysthymia and cyclothymia: historical origins and contemporary development. J Affect Disord 1997;45:117–26. [55] Akiskal HS, Chen SE, Davis GC, et al. Borderline: an adjective in search of a noun. J Clin Psychiatry 1985;46:41–8. [56] Soloff PH, George A, Nathan RS, et al. Characterizing depression in borderline patients. J Clin Psychiatry 1987;48:155–7. [57] Henry C, Mitropoulou V, New A, et al. Affective instability and impulsivity in borderline personality and bipolar II disorders: similarities and differences. J Psychiatr Res 2001;35:307–12. [58] Kessler RC, McGonagle KA, Zhao S, et al. Lifetime and 12-month prevalence of DSMIII-R psychiatric disorders in the United States: results from the National Comorbidity Survey. Arch Gen Psychiatry 1994;51:8–19. [59] Regier DA, Boyd JH, Burke JD Jr, et al. One-month prevalence of mental disorders in the United States: based on five epidemiologic catchment area sites. Arch Gen Psychiatry 1988;45:977–86. [60] Weissman MM, Bland RC, Canino GJ, et al. Cross-national epidemiology of major depression and bipolar disorder. JAMA 1996;276:293–9. [61] Judd LL, Akiskal HS. The prevalence and disability of bipolar spectrum disorders in the U.S. population: re-analysis of the ECA database taking into account subthreshold cases. J Affect Disord, in press. [62] Lewinsohn PM, Klein DL, Seeley JR. Bipolar disorders in a community sample of older adolescents: prevalence, phenomenology, comorbidity, and course. J Am Acad Child Adolesc Psychiatry 1995;34:454–63. [63] Sza´do´czky E, Papp Z, Vitrai J, et al. The prevalence of major depressive and bipolar disorder in Hungary. J Affect Disord 1998;50:155–62. [64] Benazzi F. Prevalence of bipolar II disorder in outpatient depression: a 203-case study in private practice. J Affect Disord 1997;43:163–66. [65] Koukopoulos A, Reginaldi D, Laddomada P, et al. Course of the manic-depressive cycle and changes caused by treatment. Pharmacopsychiatry 1980;13:156–67. [66] Manning JS, Haykal RF, Connor PD, et al. On the nature of depressive and anxious states in a family practice setting: the high prevalence of bipolar II and related disorders in a cohort followed longitudinally. Compr Psychiatry 1997;38:102–8. [67] Hantouche EG, Akiskal HS, Lancrenon S, et al. Systematic clinical methodology for validating bipolar-II disorder: data in mid-stream from a French national multi-site syudy (EPIDEP). J Affect Disord 1998;50:153–73. [68] Akiskal HS, Khani MK, Scott-Strauss A. Cyclothymic temperamental disorders. Psychiatr Clin N Am 1979;2:527–54. [69] Jamison KR, Gerner RH, Hammen C, et al. Clouds and silver linings: positive experiences associated with primary affective disorders. Am J Psychiatry 1980;137: 198–202. [70] Akiskal HS, Rosenthal RH, Rosenthal TL, et al. Differentiation of primary affective illness from situational, symptomatic, and secondary depressions. Arch Gen Psychiatry 1979;36:635–43. [71] Akiskal HS, Walker PW, Puzantian VR, et al. Bipolar outcome in the course of depressive illness: phenomenologic, familial, and pharmacologic predictors. J Affect Disord 1983; 5:115–28. [72] Altschuler LL, Post RM, Leverich GS, et al. Antidepressant-induced mania and cycle acceleration: a controversy revisited. Am J Psychiatry 1995;152:1130–8. [73] Benazzi F. Antidepressant-associated hypomania in outpatient depression: a 203-case study in private practice. J Affect Disord 1997;46:73–7.

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

733

[74] Post RM, Denicoff KD, Leverich GS, et al. Drug-induced switching in bipolar disorder. CNS Drugs 1997;8:352–65. [75] Strober M, Carlson G. Bipolar illness in adolescents with major depression. Arch Gen Psychiatry 1982;39:549–55. [76] Sultzer DL, Cummings JL. Drug-induced mania: causative agents, clinical characteristics and management. A retrospective analysis of the literature. Med Toxicol Adverse Drug Exp 1989;4:127–43. [77] Wehr TA, Goodwin FK. Can antidepressants cause mania and worsen the course of affective illness? Am J Psychiatry 1987;144:1403–11. [78] Weller EB, Weller RA, Fristad MA. Bipolar disorder in children: misdiagnosis, underdiagnosis, and future directions. J Am Acad Child Adolesc Psychiatry 1995;34: 709–14. [79] Rihmer Z. Dysthymia: a clinician’s perspective. In: Burton SW, Akiskal HS, editors. Dysthymic disorder. London: Gaskell, Royal College of Psychiatrists; 1990. p. 112–25. [80] Rosenthal TL, Akiskal HS, Scott-Strauss A, et al. Familial and developmental factors in characterological depressions. J Affect Disord 1981;3:183–92. [81] Himmelhoch JM. Social anxiety, hypomania and the bipolar spectrum: data, theory and clinical issues. J Affect Disord 1998;50:203–13. [82] Perugi G, Akiskal HS, Pfanner C, et al. The clinical impact of bipolar and unipolar affective comorbidity on obsessive-compulsive disorder. J Affect Disord 1997;46:15–23. [83] Perugi G, Akiskal HS, Ramacciotti S, et al. Depressive comorbidity of panic, social phobic, and obsessive-compulsive disorders re-examined: is there a bipolar II connection? J Psychiatr Res 1999;33:53–61. [84] Benazzi F. Prevalence of bipolar II disorder in atypical depression. Eur Arch Psychiatry Clin Neurosci 1999;249:62–5. [85] Coryell W, Endicott J, Andreasen N, et al. The significance of past mania or hypomania in the course and outcome of major depression. Am J Psychiatry 1987;144:309–15. [86] Rihmer Z, Pestality P. Bipolar II disorder and suicidal behavior. Psychiatr Clin N Am 1999;22:667–73. [87] Akiskal HS, Downs J, Jordan P, et al. Affective disorders in referred children and younger siblings of manic depressives: mode of onset and prospective course. Arch Gen Psychiatry 1985;42:996–1003. [88] Benazzi F, Rihmer Z. Sensitivity and specificity of DMS-IV atypical features for bipolar II disorder diagnosis. Psychiatry Res 2000;93:257–62. [89] Dunner DL, Kai Tay L. Diagnostic reliability of the history of hypomania in bipolar II patients with major depression. Compr Psychiatry 1993;34:303–7. [90] Benazzi F, Akiskal HS. The dual factor structure of self-rated hypomania: energizedactivity versus irritable-thoughts racing. J Affect Disord, in press. [91] Klein DN, Depue RA, Slater JF. Inventory identification of cyclothymia. IX. Validation in offspring of bipolar I patients. Arch Gen Psychiatry 1986;43:441–5. [92] Akiskal HS, Placidi GF, Signoretta S, et al. TEMPS-I: delineating the most discriminant traits of cyclothymic, depressive, irritable and hyperthymic temperaments in a nonpatient population. J Affect Disord 1998;51:7–19. [93] Placidi GF, Signoretta S, Liguori A, et al. The Semi-Structured Affective Temperament Interview (TEMPS-I): reliability and psychometric properties in 1010 14–26 year students. J Affect Disord 1998;47:1–10. [94] Akiskal HS, Perugi G, Hantouche E, et al. The affective temperament scales of Memphis, Pisa, Paris and San Diego: progress towards a self-rated auto-questionnaire version (TEMPS-A). J Affect Disord, in press. [95] Levitt AJ, Joffe RT, Ennis J, et al. The prevalence of cyclothymia in borderline personality disorder. J Clin Psychiatry 1990;51:335–9. [96] Sab H, Herpertz S, Steinmeyer EM. Subaffective personality disorders. Int Clin Psychopharmacol 1993;1(Suppl 1):39–46.

734

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

[97] Levy D, Kimhi R, Barak Y, et al. Antidepressant-associated mania: a study of anxiety disorders patients. Psychopharmacology 1998;136:243–6. [98] Robertson HA, Lam RW, Stewart JN, et al. Atypical depressive symptoms and clusters in unipolar and bipolar depression. Acta Psychiatr Scand 1996;94:421–7. [99] Zisook S, Shuchter SR, Gallagher T, et al. Atypical depression in an outpatient psychiatric population. Depression 1993;1:268–74. [100] Davidson J, Zysook S, Giller E, et al. Symptoms of interpersonal sensitivity in depression. Compr Psychiatry 1989;30:357–68. [101] Derecho CN, Wetzler S, Mcginn LK, et al. Atypical depression among psychiatric inpatients: clinical features and personality traits. J Affect Disord 1996;39:55–9. [102] Liebowitz MR, Klein DF. Hysteroid dysphoria. Psychiatr Clin N Am 1979;2:555–75. [103] Himmeloch JM, Thase ME, Mallinger AG, et al. Tranylcypromine versus imipramine in anergic bipolar depression. Am J Psychiatry 1991;48:910–6. [104] Liebowitz MR, Quitkin FM, Stewart JW, et al. Antidepressant specificity in atypical depression. Arch Gen Psychiatry 1988;45:129–37. [105] Quitkin FM, Stewart JW, McGrath, et al. Phenelzine versus imipramine in the treatment of probable atypical depression: defining syndrome boundaries of selective MAOI responders. Am J Psychiatry 1988;145:306–11. [106] Sotsky SM, Simmens SJ. Pharmacotherapy response and diagnostic validity in atypical depression. J Affect Disord 1999;54:237–47. [107] Ebert D, Barocka A. The early course of atypical depression. Eur Arch Psychiatry Neurosci 1991;241:131–2. [108] Zubieta JK, Pande AC, Demitrack MA. Two year follow up of atypical depression. J Psych Res 1999;33:23–9. [109] Alpert KE, Uebelacker LA, Mclean NE, et al. Social phobia, avoidant personality disorder and atypical depression: co-occurrence and clinical implications. Psychol Med 1997;27:627–33. [110] Levitan RD, Kaplan AS, Levitt AJ, et al. Seasonal fluctuations in mood and eating behavior in bulimia nervosa. Int J Eat Disord 1994;16:255–99. [111] Levitan RD, Rector NA, Bagby RM. Negative attributional style in seasonal and nonseasonal depression. Am J Psychiatry 1998;155:428–30. [112] Phillips KA, Nierenberg AA, Brendel G, et al. Prevalence and clinical features of body dysmorphic disorder in atypical major depression. J Nerv Ment Dis 1996;184:125–9. [113] Horwath E, Johnson J, Klerman GL, et al. Depressive symptoms as relative and attributable risk factors for first-onset major depression. Arch Gen Psychiatry 1992;49:817–23. [114] Angst J, Gamma A, Sellaro R, et al. The validity of atypical depression in the community: results of the Zurich Cohort Study. J Affect Disord 2002;72:125–33. [115] Dell’Osso Placidi GF, Nassi R, Freer P, Cassano GB, Akiskal HS. The manic-depressive mixed state: familial, temperamental and psychopathologic characteristics in 108 female inpatients. Eur Arch Psychiatry Clin Neurosci 1991;240:234–9. [116] Perugi G, Micheli C, Akiskal HS, et al. Polarity of the first episode, clinical characteristics, and course of manic depressive illness: a systematic retrospective investigation of 320 bipolar I patients. Compr Psychiatry 2000;41:13–8. [117] Akiskal HS, Hantouche E, Bourgeois M, Azorin JM, Sechter D, Allilaire JF, et al. Gender, temperament and the clinical picture in dysphoric mixed mania: findings from a French national study (EPIMAN). J Affect Disord 1998;50:175–86. [118] Perugi G, Maremmani I, Toni C, et al. The contrasting influence of depressive and hyperthymic temperament on psychometrically derived manic subtypes. Psychiatr Res 2001;101:249–58. [119] Winokur G, Coryell W, Endicott J, et al. Further distinctions between manic-depressive illness (bipolar) and primary depressive disorder (unipolar). Am J Psychiatry 1993; 150:1176–81.

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

735

[120] Regier DS, Farmer ME, Rae DS, et al. Comorbidity of mental disorders with alcohol and other drug abuse. JAMA 1990;264:2511–8. [121] Sonne SC, Brady KT. Substance abuse and bipolar comorbidity. Psychiatr Clin N Am 1999;22:609–27. [122] Winokur G, Turvey C, Akiskal H, et al. Alcoholism and drug abuse in three groups: bipolar I, unipolars and their acquaintances. J Affect Disord 1998;50:81–9. [123] McElroy SL, Pope HG Jr, Keck PE Jr, et al. Are impulse-control disorders related to bipolar disorder? Compr Psychiatry 1996;37:229–40. [124] Fava M, Anderson K, Rosenbaum JF. Anger attacks: possible variants of panic and major depressive disorders. Am J Psychiatry 1990;147:867–70. [125] Linden RD, Pope HG Jr, Jonas JM. Pathological gambling and major affective disorder: preliminary findings. J Clin Psychiatry 1986;47:201–3. [126] McCormick RA, Russo AM, Ramirez LF, et al. Affective disorders among pathological gamblers seeking treatment. Am J Psychiatry 1984;141:215–8. [127] McElroy SL, Hudson JI, Pope HG, et al. The DSM-III-R impulse control disorders not elsewhere classified: clinical characteristics and relationship to other psychiatric disorders. Am J Psychiatry 1992;149:318–27. [128] McElroy SL, Keck PE Jr, Phillips KA. Kleptomania, compulsive buying and binge eating disorder. J Clin Psychiatry 1995;56(Suppl 4):14–26. [129] McElroy SL, Pope HG Jr, Hudson JI, White KL. Kleptomania: a report of 20 cases. Am J Psychiatry 1991;148:652–7. [130] Ramirez LF, McCormick RA, Russo AM, et al. Patterns of substance abuse in pathological gamblers undergoing treatment. Addict Behav 1983;8:425–8. [131] Cantwell DP, Sturzenberger S, Burroughs J, et al. Anorexia nervosa: an affective disorder? Arch Gen Psychiatry 1977;34:1087–93. [132] Swift WJ, Andrews D, Barklage NE. The relationship between affective disorder and eating disorders: a review of the literature. Am J Psychiatry 1986;143:290–9. [133] Hatsukami DK, Mitchell JE, Eckert ED. Eating disorders: a variant of mood disorders? Psychiatr Clin N Am 1984;7:349–65. [134] Ghadirian AM, Steiger H, Leichner PP. A manic episode in the course of anorexia nervosa with bulimia. Psychosomatics 1989;30:101–3. [135] Herridge PL, Pope HG Jr. Treatment of bulimia and rapid-cycling bipolar disorder with sodium valproate: a case report. J Clin Psychopharmacol 1985;5:229–30. [136] Hsu LKG, Clement L, Santhouse R. Treatment of bulimia with lithium: a preliminary study. Psychopharmacol Bull 1987;23:45–8. [137] Kaplan AS, Garfinkel PE, Darby PL, et al. Carbamazepine in the treatment of bulimia. Am J Psychiatry 1983;140:1225–6. [138] Greenberg BR, Harvey PD. Affective lability versus depression as determinants of binge eating. Addict Behav 1987;12:357–61. [139] Kassett JA, Gerson ES, Maxwell ME, et al. Psychiatric disorders in the first-degree relatives of probands with bulimia nervosa. Am J Psychiatry 1989;146:1468–71. [140] Winokur G. Is there a common genetic factor in bipolar and unipolar affective disorder? Compr Psychiatry 1980;21:460–8. [141] Simpson SG, Al-Mufti R, Andersen A, et al. Bipolar II affective disorder in eating disorder inpatients. J Ment Nerv Dis 1992;180:719–22. [142] Hudson JL, Pope HG, Yurgelun-Todd D. A controlled study of lifetime prevalence of affective and other psychiatric disorders in bulimic outpatients. Am J Psychiatry 1987;144:1283–7. [143] McElroy SL, Keck PE, Pope HG. Sodium valproate: its use in primary psychiatric disorders. J Clin Psychopharmacol 1987;7:16–24. [144] Rice JP, McDonald-Scott P, Endicott J, et al. The stability of diagnosis with an application to bipolar II disorder. Psychiatry Res 1986;19:285–96.

736

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

[145] Robins LN, Helzer JE, Weissman MM, et al. Lifetime prevalence of specific psychiatric disorders in three sites. Arch Gen Psychiatry 1984;41:949–58. [146] Rosenthal NE, Heffernan MM. Bulimia, carbohydrate craving, and depression. A central connection? In: Wurtman RJ, editor. Nutrition and the brain. New York: Raven Press; 1986. p. 139–66. [147] DePaulo JR, Simpson SG, Gayle JO, et al. Bipolar II disorder in six sisters. J Affect Disord 1990;19:259–64. [148] Endicott J, Nee J, Andreasen N, et al. Bipolar II: combine or keep separate? J Affect Disord 1985;8:17–28. [149] Perugi G, Toni C, Akiskal HS. Anxious-bipolar comorbidity: diagnostic and treatment challenges. Psychiatr Clin N Am 1999;22:565–83. [150] Kaye WH, Gwirtsman HE, George DT, et al. Relationship of mood alterations to bingeing behavior in bulimia. Br J Psychiatry 1986;149:479–85. [151] Gershon ES, Schreiber JL, Hamovit JR, et al. Clinical findings in patients with anorexia nervosa and affective illness in their relatives. Am J Psychiatry 1984;141:1419–22.

Appendix 1: The cyclothymia scale of TEMPS-A : the auto-questionnaire version of the temperament evaluation of Memphis, Pisa, Paris, and San Diego, for the cyclothymic temperament* Circle T (True) for all items that are true about you for much of your life. Circle F (False) for all the rest that do not apply to you for much of your life. 1. T F I often feel tired for no reason. 2. T F I get sudden shifts in mood and energy. 3. T F My moods and energy are either high or low, rarely in between. 4. T F My ability to think varies greatly from sharp to dull for no apparent reason. 5. T F I can really like someone a lot, and then completely lose interest in him or her. 6. T F I often blow up at people and then feel guilty about it. 7. T F I often start things and then lose interest before finishing them. 8. T F My mood often changes for no reason. 9. T F I constantly switch between being lively and sluggish. 10. T F I sometimes go to bed feeling down but wake up in the morning feeling terrific. 11. T F I sometimes go to bed feeling great but wake up in the morning feeling life is not worth living. 12. T F I am told that I often get pessimistic about things and forget previous happy times. 13. T F I go back and forth between feeling overconfident and feeling unsure of myself. 14. T F I go back and forth between being outgoing and being withdrawn from others. 15. T F I feel all emotions intensely. 16. T F My need for sleep varies a lot from just a few hours to more than 9 hours.

G. Perugi, H.S. Akiskal / Psychiatr Clin N Am 25 (2002) 713–737

737

17. T F The way I see things is sometimes vivid but at other times lifeless. 18. T F I am the kind of person who can be sad and happy at the same time. 19. T F I daydream a great deal about things that other people consider impossible to achieve. 20. T F I often have a strong urge to do outrageous things. 21. T F I am the kind of person who falls in and out of love easily. *

A score of >10 is indicative of cyclothymia [67].