Dysthymia, Cyclothymia, and Related Chronic Subthreshold Mood Disorders

Dysthymia, cyclothymia, and related chronic subthreshold mood disorders.

Topics covered: 

  • Subthreshold affective conditions
  • The dysthymic spectrum
    • History
    • Clinical picture and diagnostic considerations
    • Course
    • Epidemiology
    • Aetiological considerations
    • Treatment
    • Evidence
    • Clinical management
    • Prevention opportunities
  • Cyclothymic disorder and labile–irritable variants
    • History
    • Clinical features and diagnostic considerations
    • Irritable periods
    • Romantic–conjugal failure
    • Uneven school and work record
    • Alcohol and drug abuse
    • Financial extravagance
    • Course patterns
    • Epidemiology
    • Aetiological aspects
    • Clinical management
    • Prevention
  • Complications of the hyperthymic temperament
    • History
    • Epidemiology
    • Diagnostic aspects
    • Aetiological aspects
    • Course and treatment 
  • References

Subthreshold affective conditions

Long before psychiatry moved to the outpatient arena in the latter part of the twentieth century, psychiatrists had observed milder mood disturbances among the kin of patients hospitalized for endogenous or psychotic depressions or mania. Some were described as sullen, morose, or otherwise moody, but without discrete episodes; others reported self-limited episodes, but often went untreated.

With the advent of modern treatments, practitioners are being increasingly consulted by patients presenting with attenuated affective disturbances. Although the relationship of these ambulatory mood disorders and more classical severe affective states has not been resolved, there is emerging sleep electroencephalography (EEG) and familial-genetic evidence (1,2,3 and 4) that a continuum exists between them. Along the same lines, studies conducted in the United States and Germany (5,6) into what were once described as ‘neurotic' depressions have revealed a progression to more endogenous, psychotic, or otherwise severe, mood states—even bipolar switching in 18 per cent of patients. For these and related reasons, current official classification systems such as the ICD-10 and DSM-IV, have dropped the neurotic–endogenous dichotomy.

Sceptics would perhaps argue that the new categorization of depressive disorders into dysthymic and major subtypes is not much of an improvement. None the less, the new terminology has drawn attention to a large universe of human suffering that had been neglected in the past, and the conceptualization of dysthymia as a variant of mood disorder has had a far-reaching impact on diagnostic and therapeutic habits of clinicians worldwide. (7) The emerging concept of the bipolar spectrum, (8) which does include manic, cyclic depressive (bipolar II), cyclothymic, and related conditions, is beginning to have a similar impact on psychiatric practice. (9)

The subthreshold mood disorders are not only in continuum with more pathological mood states, but they also provide a bridge with normal affective conditions. In this context, temperament, as a construct encompassing affective personalities, is currently enjoying a renaissance as one of the possible substrates for the origin of mood disorders. (10) Temperament classically refers to an adaptive mixture of traits which, in the extreme, can lead to illness or modify the expression of superimposed affective states. The subthreshold conditions covered in this article represent the extreme expressions of these temperaments.

In the current literature, various items such as ‘minor affective states', ‘intermittent depression', ‘hysteroid dysphoria', and ‘atypical depression' are often used for subthreshold disorders. (11) These terms are avoided here, because in contemporary practice these conditions are at least as ‘typical' as major mood disorders: their impact on the sufferer is not time-limited, nor minor, and involves more than a state of demoralization and moral foible. The following passage from Sir Aubrey Lewis (12) is à propos:

...Severe emotional upsets ordinarily tend to subside, but mild emotional states...tend to persist, as it were, autonomously. Hence the paradox that a gross blatant psychosis may do less damage in the long run than some meagre neurotic incubus: a dramatic attack of mania or melancholia, with delusions, wasting, hallucinations, wild excitement may have far less effect on the course of man's life than some deceptively mild affective illness which goes on so long that it becomes inveterate. The former comes as a catastrophe and when it has passed the patient takes up his life again...while with the latter he may never get rid of his burden.

It is a curious fact that most subthreshold affective conditions, while symptomatologically attenuated, tend to pursue a chronic course. This raises the question, partially addressed in this article, that these conditions in their trait expressions might serve some useful function, even as they burden the individual with cares and instability which could predispose to full-blown affective disease. By their very chronicity, these subaffective conditions pose difficult conceptual and clinical questions about their differentiation from personality disorders. (13) Sceptics might argue that subthreshold affective conditions are nothing more than expressions of ‘neuroticism'.

Actually, a close examination of the Eysenck personality inventory, which ranges over a large terrain of depressiveness, anxiousness, emotionality, and mood lability among others, (14) reveals low-grade intermittent affective symptomatology. (15) And at least one genetic investigation has reported that neuroticism and major depression in women share substantial genetic underpinnings. (16) None the less, clinicians have always preferred categorical constructs, because neuroticism and related personality constructs do not do justice to the rich clinical phenomenology of disorders within the subaffective realm. (17)

The dysthymic spectrum


The term ‘dysthymia' (meaning ‘bad mood') originated in classical Greek and is still in current use in that country with the same connotation. (18) In the Hippocratic school, it was considered as part of the broader concept of melancholia (meaning ‘black bile'). A temperament predisposed to melancholia was also delineated, and referred to individuals who were lethargic, brooding, and insecure. It was not until the early nineteenth century that dysthymia was reintroduced into medicine by the German physicians Stark and Fleming to describe depressions in inpatients that pursued a chronic course. Eventually, dysthymia came to subsume all mood disorders. The major residue of dysthymia in the latter sense in Europe today is the French rubric of les dysthymies, as a synonym for troubles de l'humeur; the DSM-IV or ICD-10 ‘dysthymic disorder' in that country is translated as le trouble dysthymique.

The more direct lineage of our current usage of the term dysthymia is to be found in the latter part of the nineteenth century in the work of Kraepelin, who delineated the depressive disposition as one of the constitutional foundations of affective episodes. (19) The condition often began early in life, such that by adolescence many showed an increased sensitivity to life's sorrows and disappointments: they were tormented by guilt, had little confidence in their abilities, and suffered from low energy. As they grew into adulthood, they experienced ‘life with its activity [as] a burden which they habitually [bore] with dutiful self-denial without being compensated by the pleasures of existence'. In some, these temperamental peculiarities were so marked that they could be considered ‘morbid without the appearance of more severe, delimited attacks...' (clearly foreshadowing the modern concept of trait dysthymia). In other cases, recurrent melancholia arose from this substrate without definite boundaries (again anticipating the concept of ‘double-depression').

Subsequently, Kurt Schneider in his opus Psychopathic Personalities (20) devoted considerable space to a depressive type whose entire existence was entrenched in suffering. Building on this rich phenomenological tradition, our research in Memphis (21) helped in operationalizing the core characteristics of such patients encountered in contemporary practice: gloomy, sombre, and incapable of having fun; brooding, self-critical, and guilt-prone; lack of confidence, low self-esteem, preoccupation with failure; pessimistic, easily discouraged; easy to tire, sluggish, and bound to routine; non-assertive, self-denying, and devoted; shy and sensitive. These traits have excellent internal consistency and discriminatory ability. (22,23) Similar concepts have also appeared in the Japanese literature, (24) with particular emphasis on self-critical attitudes, persistence in work habits, and devotion to others. Finally, the French construct of la depression constitutionelle (25) has emphasized the lethargic aspects with a sense of inadequacy.

The classical tenet in psychiatry has been that affectively ill patients recover from their acute episodes with relatively little symptomatic residua and dysfunction. Community psychiatry, which has given renewed visibility to the temperamentally expressed low-grade fluctuating depressive disorders, has challenged this classic view. With the advent of DSM-III, such patients are now officially designated as ‘dysthymic.' In the ICD-10 classification, the low-grade depressive baseline is considered the main pathology; only an occasional superimposed depressive episode is permitted, provided that it is mild. In the latest American Psychiatric Association revision (DSM-IV), at least two patterns have been described: pure dysthymia uncomplicated by major depression, and a more prevalent pattern of dysthymia complicated by major depressive episodes that could be even moderate or severe in intensity (and which has been dubbed ‘double depression').

The mystery of this incapacitating depressive subtype (26)—long recognized, but only recently sanctioned in official diagnostic manuals—is that, in their habitual condition, sufferers lack the classical ‘objective' or ‘major' signs of acute clinical depression, such as profound changes in psychomotor and vegetative functions. Instead, patients consult their doctors for more fluctuating complaints consisting of gloominess, lethargy, self-doubt, and lack of joie de vivre; they typically work hard, but do not enjoy their work; if married, they are deadlocked in bitter and unhappy marriages which lead neither to reconciliation nor separation; for them, their entire existence is a burden: they are satisfied with nothing, complain of everything, and brood about the uselessness of existence. As a result, in the past those who could afford it were condemned to the couch for what often proved to be interminable analysis. The legitimization of dysthymia as a clinically significant variant of affective disorder in both the United States and WHO classifications has helped the cause of more cost-effective treatments.

To sum up, for nearly 2500 years physicians have described individuals with a low-grade chronic depressive profile marked by gloominess, pessimism, low enjoyment of life, relatively low drive, yet endowed with self-critical attitudes and suffering for others. This constellation is as much a virtue as it is a disposition to melancholy, and many dysthymic patients presenting clinically have various admixtures of major depression. This is compatible with a spectrum-concept of depressive illness, which defines various degrees of severity.

Clinical picture and diagnostic considerations

Diagnostic criteria for dysthymia in both DSM-IV and ICD-10 stipulate a 2-year duration of low-grade depressive symptoms, exclusive of such indicators of severity as suicidality and psychomotor disturbances. Dysthymia is distinguished from chronic major depressive disorder by the fact that it is not a sequel to well-defined major depressive episodes. Instead, patients often complain that they have always been depressed. (27) Most are of early onset (less than 20 years). A late-onset subtype (28) first manifesting after the age of 50 is much less prevalent and has not been well characterized clinically, but it has been identified largely through studies in the community.

At their best, dysthymic individuals invest whatever energy they have in work, leaving none for leisure or social activities. According to Tellenbach, (29) such dedication to work represents an overcompensation against depressive disorganization. Kretschmer (30) had earlier suggested that such persons were the backbone of society, devoting their lives to jobs that require dependability and great attention to detail. These features represent the obsessoid facet of dysthymia. Such individuals may seek outpatient counselling and psychotherapy for what some clinicians might consider ‘existential depression': individuals who complain that their life lacks lustre, joy, and meaning. Others present clinically because of an intensification of their gloom to the level of clinical depression; history of lifelong low-grade depressive symptoms would distinguish them from episodic major depressive patients.

The proverbial dysthymic patient will often complain of having been ‘depressed since birth'. (21) In the eloquent words of Kurt Schneider, (20) ‘they view themselves as belonging to an “aristocracy of suffering”'. These hyperbolic descriptions of suffering in the absence of more objective signs of depression earn such patients the label of ‘characterological depression'. (27) The description is further reinforced by the fluctuating depressive picture that merges imperceptibly with the patient's habitual self, leading to the customary clinical uncertainty as to whether dysthymic disorder belongs to the affective or personality disorder domains.

At their worst, patients with low-grade depression having an intermittent course can present such instability in their life, including suicidal crises, that some clinicians would entertain the diagnosis of borderline personality disorder. This is not consistent with the classic picture of dysthymia arising from a temperamental type with more mature ego structure (31) described above. Depressives with unstable (that is to say, ‘borderline') personality structure more often belong to the irritable cyclothymic–bipolar II spectrum.

The greatest overlap of dysthymia is with major depressive disorder, but differs from it in that symptoms tend to outnumber signs (more subjective than objective depression). Thus, marked disturbances in appetite and libido are uncharacteristic, and psychomotor agitation or retardation is not observed. Nonetheless, subtle ‘endogenous' features are not uncommonly reported: inertia, lethargy, and anhedonia that are characteristically worse in the morning. (27) Because many patients with dysthymia presenting clinically fluctuate in and out of a major depression, the core DSM-IV criteria for dysthymia tend to emphasize vegetative dysfunction, whereas the alternative criterion B for dysthymia in a DSM-IV appendix lists cognitive symptoms; although the latter appear more characteristic of trait dysthymia, the DSM-IV field trial (32) could not demonstrate their specificity for dysthymia.

A recent Italian investigation (33) of a large sample from community and primary-care medical settings revealed that negative mood (by definition), along with low energy, poor concentration, low self-esteem, sleep and appetite disturbance, and hopelessness (in descending order) were the most common symptoms of dysthymia. These data suggest that the cognitive and somatic symptoms are not easily separable in practice. None the less, this study did raise the possibility that factors could be discerned along two different axes: ‘negative affectivity' and ‘lassitude with poor concentration'. In our experience, patients loading on the latter factor often complain of hypersomnia and may exhibit subtle bipolar signs; alternatively, they might have some link to the poorly defined constructs of neurasthenia, chronic fatigue syndrome, and fibromyalgia. In terms of differential diagnosis, patients with chronic fatigue syndrome present with disabling fatigue and, typically, deny depressive symptoms; patients with fibromyalgia complain of pain; by contrast, the typical patient with dysthymia cannot stop relating to the physician his or her litany of depressive symptoms. Polysomnography, though not yet definitive, may shed some light on differentiating fibromyalgia from dysthymia proper. (34)

Although dysthymic disorder represents a more restricted concept than does its parent, neurotic depression, it is still quite heterogeneous. Anxiety is not a necessary part of its clinical picture, yet dysthymia is sometimes diagnosed in patients with anxiety and neurotic disorders. That clinical situation is perhaps to be regarded as a secondary or ‘anxious dysthymia' or, as some British authors seem to prefer, as part of a ‘general neurotic syndrome' (35) (an implicit partial return to the now defunct concept of neurotic depression).

The clinical picture of dysthymic disorder that emerges from the foregoing descriptions is quite varied, with many who fluctuate in and out of major depression, (36) whereas in others the pathology is woven into the habitual self. (27) These considerations suggest that a clinically satisfactory operationalization of dysthymia must include both symptoms and trait characteristics (Table 1). The following vignette illustrates this more prototypical form of dysthymic suffering.

Table 1 The core characteristics of dysthymia

  • Long-standing subthreshold depression of a fluctuating or persistent nature
  • Gloomy and joyless disposition
  • Brooding about the past and guilt prone
  • Low drive and lethargy
  • Low self-esteem and preoccupation with failure
  • Identifies suffering as part of the habitual self
Case Study

This 37-year-old never-married male teacher presented with the complaint that he was ‘tired of living' and was considering ‘ending it all'. He said that much of his life had been ‘wasted', he had never known any joy, that all human existence for him was a ‘tragic mistake of God'. He was known to be a dedicated and talented teacher, but he felt all his efforts had been ‘useless and in vain'. He said he probably was ‘born depressed', because he had not known any happiness and that the only utility he could have for mankind was ‘to serve as a specimen to be researched—to shed light on human misery'. Although he conceded that some women found him interesting, even intellectually stimulating, he said he could not enjoy physical intimacy, that even orgasm lacked passion; nonetheless, he masturbated frequently, fantasizing about married female teachers—only to feel guilty. We could not document any major affective episodes. He stated that he had always functioned at a ‘mediocre level' (which was at variance with the good feedback students had given him year after year); but did admit he ‘appreciated work, because there was nothing else to do'. He denied alcohol and drug habits. There had never been any periods of hypomania, but one of his maternal aunts had been treated for a ‘cyclical depression' and was apparently doing well on lithium. The patient's mother was a sombre serious work-oriented woman who had raised three children and had done voluntary work for the church, but had no depressive complaints. His father had died from a coronary attack, but his side of the family was otherwise unremarkable.

Although both DSM-IV and ICD-10 omit suicidal preoccupations in their diagnostic criteria for dysthymia, as testified by the above case, this is what often brings patients to clinical attention.


An insidious onset of depression dating back to late childhood or the teens, preceding any superimposed major depressive episodes by years, even decades, is the most typical developmental background of dysthymic disorder. A return to the low-grade depressive pattern is the rule following recovery from superimposed major depressive episodes, if any—hence the designation ‘double depression'. (36)

A long-term prospective study of prepubertal children (37) has revealed an episodic course of dysthymia with remissions and exacerbations, and eventual complication by major depressive episodes, as well as hypomanic, manic, or mixed episodes postpubertally. Persons with dysthymic disorder presenting clinically as adults tend to pursue a chronic ‘unipolar' course, which may or may not be complicated by major depression: they rarely develop spontaneous hypomania or mania. However, when treated with antidepressants, some adult patients with dysthymia may experience brief hypomanic switches (38) that typically disappear when the antidepressant dose is decreased. Although ICD-10 and DSM-IV would not ‘allow' the occurrence of such switches in dysthymia, systematic clinical observation (39,40) have verified their occurrence in between 10 and 30 per cent of dysthymic patients. In that special dysthymic subgroup, the family histories are typically positive for bipolar disorder. (38) Such patients, often conforming to the double depressive pattern, represent a clinical bridge between major depressive disorder and bipolar II. (8)

A recent 12-year prospective study (41) has shown that patients with major depressive disorder spent 44 per cent of their course in low-grade depression (versus 15 per cent of time in major depressive episodes). This suggests that major depression, dysthymia, or otherwise subsyndromal depression constitute somewhat artificial conventions on the threshold and duration of depressive illness, representing alternative manifestations of the same diathesis. In this context, residual intermorbid depressive symptoms have been confirmed as being strongly predictive of a rapid relapse into a new major depressive episode. Various ‘major' and ‘minor' depressive conditions described in DSM-IV and its appendix must not be viewed as distinct depressive subtypes, but part of a symptomatic continuum. (42) 


From 1966 to 1980, Index Medicus listed no more than 10 articles on chronic depressions. Since 1980, when dysthymia was first introduced in DSM-III, at least 200 articles have appeared on chronic depression, mostly on dysthymia. This phenomenal growth in research interest parallels the increasing public health significance of this disorder. It is estimated that 3 per cent of the North American population is suffering from this condition; (43) the figures in other parts of the world are comparable, at least on average. (44) Like major depressive disorder, dysthymia is twice as common in women as in men. Because of its chronicity, dysthymia is among the most common clinical presentations of a psychiatric condition. Dysthymia is more disabling, as far as quality of life in social and personal areas, work, and leisure, than depression in the setting of a severe anxiety disorder like agoraphobia. (45) Celibacy is also common in early-onset dysthymia, (46) but not for long; modern successful treatments often lead to a change in marital status!

Other research in primary care (47) has focused on depressive symptoms falling short of the major depressive threshold, as far as symptom intensity is concerned, as well as falling short of the 2-year duration criterion for dysthymia. Despite its chronicity, 50 per cent of people remain unrecognized by general practitioners. Despite the low-grade nature of their depressive complaints, these patients report high degrees of morbidity and impairment in a variety of health domains and quality of life, including ‘bed days' (namely the number of days per year they stayed ill in bed). Actually, these impairments are generally more pronounced than those of patients with a variety of medical conditions, such as hypertension, diabetes, arthritis, and chronic lung disease; only coronary artery disease exceeded the disability of low-grade depression in several domains.

In light of the foregoing developments, both the World Psychiatric Association (48) and the World Health Organization (49) have developed programmes to address the challenges of educating general practitioners in the proper recognition and treatment of dysthymia.

Aetiological considerations

Some sensitivity to suffering, a cardinal feature of the depressive temperament, represents an important attribute in a species like ours, where caring for young and sick individuals is necessary for survival. This temperament, historically the Anlage of dysthymia, in the extreme often leads to clinical depression. The constitutional viewpoint, while dominant in the early part of the twentieth century, (19) gradually disappeared from psychiatric thinking. One reason was that Kurt Schneider (20) preferred to conceptualize this form of pathology as ‘psychopathy', by which he meant abnormal personality development. Independently, Freud's disciples took this one step further and, eventually, in outpatients, all milder depressions with a tendency to chronicity came to be considered as the expressions of a character neurosis. (50) In support for this position, these authors could point to the long-standing nature of the interpersonal difficulties in the lives of these individuals. When, and if, antidepressants were prescribed, they were given in homeopathic doses; worse, many patients received stimulants or benzodiazepines rather than genuine antidepressants. Failure to respond to these incorrectly chosen pharmaceutical agents seemed to further reinforce the notion of a ‘character' defect.

Several lines of observation have challenged the concept of ‘character neurosis' as an explanation for low-grade depression, and thereby forced a return to the more classical European concept of temperament with its biological underpinnings. First, in a 1980 study of rapid eye movement ( REM) latency (normally 90 min, measured from sleep onset to the first REM period) conducted in ‘depressive characters' who were not in a state of major depression, we reported that REM latency was less than 60 min, and sleep was redistributed to the early part of the night (27) (which was the reverse of what we observed in chronic anxious patients (51)). Moreover, a family history for major affective illness (including bipolar) was significantly high in short-REM latency patients. (38) (The reverse was true for those with familial alcoholism and sociopathy.)

The sleep findings were so reminiscent of those seen in major affective illness that we were compelled to give our patients systematic open trials with desipramine and nortriptyline (the best-tolerated secondary amine tricyclics in those days) or lithium carbonate if antidepressants failed (based on the observation of familial bipolar disorder in some). (27) Nearly 40 per cent remitted, of whom one out of three developed brief hypomania. The sleep findings have been replicated in other laboratories, (52) and Rihmer (39) has shown that patients with dysthymia experience transient lifting of their mood with sleep deprivation. Klein et al. (53) have shown high rates of affective illness in a systematic familial investigation of dysthymic probands. There also exist dysthymic patients whose lifelong suffering and discontent appear, in retrospect, (38) a legacy of an unsatisfactory childhood marked by deprivation or abuse at the hands of alcoholic and/or sociopathic parents or step-parents. Although it is clinically attractive to invoke the notion of ‘learned helplessness' secondary to such inescapable childhood traumata, an alternative hypothesis (54) is that the helplessness of these individuals might develop secondary to an inherited diathesis which biases these children's early experiences in a dysphoric direction.

As for neuroendocrine markers, (52) thyroid-releasing hormone–thyroid-stimulating hormone challenge and electrodermal activity similar to those with major depressive disorders are the main findings; by contrast, dexamethasone suppression and catecholamine metabolism are essentially unaltered in dysthymia. These observations, along with the REM latency findings, suggest that dysthymia represents trait depression. Coupled with the family history data, this traitness can be postulated to be of constitutional origin. Certainly, the occurrence of major affective episodes in the long-term course of dysthymia, in both community and clinical samples, (50,51) is in line with this position. It is, therefore, of great theoretical and practical significance that shortened REM latency has been reported in the offspring of the affectively ill. (55,56) 

There are also medical and neurological factors that may contribute to dysthymic symptom formation. (52) Actually, joint medical–neurological and non-affective psychiatric disease is often contributory to extreme refractoriness among the chronic depressive states of these patients. Such patients are at risk for suicide, especially those with epilepsy or progressive degenerative neurological disease. Interestingly, living with a medically disabled spouse or family member, too, can be associated with some chronicity of depression. The emergence of pathogenetic understanding, as outlined above, is all the more impressive, given the controversies on the very nature of dysthymia and its legitimacy as a nosological entity. (57)


The trait nature of dysthymia can be further observed in the fact that dysthymia often pursues an unrelenting course towards chronicity. (11) Thus, spontaneous recovery has been shown to occur in no more than 13 per cent of subjects in the community over 1 year. (58) In outpatient clinics, the outcome is somewhat better, but this is probably due to the treatment received and a longer follow-up. (11)


Most classes of antidepressants (59,60,61,62,63,64,65 and 66) have been shown to be effective in dysthymia in double-blind studies. The rationale for using classic antidepressants such as tricyclic compounds was the observation of shortened REM latency in dysthymic subjects in our sleep laboratory. (27) Irreversible monoamine oxidase inhibitors such as phenalzine were used because of the belief that non-classical depressions respond preferentially to this class of drugs; (61) the same can be said for the reversible inhibitors of type A monoamine oxidase. (66) Ritanserin (62) (now withdrawn because of cardiotoxicity) and amisulpride (65) were tried, because both compounds reverse ‘negative' symptoms in schizophrenia and, by analogy, it was hypothesized that the low motivation and lethargy seen in some patients with dysthymia reflected a shared underlying dimensional transnosological pathology. The selective serotonin re-uptake inhibitors (SSRIs) were used empirically, (63,64) because of good tolerance compared with the tricyclic antidepressants, and later it was suggested that improvement in dysthymia correlates with normalization of serotonergic indices. (67) The foregoing trials, conducted in different countries during the past 12 years represent the most eloquent evidence for the increasing worldwide acceptance of the concept of dysthymia as a clinically significant variant of affective disorder.

Clinical management

The treatment of dysthymia should continue in most cases for 2 years or more. Tricyclic antidepressants have too many side-effects in clinically effective doses (desipramine equivalent of 150 mg or more per day). Given dietary and medication prohibitions, monoamine oxidase inhibitors are also not practical as first-line drugs. Overall good tolerance in long-term use, despite sexual side-effects, has made the SSRIs the first-line intervention treatment for dysthymia; given that many people with dysthymia are young individuals who should be eager to form families, their acceptance of long-term SSRI use is an indication that the alleviation of the depressive suffering of dysthymia is genuine and far outweighs the sexual dysfunction. However, 75 to 150 mg bupropion-SR (an antidepressant available in the United States) can be taken in the morning on the desired day of sexual union, but preferably no more than once a week. If severe, sexual dysfunction can be reversed by switching to nafazodone with sex-sparing 5-HT2 action (which might work on clinical grounds, though its antidysthymic action is still under investigation). Moclobemide also spares sexual function, but seems more effective in anxious and milder cases of dysthymia. Amisulpride, which rarely causes amenorrhoea and/or galactorrhoea, is otherwise well tolerated in dysthymia in the more lethargic forms of the illness seen in general medical practice.

The dosage of nearly all antidepressants in dysthymia is in the full range for that recommended for major depression (20–40 mg for fluoxetine). In the case of amisulpride, the dosage is low (25–50 mg), because at this dosage the drug is a dopamine agonist, believed to be the necessary ingredient for its mechanism of action in dysthymia. Both dysthymia and double depression respond equally well, and the duration of underlying dysthymia does not seem to matter. The main difference in treatment for these two course patterns is that dysthymia need not be treated for a lifetime, but double depression should probably be treated indefinitely. Women seem to have a preferentially better response to SSRIs, (46) which have the added benefit of treating the premenstrual accentuation of dysthymic symptoms.

Borderline thyroid function (e.g. a high baseline thyroid-stimulating hormone level) preferentially occurs in women with dysthymia, so that these women would benefit from thyroid augmentation (levothyroxine 175 mg/day) of the antidepressant. In those patients who oversleep in the morning, terminal sleep deprivation and/or morning phototherapy represent useful adjuncts to antidepressants. Although there are no controlled studies in children, our clinical experience indicates that SSRIs often prove effective in this population, with the appropriate dosage reduction for body-weight. In adults, concurrent personality disturbances (for instance, obsessoid, avoidant, dependent, and hostile features) do not compromise therapeutic responses. (46) Indeed, more often than not, such personality disturbances recede with the successful alleviation of dysthymic suffering; social function improves in tandem. (68) (However, extremely hostile patients, who may meet symptomatological criteria for ‘dysthymia' but whose irritable dysphoria more appropriately belongs to the cyclothymic domain, are best managed with mood-stabilizing anticonvulsants, for example divalproex 600 to 1200 mg/day.)

Together with Haykal, we have shown that, with the judicious use of the foregoing modalities in private practice, (46) three out of four patients with dysthymia engulfed in gloom for much of their lives had sustained remissions for five or more years. Depressive episodes and suicidal preoccupation and/or crises were prevented, in tandem with the alleviation of the dysthymia. Approximately 15 per cent experienced ‘overcorrection' of their dysthymia in the direction of mild hypomania; this is particularly likely in the presence of inhibited-social phobic traits as part of dysthymia, and when the family history is bipolar. The hypomania is typically short-lived, and tends to disappear when the antidepressant dose is reduced; in some cases, it is necessary to provide lithium (600–900 mg/day) or valproate augmentation (500–750 mg/day).

The question has been raised whether selective serotonin-reuptake inhibitors, in particular fluoxetine, change the personality in a hyperthymic direction. In our experience, most observed changes are compatible with adaptive behaviour that emerge as a result of alleviation of depressive suffering; the more distinctly protracted hypomanic changes nearly always require familial bipolar diathesis. (46) It is, nonetheless, true that with SSRIs we have entered an era of ‘dimensional psychopharmacology', whereby the clinician could dose the patient to a desired end from a functional standpoint. (69) Many become care-less rather than careless. (70) The present author has also encountered some patients treated with SSRIs who view a life without cares as negative; in such cases, one should opt for very low doses and a more gradual lifting of the dysthymia, and help the patient adjust to a new self-image of normalcy.

As for psychotherapy, there is little credible evidence for its efficacy as a monotherapy in the treatment of dysthymia. (71) Actually, a panel of female experts (72) has argued that exploration of one's mental inadequacies, in the ‘passive' psychoanalytical situation, is particularly negative for women; the more ‘active' cognitive–behavioural approaches, which encourage thinking, and behaviours reinforcing for the individual, are preferable. Many clinicians profitably use the latter strategy along with pharmacotherapy—to boost the self-esteem of the patient. In a more practical vein, there are clinical management strategies (73) that are specifically useful for both the patients and their clinicians (Table 4). It is particularly important for the clinician not to be submerged by the negative thinking of the patient, and it is even more crucial for the therapist to recognize that a relative lack of progress can generate feelings of ‘impotence' and countertransference; periodic consultation with more experienced clinicians in the treatment of chronic depression would be desirable.

Interpersonal psychotherapy has been used in medication failures by Markowitz. (71) This is best viewed as a more practical abbreviation of psychodynamic psychotherapy, with a strong emphasis on support and encouragement for patients with dysthymia who seek help at a time of loss or role transition in their lives. Knowledge of the interpersonal context of depression is obviously important in formulating how the clinician would stage the psychological recovery process from dysthymia. None the less, there are some suggestions that SSRIs often lead to improved coping behaviour, even without formal psychotherapy. Indeed, it has been shown by Ravindran and colleagues (74) that a successful response to SSRIs is often associated with decreased emotion-focused coping and decreased perception of daily hassles, as well as alleviation of the sense of loneliness one experiences in chronic depression.

No matter what the active ingredients in antidysthymic treatments, there is little doubt that for the first time we have potent practical treatments to alleviate a major source of chronic human suffering, including what were once deemed depressive characterological attributes inseparable from the habitual self. Helping patients attain a new homeostasis of the self is an art unparalleled in the history of medical science. It does not constitute ‘cosmetic psychopharmacology'. (75)

Prevention opportunities

Community subjects with pure dysthymia have been found in two prospective studies (76,77) to be at risk for major depressive episodes. Because dysthymia often makes its first appearance in juvenile years, identifying the disorder at this early stage represents a special opportunity for prevention in child psychiatry and paediatrics.

In still another group of patients, low-grade chronic depressive developments occur in the setting of disabling systematic medical and neurological disorders, and are best categorized as ‘secondary dysthymias'. (78) For instance, poliomyelitis may not only lead to deformities in musculoskeletal structures in children, but could permanently scar the sufferer's sense of enjoyment, fulfilment, and outlook of life. Likewise, low-grade chronic depressive development often complicates neurodegenerative cerebrovascular disease later in life. In both situations, psychological factors might be operative, yet the contribution of specific cerebral lesions to the subthreshold mood disturbance may be substantial. This group as a whole is not well captured by the conventional depressive categories in ICD-10 and DSM-IV. In these subacute dysthymic-like conditions, the affective state is often disabling, yet symptomatologically less severe than major depression; it is low grade, yet not as chronic as dysthymia. ‘Minor depression' would not capture the clinical significance of their condition. Indeed, there is emerging data that treating these subacute dysthymias may improve the prognosis of the underlying neurological disorder. (49)

Cyclothymic disorder and labile–irritable variants


Kraepelin (19) included the cyclothymic disposition as one of the temperamental foundations from which manic–depressive illness arose. Kretschmer (30) went one step further and proposed that this constitution represented the core characteristic of the illness: some patients were more likely to oscillate in a sad direction, while others would more readily resonate with cheerful situations; these were merely viewed as variations in the cyclothymic oscillation between these two extremes. Kurt Schneider,(20) who did not endorse the concept of ‘temperament', instead referred to ‘labile psychopaths' whose moods constantly changed in a dysphoric direction, and who bore no relationship to patients with manic depression. To confuse matters further, Schneider used the term ‘cyclothymia' as a synonym for all manic depressive illness, from the mildest to the most severe psychotic forms. Today, ‘cyclothymia' is still sometimes used in this broader sense in Germanophone psychiatry. But in much of the rest of the world, cyclothymia (short for ‘cyclothymic disorder') is reserved for a form of extreme temperament related to bipolar disorder.

Cyclothymia, which in ICD-9 and DSM-II was subsumed under the affective personalities, was first introduced into DSM-III and DSM-IV and subsequently into ICD-10 as a form of attenuated chronic mood disorder. The diagnosis is not commonly made in clinical practice, because it is almost always seen when a patient presents with major depressive episodes, warranting the designation of ‘bipolar II'. (8,18) None the less, the construct is of great theoretical and practical significance as one of the possible substrates for major mood disorders. Moreover, it could shed light on social and occupational maladjustment and/or addictive behaviours that could otherwise be misattributed to personality disorder.

Clinical features and diagnostic considerations

By definition, individuals with cyclothymia report short cycles of depression and hypomania that fail to meet the sustained duration criterion for major affective syndromes. At various times, they exhibit the entire range of manifestations required for the diagnosis of depression and hypomania, but only from a few days at a time up to 1 week, rarely longer. (79) These cycles follow each other in an irregular fashion, often changing abruptly from one mood to another, with only rare interposition of ‘even' periods. The unpredictability of mood swings is a major source of distress for cyclothymes, as they do not know from moment to moment, how they will feel. (80) As one patient put it, ‘my moods swing like a pendulum, from one extreme to another'. The rapid mood shifts, which undermine the patients' sense of self, may lead to the misleading diagnostic label of borderline personality. But unlike a personality disorder, the mood changes in cyclothymes have a circadian component. One patient described it as follows: ‘I would go to bed in a cheerful mood and wake up down in the dumps'. This observation is in line with psychophysiological data on mood-switching occurring out of the rapid eye movement sleep phase, as reported in more typical cases of manic depression. (81)

The mood swings of cyclothymes are biphasic: eutonia versus anergic periods; people-seeking versus self-absorption; sharpened thinking versus mental paralysis. 

Irritable periods

At one time or another, labile angry or irritable moods are observed in virtually all these patients. (80) Cyclothymes, unlike patients with epilepsy, are aware of their ‘fits of anger', which lead to considerable personal and social embarrassment after they subside. The patients often feel ‘on edge, restless, and aimlessly driven'; family and friends report that during these periods patients seem inconsiderate and hostile toward people around them. The contribution of alcohol and sedative-hypnotic drugs to these moods cannot be denied, but the moods often occur in the absence of drugs. Electroencephalography typically reveals no seizure or subseizure activity. The interpersonal costs of such unpredictable interpersonal explosiveness can be quite damaging. One of our patients reported frequent periods where he would start unprovoked fights with very close friends, only to shift into periods of prolonged ‘soul-searching, guilt, shame, and embarrassment'. In other patients, outbursts of anger are ‘reactive' to minor interpersonal disputes (82)—but once in full force, they are like emotional avalanches with the distinct potential to destroy relationships. Should they dominate the clinical picture, especially among young women who hurt themselves in response to interpersonal contexts, the problematic diagnosis of borderline personality disorder is often invoked (more so in North America than elsewhere). Although controversial, contemporary research suggests that many ‘borderline' patients represent a severe labile–irritable variant of cyclothymia on the border of manic–depressive psychosis. (17,83,84) On the other hand, bizarre episodes of self-harm, recurrent illusions, and dissociative symptoms of the post-traumatic stress disorder type are uncharacteristic of cyclothymia, and suggest other diagnoses.

Romantic–conjugal failure

It is easy to understand how individuals with mercurial moods would charm others when in an expansive people-seeking mode, and rapidly alienate them when dysphoric. In effect, the life of many of these patients is a tempestuous chain of intense but brief romantic liaisons, (80) often with a series of unsuitable partners. Some rationalize their behaviour on the grounds that their spouse or partner is ‘too conservative in sex, too unimaginative, too unaware of the intensity' needed to stimulate them. As expected, frequent marital separations, divorces, and remarriage to the same person occur.

Uneven school and work record

Repeated and unpredictable shifts in work and study habits occur in most people with cyclothymia, giving rise to a dilettante biography. (79,80) Although some do better during their ‘high' periods—for example, one car salesman would sell cars only ‘when up'—for others, the occasional ‘even' periods were more conducive to meaningful work. It is sometimes unappreciated by clinicians inexperienced with bipolarity that the hypomanic period can be one of disorganized and unpatterned busyness that could easily lead to a serious drop in net productivity. For instance, one insurance salesman related that he was less successful when ‘high', because he tended to enter into unproductive arguments with his clients. When ‘down', productivity obviously abates, although two creative individuals in our case series (80)—one inclined poetically, the other towards painting—produced their better work when coming out of mini-depressions.

Alcohol and drug abuse

An alternating pattern of the use of ‘uppers' and ‘downers' occurs in at least 50 per cent of patients. (79) We have clinically evaluated at least five cyclothymes who ‘sold dope' to maintain their habit: two went to prison. These and other observations (85,86 and 87) suggest that a proportion of substance-abusing, especially stimulant-abusing, patients might be suffering from subtle or cryptic forms of bipolar disorder. The bipolar nature of mood swings in alcohol- or substance-abusing individuals can be documented by demonstrating mood swings well past the period of detoxification; in some cases, escalating mood instability makes its first appearance following abrupt drug or alcohol withdrawal. The DSM-IV criteria for drug-induced or drug withdrawal-induced mood disorder are, in our opinion, biased against the diagnosis of otherwise treatable bipolar spectrum disorders. (8)

Financial extravagance

One patient in a psychiatrist's case series reported going to bars and buying people drinks because he wanted everybody to feel like him. Another patient intermittently showered his lovers with expensive jewellery. In general, however, the extravagance of the cyclothymic group reflects gregariousness and tends to occur on a smaller scale compared to the psychotic manner in which manic patients bring financial ruin to themselves and their families.

The social warmth observed among most people with cyclothymia distinguishes them from adult attention-deficit hyperactivity disorder ( ADHD). Also, elation and inflated self-confidence, which occur periodically in cyclothymia, are uncharacteristic of ADHD; the moodiness in the latter is largely depressive in nature. Finally, antidepressants and stimulants typically worsen the moods in cyclothymia; they treat ADHD. In rare cases, however, cyclothymia and ADHD coexist. (80)

Course patterns

In cyclothymia, hypomania and mini-depression alternate with each other from adolescence. For instance, the optimistic, overconfident, people-seeking phase can give way to self-absorption, self-doubt, pessimism, and a sense of futility, emptiness, and suicidal ideation. More commonly, depressive periods dominate the clinical picture, interspersed by ‘even', ‘irritable', and occasional hypomanic periods. Indeed, most people with cyclothymia who present clinically do so because of depression. (80) These depressions are typically short-lived, yet unrelenting in their cyclic course, creating much interpersonal havoc for the patient. The following vignette illustrates the cardinal clinical features of cyclothymia that has not yet progressed to major depression.

Case Study

This 24-year-old songwriter presented with the chief complaint of ‘depression so bad that I become totally dysfunctional—I cannot even get out of bed'. Since her mid-teens she had experienced periods lasting from a few days up to a week, during which she would withdraw into herself, losing confidence and interest, feeling drained of energy, and crying when approached by anybody. These periods were particularly prevalent during the autumn and winter months, but they did not coincide with the premenstrual phase. All she needed sometimes was restful sleep to ‘feel alive again'; at other times she would have little sleep, and would wake up ‘wired', ‘ready to go', ‘open to experience all the joy waiting for me out there'; she would exude confidence, ‘sensuality and sexual aroma'. These occurred less frequently than the ‘down' periods and usually lasted for 1 to 3 days, but were not associated with creative spurts. The latter came as she was descending from ‘highs' into a more ‘mellow depression'. Her success in music had been sporadic, paralleling the sporadic nature of her ‘muses' that visited her on the descending limb of ‘hypomania merging with tears'. However, the major toll of her mood swings had been in her personal life, the intensity of her moods had driven away most men she had loved, of whom she had lost count. She described periods of such intense sexual arousal, that sometimes she would go to bed with anybody, including ‘women of all ages, shapes, and description'. But, she added: ‘I am not a lesbian—oral love is just one way of relating to these women—why not?' She had also experimented with drugs, such as stimulants, which had made her moodiness worse. More recently, she had been prescribed at least two SSRIs, which after a period of ‘success' for a few months, had made her depressive swings more frequent and lower in amplitude, leading to the present consultation in the psychiatrist's clinic.

As documented in this case, sexual excesses with both sexes are often readily admitted by patients. Winter accentuation or clustering of depressive periods, as exemplified here, is not uncommon in cyclothymia. We also would like to point out the special relationship of the moods to artistic productivity which occur in up to 8 per cent of cyclothymic depressions. (88) The 4-day threshold for hypomania in the official diagnostic manuals is too conservative; as shown in this case, most patients with cyclothymia (and bipolar II disorder) report a threshold of 1 to 3 days (though on occasion, one would observe a hypomanic duration of 1 week or longer). (79,89) It is also noteworthy that the episodes are short-lived and do not reach the duration threshold for rapid cycling. Sometimes, the term ‘ultra rapid cycling' is used for these patients, but we prefer to reserve this for extremely severe cases who require hospitalization. The short cycle length in cyclothymia is, in part, a selection artefact: the universe of patients with bipolar disorder is composed of an extreme variety of overlapping patterns. (8)

The relationship of a cyclothymic temperament to the bipolar spectrum is more complex than that of dysthymia to major depressive disorder. Although cyclothymia can be observed in some patients with full-blown manic–depressive illness (bipolar I with severe or hospitalized mania), it is more commonly associated with the bipolar II pattern (of recurrent major depression with self-limited hypomanias). In a recent French national study of patients with major depression, (9) 88 per cent of those with a cyclothymic disposition belonged to the bipolar II subtype. (Mania, by contrast, has been reported to more likely represent either an extension of hyperthymic traits, or a reversal from a depressive temperamental baseline.(90))

One-third of patients with cyclothymia studied by a group of psychiatrist's on a prospective basis, progressed to spontaneous affective episodes with more protracted hypomanias and clinical (major) depression. Thus, 6 per cent of the original cyclothymic cohort could be reclassified as bipolar I, and 30 per cent as bipolar II. The tendency to switch to hypomania was further augmented by the administration of antidepressants. A larger and more recent National Institute of Mental Health study (91) of patients with major depression who switched to bipolar II during a prospective observation period of up to 11 years, found that a temperamental mix of ‘mood-labile', ‘energetic-active', and ‘daydreaming' traits (reminiscent of Kretschmer's concept of the cyclothymic temperament (30)) were the most specific predictors of such outcome. Limited data (92) also exist on the importance of these temperamental factors in predicting which of the offspring of bipolar probands will progress to clinical episodes.

The foregoing clinical and course characteristics suggest that a cyclothymic temperament leading to major depressive recurrences represents a distinct longitudinal pattern of ‘cyclothymic depression', and which appears to capture the core features of bipolar II disorder in contemporary clinical practice. (8,91) Because hypomanic episodes cannot be easily ascertained by history, assessing cyclothymia in clinically depressed patients represents a more sensitive and specific approach to the diagnosis of bipolar II.


An excess of interpersonal difficulties and psychiatric consultations distinguish people with cyclothymia in the community from controls; excessive use patterns of stimulants, caffeine, nicotine, and alcohol, have also been well documented. (93,94) Explosive traits, probably representing the irritable component of cyclothymia, have been reported to be prevalent in the community in a British study. (95) More recently, we found that 6.3 per cent of a national cohort of 1010 Italian students between the ages of 14 and 26 years of age (22) scored above two standard deviations for cyclothymia; this was more prevalent in females, with a ratio of 3:2. Overall, the foregoing data testify to the fact that a cyclothymic and/or labile disposition can be accurately measured, is prevalent, and represents a population at risk for affective disorders. 

Aetiological aspects

The flamboyant behaviour and the restless pursuit of romantic opportunities in cyclothymia suggest the hypothesis that its constituent traits may have evolved as a mechanism in sexual selection. Even their creative bent—in poetry, music, painting, or fashion design—may have evolved to subserve such a mechanism. Cyclothymic traits appear to lie on a polygenic continuum between excessive temperament and manic depression. Indeed, clinically identified cyclothymes have patterns of familial affective illness, as one would expect for a forme fruste disorder.(79)

Cyclothymia has also been observed in the offspring of manic–depressive probands, with onset in the postpubertal period. (92) Finally, family studies of patients with a bipolar disorder have revealed an excess of cyclothymia. (97) Hypothetically, this temperament might represent one of, if not the most important, inherited trait diathesis for bipolar disorder. For instance, moody-temperamental individuals are over-represented in the ‘discordant' monozygotic co-twins of manic–depressives. (98) Alternatively, and in a more theoretical vein, manic–depressive illness might be the genetic reservoir for the desirable cyclothymic traits in the population at large.

Clinical management

The proper pharmacological treatment for cyclothymic excesses is low doses of lithium (600–900 mg/day) or valproate (500–750 mg/day). These are based on open systematic studies. (80,99,100) There is some data from a controlled trial with lithium about the prevention of depression in cyclothymic individuals. (101) Similarly controlled data exist for a related construct—‘labile personality'. (102) Generally speaking, patients with cyclothymia object to the ‘overcontrol' that may come from mood stabilizers, and this is particularly the case with lithium.

Patients should be taught how to live with the extremes of their temperamental inclinations, and to seek professions where they determine the hours that they work. (103) Marriage to a rich older spouse might sustain them for a while, but eventually interpersonal friction and sexual jealousy terminate such marriages. The artistically inclined among them should be encouraged to live in those parts of a city inhabited by artists and other intellectuals, where temperamental excesses are better tolerated. Ultimately, the decision to use mood stabilizers in such individuals should balance any benefits from decreased mood instability against the social and creative spurts that the cyclothymic disposition can bring to them. Their clinical management represents a challenging task for the psychiatrist who is willing to learn about the lifestyle of these individuals, not prejudging them by the more mundane norms of society. But the psychiatrist should also be there to help them during the multiple crises of their lives. Low-dose sedating neuroleptics, both classical (e.g. thioridazine 50 mg at bedtime) and atypical (e.g. olanzapine 2.5 mg at bedtime) may temporarily help to diffuse such crises. It is only when a clinician has earned therapeutic alliance with a patient that the latter will permit limit-setting on his or her extravagant or outrageous behaviour. Parents might also benefit from some counselling, because the dilettante life of their children is often a source of great sorrow for them. Rarely, parents or spouses are rewarded by great artistic or intellectual achievement, which does not necessarily reduce the pain that the volatile cyclothymes bring to their loved ones.

Kurt Schneider (20) admonished the kin of labile individuals (who might approximate the contemporary concept of borderline personality disorder) ‘on their bad days...to keep out of their way as far as possible'. Cyclothymes with some insight into their own temperament would give the same advice to their loved ones. A cautious trial of anticonvulsants will often prove effective in those distressed enough by their behaviour to comply with such treatment.


The offspring of patients with bipolar disorder who exhibit a cyclothymic level of temperamental dysregulation represent a logical population for prevention studies. (92) This is a challenge for the twenty-first century. Presumably molecular genetic testing will one day identify those moody individuals who carry the genes for bipolar disorder. At present, it would be useful to conservatively follow-up the cyclothymic offspring of people with bipolar disorders and provide them with psychoeducation about the necessity of avoiding stimulants and sleep deprivation. It may not be entirely possible to prohibit the use of moderate alcohol consumption, but benzodiazepines should not be used. It is also imperative, should they get depressed, to protect them from the indiscriminate prescription of antidepressants. Mood-labile female prisoners, commonly given the diagnosis of antisocial or borderline personality disorder, (104) may represent affective variants with irritable cyclothymic features. Formal studies are needed in prison populations, to assess more precisely the rates of preventable cryptic bipolarity among female and male offenders.

Complications of the hyperthymic temperament


Although well described by classical German psychiatrists (e.g. Schneider (20)), the hyperthymic type appears neither in DSM-IV nor in ICD-10. A lifelong disposition, hyperthymia must be distinguished from short-lived hypomanic episodes. Alternatively, this disposition can be characterized as trait hypomania. It derives from the ancient Graeco-Roman sanguine temperament, believed to represent the optimal mixture of behavioural traits. They are full of zest, fun-loving, and prone to lechery: their habitual disposition is one of buoyant action-orientation, extraverted people-seeking, overconfidence, and swift thinking. Typically short sleepers, they possess boundless energy to invest in sundry causes and projects, which often earn them leadership positions in the various professions and politics, yet their carefree attitudes and propensity for risk-taking can bring them to the brink of ruin; this is particularly true for their finances and sexual life, which can be marred by scandal. (105) A hyperthymic lifestyle is so reinforcing that some resort to ‘augmenting' it with stimulants such as amphetamines or cocaine. In brief, while a hyperthymic temperament per se does not constitute affective pathology—indeed, it represents a constellation of adaptive traits—in excess it could lead to undesirable complications. The latter will be our main focus here.


The foregoing considerations partly explain why such a temperament has received scant research attention in the community. Extrapolating from studies on intermittent hypomania in the community,(94,106) if strictly limited to hypomania with early onset and persistent course, the prevalence of hyperthymic temperament can be estimated to be slightly under 1 per cent. On the other hand, the traits that constitute the hyperthymic profile are so desirable that normal individuals tend to endorse them; in a recent Pisa–San Diego collaboration (22) involving 1010 students aged between 14 and 26, of whom 8.2 per cent met the full criteria for hyperthymic temperament, all participants scored between the first and second positive standard deviation. Thus, more work needs to be done on the psychometric standardization of this temperamental construct; on the other hand, all studies are consistent in showing marked male predominance.

Diagnostic aspects

On the positive side, hyperthymic individuals are enterprising, ambitious, and driven, often achieving considerable social and vocational prominence. (105) Abuse of stimulants is not so much an attempt to ward off depression and fatigue as an effort to enhance their already high-level drive and, sometimes, to further curtail their already reduced need for sleep. Hyperthymic individuals typically marry three or more times. Others, without entering into legally sanctioned matrimony, form three or more families in different cities; these men are capable of maintaining such relationships for long periods, testifying to their financial and personal resourcefulness, as well as their generosity towards their lovers and the offspring from such unions. Unlike the antisocial psychopath who is predatory on others and neglects or abuses his women and children, these men care for their loved ones. But obviously the ‘arrangement' involving women of different generations is complex, and a fertile soil for jealousy, drama, scandal, and tragedy. Nonetheless, it is not uncommon to see more than three or four women crying profusely and expressing their common grief at the funerals of these men!

Although individuals with hyperthymia optimally enjoy the advantage of their reduced need for sleep (giving more time and energy to invest in work and pleasure), some present clinically because of insomnia. Thus, in a predominantly male sample of executives presenting to a sleep centre, (105) habitual sleep need was 4 to 5 h; however, they had been intermittently bothered by ‘nervous energy' and difficulty falling asleep. Now, in late middle age, alcohol was no longer an effective hypnotic. Although they vigorously denied depressive and other mental symptoms—indeed, they had extremely low scores on self-rated depression—spouses or lovers provided collateral information about brief irritable–depressive dips, especially in the morning and, in some cases, more protracted ‘fatigue states' of days to weeks during which the subject would vegetate. Despite these depressive dips, these patients were distinguished from the constantly shifting moods of cyclothymic patients by the fact that the depressions arose from a baseline of trait hypomania of a more or less stable course. Our most current diagnostic guidelines for a hyperthymic temperament consist of the following traits on a habitual basis since at least early adulthood: cheerful, overoptimistic, or exuberant; extraverted and people-seeking, often to the point of being overinvolved or meddlesome; overtalkative, eloquent, and jocular; uninhibited, stimulus-seeking, and sexually-driven; vigorous, full of plans, improvident; overconfident, self-assured, and boastful attitudes that may reach grandiose proportions.

A systematic retrospective review of the case records of people with manic depression, whose course was dominated by manic episodes, was recently undertaken in Munich, (107) yielding attributes overlapping with our proposed list: active, vivid, extraverted, verbally aggressive, self-assured, strong willed, engaged in self-employed professions, risk-taking, sensation-seeking, breaking social norms, spendthrift, and generous. The fact that at least 10 per cent of patients with major depression in an Italian study (108) could be characterized as premorbidly hyperthymic, suggests that this temperament has relevance to both major affective poles. This is an important diagnostic consideration, because rather than being considered narcissistic depressions, these should be recognized as a soft bipolar variant. (8)

Aetiological aspects

It is of interest that Gardner's ethological analysis (109) of what constitutes ‘leadership' led to a description that overlaps with a hyperthymic profile: cheerfulness, joking, irrepressible infectious quality, unusual warmth, expansive, strong sense of confidence in one's abilities, scheming, robust, tireless, pushy, and meddlesome. Hypothetically, this temperament evolved in primates whose social life required leadership to better face challenges to the group from within and without.

In the only sleep electroencephalography study conducted to date, (105) REM latency was found to be shortened; similar findings have been reported on the sleep of manic patients,(110) thereby supporting the notion of a trait–state continuum at the neurophysiological level. (Although counterintuitive, this neurophysiological marker appears to be shared between the depressive and hyperthymic poles.) Finally, family history for frank bipolar disorder characterizes many such individuals. The foregoing data, albeit limited, suggest that hyperthymic traits share several key biological underpinnings of affective disorder.

Course and treatment

Little is known about the natural course of the hyperthymic temperament, except what can be reconstructed retrospectively from biographical and clinical studies. Given their overoptimistic and self-assured style of thinking, these individuals feel perfectly fit in all areas of functioning and thus have no need to consult a psychiatrist. They do so only when forced by loved ones. There are no systematic treatment studies on hyperthymia. Anectodally, (111) low doses of anticonvulsants such as valproate (e.g 500–750 mg/day) can be useful in reducing drivenness, deemed to be damaging to the cardiovascular system, or the enormous sexual appetite that places them at risk for social scandals and, in some cases, exposure to HIV infection. (112) Drug-abusing subjects with hyperthymia can be detoxified with valproate, carbamazepine, or gabapentin. Clinically depressed subjects with a hyperthymic temperament often respond poorly to antidepressants. In our opinion, the efficacy of lithium augmentation in resistant depression is partly based on the high prevalence of hyperthymia in resistant populations; (113) it would be wise to keep the dose of lithium in augmentation in such individuals to a lower to middle range (i.e. 600–900 mg/day).

People with hyperthymia are action-oriented, and are not inclined to any type of self-examination. Furthermore, their hypertrophied sense of denial (114) makes them poor candidates for psychotherapy. The physician must, none the less, attempt psychoeducation about the harm that can come to them and their loved ones because of their temperamental excesses. Alcohol consumption, which is common in these individuals, should not be abruptly interrupted because of the risk of the switch to a suicidal depression. If detoxification is necessary for health reasons, admission to a suitable inpatient facility should be arranged. The occasion might be profitably used for whatever counselling is deemed appropriate for life and health situations confronting them at the time. 


1. Akiskal, H.S., Judd, L.L., Gillin, J.C., and Lemmi, H. (1997). Subthreshold depressions: clinical and polysomnographic validation of dysthymic, residual and masked forms. Journal of Affective Disorders, 45, 53–63.

2. Meier, W., Lichterman, D., Minges, J., Heun, R., and Mallmayer, J. (1992). The risk of minor depression in families of probands with major depression: sex differences and familiality. European Archives of Psychiatry and Clinical Neuroscience, 242, 89–92.

3. Kendler, K.S., Neale, M.C., and Kessler, R.C. (1992). A population-based twin study of major depression in women: the impact of varying definitions of illness. Archives of General Psychiatry, 49, 257–66.

4. Remick, R.A., Sadovnick, A.D., Lam, R.W., Zis, A.P., and Yee, I.M. (1996). Major depression, minor depression, and double depression: are they distinct clinical entities? American Journal of Medical Genetics, 67, 347–53.

5. Akiskal, H.S., Bitar, A.H., Puzantian, V.R., Rosenthal, T.L., and Walker, P.W. (1978). The nosological status of neurotic depression: a prospective three-to-four year examination in light of the primary–secondary and unipolar–bipolar dichotomies. Archives of General Psychiatry, 35, 756–66.

6. Bronisch, T., Wittchen, H.-U., Krieg, C., Rupp, H.-U., and von Zerssen, D. (1985). Depressive neurosis: a long-term prospective and retrospective follow-up study of former inpatients. Acta Psychiatrica Scandinavica, 71, 237–48.

7. Akiskal, H.S. and Cassano, G.B. (ed.) (1997). Dysthymia and the spectrum of chronic depressions. Guilford Press, New York.

8. Akiskal, H.S. (1996). The prevalent clinical spectrum of bipolar disorders: beyond DSM-IV. Journal of Clinical Psychopharmacology, 16 (Supplement), 4s–14s.

9. Hantouche, E.G., Akiskal, H.S., Lancrenon, S., et al. (1998). Systematic clinical methodology for validating bipolar-II disorder: data in mid-stream from a French national multisite study (EPIDEP). Journal of Affective Disorders, 50, 163–73.

10. Akiskal, H.S. and Akiskal, K. (1996). The temperamental foundations of mood disorders. In Interpersonal factors in the origin and course of affective disorders (ed. C.H. Mundt), pp. 3–30. Gaskell, London.

11. Akiskal, H.S. and Weise, R.E. (1992). The clinical spectrum of so-called ‘minor' depressions. American Journal of Psychotherapy, 46, 9–22.

12. Lewis, A. (1936). Melancholia: a prognostic study. Journal of Mental Science, 82, 488–558.

13. Herpertz, S., Steinmeyer, E.M., and Sa, H. (1998). On the conceptualisation of subaffective personality disorders. European Psychiatry, 13, 9–17.

14. Eysenck, H.F. (1987). The definition of personality disorders and the criteria appropriate for their description. Journal of Personality Disorder, 1, 211–19.

15. Snaith, R.P. (1991). Measurement in psychiatry. British Journal of Psychiatry, 159, 78–82.

16. Kendler, K.S., Neale, M.C., Kessler, R.C., Heath, A.C., and Eaves, L.J. (1993). A longitudinal twin study of personality and major depression in women. Archives of General Psychiatry, 50, 853–62.

17. Akiskal, H.S. (1981). Subaffective disorders: dysthymic, cyclothymic and bipolar II disorders in the ‘borderline' realm. Psychiatric Clinics of North America, 4, 25–46.

18. Brieger, P. and Marneros, A. (1997). Dysthymia and cyclothymia: historical origins and contemporary development. Journal of Affective Disorders, 45, 117–26.

19. Kraepelin, E. (1921). Manic–depressive insanity and paranoia. Livingstone, Edinburgh.

20. Schneider, K. (1958). Psychopathic personalities. Charles C. Thomas, Springfield, IL.

21. Akiskal, H.S. (1983). Dysthymic disorder: psychopathology of proposed chronic depressive subtypes. American Journal of Psychiatry, 140, 11–20.

22. Placidi, G.F., Signoretta, S., Liguori, A., Gervasi, R., Maremmani, I., and Akiskal, H.S. (1998). The Semi-Structured Affective Temperament Interview (TEMPS-I): reliability and psychometric properties in 1010 14–26 year students. Journal of Affective Disorders, 47, 1–10.

23. Akiskal, H.S., Placidi, G.F., Signoretta, S., et al. (1998). TEMPS-I. Delineating the most discriminant traits of cyclothymic, depressive, irritable and hyperthymic temperaments in a nonpatient population. Journal of Affective Disorders, 51, 7–19.

24. Kasahara, Y. (1991). The practical diagnosis of depression in Japan. In The diagnosis of depression (ed. J.P. Feighner and W.F. Boyer), pp. 163–75. Wiley, Chichester.

25. Montassut, M. (1938). La dépression constitutionnelle: l'ancienne neurasthénie dans ses rapports avec la médecine générale; clinique biologie, thérapeutique. Masson, Paris.

26. Akiskal, H.S. (1996). Dysthymia as a temperamental variant of affective disorder. European Psychiatry, 11 (Supplement), 117s–22s.

27. Akiskal, H.S., Rosenthal, T.L., Haykal, R.F., Lemmi, H., Rosenthal, R.H., and Scott-Strauss, A. (1980). Characterological depressions: clinical and sleep EEG findings separating ‘subaffective dysthymias' from ‘character-spectrum' disorders. Archives of General Psychiatry, 37, 777–83.

28. Devinand, D.P., Nobler, M.S., Singer, T., et al. (1994). Is dysthymia a different disorder in the elderly? American Journal of Psychiatry, 151, 1592–9.

29. Tellenbach, H. (1980). Melancholia. Duquesne University Press, Pittsburgh, PA.

30. Kretschmer, E. (1936). Physique and character. Kegan Paul, Trench, Trubner, London.

31. Kernberg, O.F. (1970). A psychoanalytic classification of character pathology. Journal of the American Psychoanalytic Association, 18, 800–22.

32. Keller, M.B., Klein, D.N., Hirschfeld, R.M.A., et al. (1995). Results of the DSM-IV mood disorders field trial. American Journal of Psychiatry, 152, 843–9.

33. Seretti, A., Jori, M.C., Cassadei, J., Ravizza, L., Smeraldi, E., and Akiskal, H.S. Delineating psychopathologic clusters within dysthymia: a study of 512 patients without major depression. Journal of Affective Disorders, in press.

34. Madhulika, A., Gupta, M.D., and Moldofsky, H. (1986). Dysthymic disorder and rheumatic pain modulation disorder (fibrositis syndrome): a comparison of symptoms and sleep physiology. Canadian Journal of Psychiatry, 31, 608–16.

35. Tyrer, P., Seivewright, N., Ferguson, B., and Tyrer, J. (1992). The general neurotic syndrome: a coaxial diagnosis of anxiety, depression and personality disorder. Acta Psychiatrica Scandinavica, 85, 201–6.

36. Keller, M.B., Lavori, P.W., Endicott, J., Coryell, W., and Lerman, G.L. (1983). Double depression: two-year follow-up. American Journal of Psychiatry, 140, 689–94.

37. Kovacs, M., Akiskal, H.S., Gatsonis, C., and Parrone, P.L. (1994). Childhood-onset dysthymic disorder: clinical features and prospective naturalistic outcome. Archives of General Psychiatry, 51, 365–74.

38. Rosenthal, T.L., Akiskal, H.S., Scott-Strauss, A., Rosenthal, R.H., and David, M. (1981). Familial and developmental factors in characterological depressions. Journal of Affective Disorders, 3, 183–92.

39. Rihmer, Z. (1993). Dysthymia: a clinician's perspective. In Dysthymic disorder (ed. S. Burton and H.S. Akiskal), pp. 112–25. Royal College of Psychiatrists, London.

40. Klein, D.N., Taylor, E., Harding, K., et al. (1988). Double depression and episodic major depression: demographic, clinical, familial, personality, and socioenvironmental characteristics and short-term outcome. American Journal of Psychiatry, 145, 1226–31.

41. Judd, L.L., Akiskal, H.S., Maser, J.D., et al. (1998). A prospective 12-year study of sub-syndromal and syndromal depressive symptomatology in patients with unipolar major depressive disorder. Archives of General Psychiatry, 55, 694–700.

42. Akiskal, H.S. (1994). Dysthymia: clinical and external validity. Acta Psychiatrica Scandinavica, 89 (Supplement), 19–23.

43. Weissman, M.M., Leaf, P.J., Bruce, M.L., and Florio, L. (1988). The epidemiology of dysthymia in five communities: rates, risks, comorbidity, and treatment. American Journal of Psychiatry, 145, 815–19. 44. Waintraub, L. and Guelfi, J.D. (1998). Nosological validity of dysthymia. Part I, historical, epidemiological and clinical data. European Psychiatry, 13, 173–80.

45. Perugi, G., Akiskal, H.S., Musetti, L., Simonini, E., and Cassano, G.B. (1994). Social adjustment in panic-agoraphobic patients reconsidered. British Journal of Psychiatry, 164, 88–93.

46. Haykal, R. and Akiskal, H.S. (1999). The long-term outcome of dysthymia in private practice. Clinical features, temperament, and the art of management. Journal of Clinical Psychiatry, 60, 508–18.

47. Wells, K.B., Stewart, A., Hays, R., et al. (1989). The functioning and wellbeing of depressed patients: results from the medical outcomes study. Journal of the American Association, 262, 914–19.

48. The WPA Dysthymia Working Group (1995). Dysthymia in clinical practice. British Journal of Psychiatry, 166, 174–83.

49. Licinio, J., Prilpko, I., and Bolis, C.L. (ed.) (1997). Dysthymia in neurological disorders. In Proceedings of the WHO Meeting. World Health Organization, Geneva.

50. Arieti, S. and Bemporad, J. (1978). Severe and mild depression. Basic Books, New York.

51. Akiskal, H.S., Lemmi, H., Dickson, H., King, D., Yerevanian, B.I., and VanValkenburg, C. (1984). Chronic depressions: Part 2. Sleep EEG differentiation of primary dysthymic disorders from anxious depressions. Journal of Affective Disorders, 6, 287–95.

52. Howland, R.H. and Thase, M.E. (1991). Biological studies of dysthymia. Biological Psychiatry, 30, 283–304.

53. Klein, D.N., Riso, L.P., Donaldson, S.K., et al. (1995). Family study of early-onset dysthymia. Mood and personality disorders in relatives of outpatients with dysthymia and episodic major depression and normal controls. Archives of General Psychiatry, 52, 487–96.

54. Hamburg, S.R. (1998). Inherited hypohedonia leads to learned helplessness: a conjecture updated. Reviews of General Psychology, 2, 384–403.

55. Giles, D.E., Kupfer, D.J., and Roffwarg, H.P. (1988). Polysomnographic parameters in first-degree relatives of unipolar probands. Psychiatry Research, 27, 127–36.

56. Kreig, J.C., Lauer, C.J., and Hermle, L. (1993). Psychometric, polysomnographic, and neuroendocrine measures in subjects at high risk for psychiatric disorders: preliminary results. Neuropsychobiology, 23, 57–67.

57. Burton, S.W. and Akiskal, H.S. (ed.) (1993). Dysthymic disorder. Gaskell and Royal College of Psychiatrists, London.

58. McCullough, J.P., McCune, K.J., Kaye, A.L., et al. (1994). One-year prospective replication study of an untreated sample of community dysthymia subjects. Journal of Nervous and Mental Disease, 182, 396–401.

59. Vallejo, J., Gasto, C., Catalan, R., and Salamero, M. (1987). Double-blind study of imipramine versus phenelzine in melancholias. British Journal of Psychiatry, 151, 639–42.

60. Kocsis, J.H., Zisook, S., Davidson, J., et al. (1997). Double-blind comparison of sertraline, imipramine, and placebo in the treatment of dysthymia: psychosocial outcomes. American Journal of Psychiatry, 154, 390–5.

61. Stewart, J., Quitkin, F.M., McGrath, P.J., et al. (1989). Social functioning in chronic depression: effect of 6 weeks of antidepressants treatment. Psychiatry Research, 25, 213–22.

62. Bakish, D., Lapierre, Y.D., Weinstein, R., et al. (1993). Ritanserin, imipramine, and placebo in the treatment of dysthymic disorder. Journal of Clinical Psychopharmacology, 13, 409–14.

63. Thase, M.E., Fava, M., Halbreich, U., et al. (1996). A placebo-controlled, randomized clinical trial comparing sertraline and imipramine for the treatment of dysthymia. Archives of General Psychiatry, 53, 777–84.

64. Vanelle, J.M., Attar-Levy, D., Poirier, M.F., Bouhassira, M., Blin, P., and Olié, J.P. (1997). Controlled efficacy study of fluoxetine in dysthymia. British Journal of Psychiatry, 170, 345–50.

65. Lecrubier, Y., Boyer, P., Turjanski, S., and Rein, W. (1997). Amisulpride versus imipramine and placebo in dysthymia and major depression. Journal of Affective Disorders, 43, 95–103.

66. Versiani, M., Amrein, R., and Stabl, M. (1997). Moclobemide and imipramine in chronic depression (dysthymia): an international double-blind, placebo-controlled trial. Clinical Psychopharmacology, 12, 183–93.

67. Ravindran, A.V., Bialik, R.J., and Lapierre, Y.D. (1994). Therapeutic efficacy of specific serotonin re-uptake inhibitors (SSRIs) in dysthymia. Canadian Journal of Psychiatry, 39, 21–6.

68. Kocsis, J.H., Zisook, S., Davidson, J., et al. (1997). Double-blind comparison of sertraline, imipramine and placebo in the treatment of dysthymia: psychosocial outcomes. American Journal of Psychiatry, 154, 390–5.

69. Knutson, B., Wolkowitz, W., Cole, S.W., et al. (1998). VI. Selective alteration of personality and social behavior by serotonergic intervention. American Journal of Psychiatry, 155, 373–9.

70. Andrews, W., Parker, G., and Barrett, E. (1998). The SSRI antidepressants: exploring their ‘other' possible properties. Journal of Affective Disorders, 49, 141–4.

71. Markowitz, J.C. (1994). Psychotherapy of dysthymia. American Journal of Psychiatry, 151, 1114–21.

72. McGrath, E., Keita, G.P., Strickland, B.R., and Russo, N.P. (ed.) (1993). Women and depression: risk factors and treatment issues. American Psychological Association, Washington, DC.

73. Akiskal, H.S. (1992). Psychopharmacological and psychotherapeutic strategies in intermittent and chronic affective conditions. In Long-term treatment of depression (ed. S.A. Montgomery and F. Rouillon), pp. 245–63. Wiley, Chichester.

74. Ravindran, A.V., Griffiths, J., Waddell, C., and Anisman, H. (1995). Stressful life events and coping styles in relation to dysthymia and major depressive disorder: variations associated with alleviation of symptoms following pharmacotherapy. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 19, 634–53.

75. Kraemer, P.D. (1993). Listening to Prozac. Viking Press, New York.

76. Broadhead, W.E., Blazer, D.G., George, L.K., and Tse, C.K. (1993). Depression, disability, and days lost from work in a prospective epidemiological survey. Journal of the American Medical Association, 264, 2524–8.

77. Horwath, E., Johnson, J., Klerman, G.I., and Weissman, M.M. (1992). Depressive symptoms as relative and attributable risk factors for first-onset major depression. Archives of General Psychiatry, 49, 817–23.

78. Akiskal, H.S., King, D., Rosenthal, T.L., Robinson, D., and Scott-Strauss, A. (1981). Chronic depressions: Part 1. Clinical and familial characteristics in 137 probands. Journal of Affective Disorders, 3, 297–315.

79. Akiskal, H.S., Djenderedjian, A.H., Rosenthal, R.H., and Khani, M.K. (1977). Cyclothymic disorder: validating criteria for inclusion in the bipolar affective group. American Journal of Psychiatry, 134, 1227–33.

80. Akiskal, H.S., Khani, M.K., and Scott-Strauss, A. (1979). Cyclothymic temperamental disorders. Psychiatric Clinics of North America, 2, 527–54.

81. Gillin, J.C., Mazure, C., Post, R.M., Jimerson, D., and Bunney, W.E., Jr (1977). An EEG sleep study of a bipolar (manic–depressive) patient with a nocturnal switch process. Biological Psychiatry, 12, 711–18.

82. Akiskal, H.S. (1992). Delineating irritable-choleric and hyperthymic temperaments as variants of cyclothymia. Journal of Personality Disorders, 6, 326–42.

83. Stone, M.H. (1980). The borderline syndrome: constitution, personality and adaptation. McGraw-Hill, New York.

84. Levitt, A.J., Joffe, R.T., Ennis, J., MacDonald, C., and Kutcher, S.P. (1993). The prevalence of cyclothymia in borderline personality disorder. Journal of Clinical Psychiatry, 51, 335–9.

85. Gawin, F.H. and Ellinwood, E.H., Jr (1988). Cocaine and other stimulants. Actions, abuse, and treatment. New England Journal of Medicine, 318, 1173.

86. Mirin, S.M., Weiss, R.D., Griffin, M.L., and Michael, J.L. (1991). Psychopathology in drug abusers and their families. Comprehensive Psychiatry, 32, 36–51.

87. Brady, K.T. and Sonne, S.C. (1995). The relationship between substance abuse and disorder. Journal of Clinical Psychiatry, 56 (Supplement), 19–24.

88. Akiskal, H.S. and Akiskal, K. (1988). Re-assessing the prevalence of bipolar disorders: clinical significance and artistic creativity. Psychiatrie et Psychobiologie, 3, 29s–36s.

89. Angst, J. (1998). The emerging epidemiology of hypomania and bipolar II disorder. Journal of Affective Disorders, 50, 143–51.

90. Akiskal, H.S., Hantouche, E., Bourgeois, M., et al. (1998). Gender, temperament and the clinical picture in dysphoric mixed mania: findings from a French national study (EPIMAN). Journal of Affective Disorders, 50, 175–86.

91. Akiskal, H.S., Maser, J.D., Zeller, P., et al. (1995). Switching from ‘unipolar' to bipolar II: an 11-year prospective study of clinical and temperamental predictors in 559 patients. Archives of General Psychiatry, 52, 114–23.

92. Akiskal, H.S., Downs, J., Jordan, P., Watson, S., Daugherty, D., and Pruitt, D.B. (1985). Affective disorders in the referred children and younger siblings of manic–depressives: mode of onset and prospective course. Archives of General Psychiatry, 42, 996–1003.

93. Depue, R.A., Slater, J.F., Wolfstetter-Kausch, H., et al. (1981). A behavioral paradigm for identifying persons at risk for bipolar depressive disorder: a conceptual framework and five validation studies. Journal of Abnormal Psychology, 90, 381–437.

94. Wicki, W. and Angst, J. (1991). The Zurich study. X. Hypomania in a 28–30-year-old cohort. European Archives of Psychiatry and Clinical Neuroscience, 240, 339–48.

95. Casey, P.R. and Tyrer, P.J. (1986). Personality, functioning and symptomatology. Journal of Psychiatry Research, 20, 363–74.

96. Weissman, M.M. and Myers, J.K. (1978). Affective disorders in a US urban community, the use of research diagnostic criteria in an epidemiological survey. Archives of General Psychiatry, 35, 1304–11.

97. Gershon, F.S., Hamovit, J., Guroff, J.J., et al. (1982). A family study of schizoaffective, bipolar I, bipolar II, unipolar, and normal control probands. Archives of General Psychiatry, 39, 1157–67.

98. Bertelsen, A., Harvald, B., and Hauge, M. (1977). A Danish twin study of manic–depressive disorders. British Journal of Psychiatry, 130, 330–51.

99. Jacobsen, F.M. (1993). Low-dose valproate: a new treatment for cyclothymia, mild rapid cycling disorders, and premenstrual syndrome. Journal of Clinical Psychiatry, 54, 229–34.

100. Deltito, J. (1993). The effect of valproate on bipolar spectrum temperamental disorder. Journal of Clinical Psychiatry, 54, 300–4.

101. Peselow, E.D., Dunner, D.L., Fieve, R.R., and Lautin, A. (1982). Lithium prophylaxis of depression in unipolar, bipolar II, and cyclothymic patients. American Journal of Psychiatry, 139, 747–52.

102. Rifkin, A., Quitkin, F., Carrillo, C., et al. (1972). Lithium carbonate in emotionally unstable character disorder. Archives of General Psychiatry, 27, 519–23.

103. Akiskal, H.S. (1994). Dysthymic and cyclothymic depressions: therapeutic considerations. Journal of Clinical Psychiatry, 55, 46–52.

104. Coid, J.W. (1993). An affective syndrome in psychopaths with borderline personality disorder? British Journal of Psychiatry, 162, 641–50.

105. Akiskal, H. (1984). Characterologic manifestations of affective disorders: toward a new conceptualization. Integrative Psychiatry, 2, 83–8.

106. Eckblad, M. and Chapman, L.J. (1986). Development and validation of a scale for hypomanic personality. Journal of Abnormal Psychology, 95, 214–22.

107. Pössi, J. and von Zerssen, D. (1993). A case history analysis of the ‘manic type' and the ‘melancholic type' of premorbid personality in affectively ill patients. European Archives of Psychiatry and Neurological Science, 239, 347–55.

108. Cassano, G.B., Akiskal, H.S., Savino, M., Musetti, L., Perugi, G., and Soriani, A. (1992). Proposed subtypes of bipolar II and related disorders: with hypomanic episodes (or cyclothymia) and with hyperthymic temperament. Journal of Affective Disorders, 26, 127–40.

109. Gardner, R. Jr (1982). Mechanisms in manic–depressive disorder: an evolutionary model. Archives of General Psychiatry, 39, 1436–41.

110. Hudson, J.I., Lipinski, J.F., Franjenburg, F.R., Grochocinski, V.J., and Kupfer, D.J. (1988). Electroencephalographic sleep in mania. Archives of General Psychiatry, 45, 267–73.

111. Akiskal, H.S. (1997). Chronic disturbances of temperament. Bibliotheca Psychiatrica (Basel), 167, 29–32.

112. Perretta, P., Akiskal, H.S., Nisita, C., et al. (1998). The high prevalence of bipolar II and associated cyclothymic and hyperthymic temperaments in HIV patients. Journal of Affective Disorders, 50, 215–24.

113. Akiskal, H.S. and Mallya, G. (1987). Criteria for the ‘soft' bipolar spectrum: treatment implications. Psychopharmacology Bulletin, 23, 68–73.

114. Akhtar, S. (1988). Hypomanic personality disorder. Integrative Psychiatry, 6, 37–52.