CLINICAL AND EVOLUTIONAL ASPECTS IN BIPOLAR DISORDER, MANIC EPISODE (2)

Acknowledgement:
“This paper is published under the frame of European Social Fund, Human Resources Development Operational P r o g r a m m e 2 0 0 7 – 2 0 1 3 , p r o j e c t n o . POSDRU/159/1.5/S/138776″/ “Acest articol este publicat prin Fondul Social European, Programul Operational Dezvoltarea Resurselor Umane 2007-2013, contract nr. POSDRU/159/1.5/S/138776”

Cycle length is defined as the time from the onset of one episode to the onset of the next. Variations in cycle length reflect primarily variations in the length of the symptom- free interval, because the duration of episodes tends to be relatively constant in a given individual. Onset is used to calculate cycle length because it is generally easier to identify than termination of an episode. In addition, treatment can easily obscure an episode’s natural length. Most investigators agree that cycle length tends to grow shorter with subsequent recurrences; that is, episodes become more frequent. What is not clear is the proportion of bipolar patients whose illness accelerates. After three to five episodes, however, the extent of shortening slows down considerably and approaches a levelling off or maximum frequency of episodes. Episode frequency might represent a familial trait, as suggested by Gershon and colleagues (1).
Kraepelin (2) was the first to report that intervals of euthymia appear to decrease in duration (in manic- depressive illness), with increasing numbers of affective episodes. Later this observation became a central impetus for the kindling model, with its prediction that episodes become more frequent over time, and that euthymic intervals between episodes become increasingly shorter. This observation was subsequently confirmed in a number of clinical studies.
Despite substantial individual variability, the averages show a remarkably consistent pattern across the studies. Note the pattern in 105 bipolar patients of Zis and colleagues (3): the average cycle length between the first and the second episodes was 36 months, diminishing to about 24 months, then to 12 months. In 46 bipolar patients studied by Roy-Byrne and colleagues ( 4), the mean cycle lengths for the first seven episodes were 53, 28, 25, 20, 12,
15, and 9 months. In 95 bipolar patients from Angst’s clinic in Zurich, the median first cycle length was 48 months, compared to 22 months for the second cycle, 24 for the third, 14 for the fourth, and 12 for the fifth (5). Taschev (6) found the average second cycle to be as long as long as the first, but the fourth cycle to be half as long as the second.
A prospective study of patients in the 1980s (7) also documented the increasing probability of an early relapse with each episode, a finding suggesting that some fundamental features of the illness evolution persist despite treatment. Methodological considerations of Slater’s fallacy –must be taken into account when evaluating these studies.
Recent studies did not agree on the progressive shortening of cycle length. In a reanalysis of Zurich’s group (identified in 1959-1963, followed-up until 1997, N = 220 bipolar patients), Angst and Selloro (8) discovered the shortening of cycle length in the first couple of episodes, but not in the later episodes. In the studies carried out by Winokur and colleagues (9) and Turvey and colleagues (10), which analyzed the data of the same large NIMH group, bipolar patients did not seem to have more frequent or shorter episodes over time and there was not any correlation between poor results and such an evolution. Rather, poor results were associated to “polyphasic” mood episodes they meaning the immediate passage from mania to depression and again to mania without euthymia.
The results of other studies suggest that at least a subgroup of patients suffering from bipolar disorder will experience a progressive reduction of cycle length. Goldberg and Harrow (11) reported that 50 percent of the 20 bipolar patients having two or more prior hospitalizations experienced intervals between episodes shorter than 1 year (defined as “kindlers”), whereas other 19 patients having less than two hospitalizations experienced intervals between episodes with a duration longer than 1 year (“nonkindlers”). After 4.5 years of follow-up, “kindlers” had more recurrences than “nonkindlers”. Kessing and colleagues (12) found that 40 percent of a large sample (N = 1712) experienced “shortening” between the first and second, and 25 percent of the same sample showed a progressive shortening along two consecutive intervals.
The complexity of this subject increases when we pay attention to the methodological issue mentioned above approached before by Eliot Slater, in 1938, but ignored by most researchers until recently. “Slater’s fallacy”, revived by Oepen and colleagues (13), postulates that patients having more episodes tend to have shorter cycle lengths, and those with less episodes to have longer cycle lengths. Thus, if the two groups are put together, as in Kraepelin’s original data (2), it will seems that the cycle length decreases in more episodes for the sample taken as a whole, whereas this apparent effect maybe an artifact of putting together the two distinct patient subgroups, those with few and episodes and those with many episodes. To really demonstrate the shortening of cycle length, one should adjust the number of episodes.
In two studies (14, 15), the correction of the episode number led to similar results with those reported by Slater, namely no consistent evidence of progressive shortening of wellness intervals (euthymia). In other two studies (4,11), a pattern of “awareness” or “kindling” was identified in about half of the patients suffering from refractory bipolar disorder, but not in others. In a large Norwegian study with hospitalized bipolar patients, half of those having three or less episodes seemed to experience the progressive shortening of wellness intervals, but a similar model was observed in only 25 percent of the patients with four or more episodes (12). Subsequently, in these studies, the correction of episode frequency showed the decrease of euthymia length (wellness intervals) mainly in the first three episodes, with unchanged wellness intervals of about one per year for the next episodes. This was in fact what the Goodwin group had found out beforehand (16). These findings suggest that the studies of wellness intervals which do not manage to catch the first three episodes of the illness have very few chances to notice any “kindling” effect. The data from McLean- Harvard First-Episode Mania Study do not show evidence of shorter wellness intervals, though the follow- up period was short (2-4 years) (17).
Obviously, “kindling” model does not depend only on the prediction of shortening of euthymic wellness episodes. The model also makes other predictions such as the increase of episode frequency, the increase of episode length over time, the decrease of importance psychosocial triggers over time and the increase of resistance to treatment once with the increase of episode number (or different efficacy in the early phases as compared to tardy illness phases). An important aspect that bears the generalization of kindling model is the proportion of patients with bipolar disorder whose evolution is in concordance with these predictions. Research on psychosocial triggers of episodes showed interesting data in this respect. These results indicate the observable decline in the association between stressful life events and the depressive onset with an increasing number of prior depressive episodes.
The increase of resistance to treatment across episode number was reported in some studies (18), but not in others (19). Kessing and colleagues (20) did not manage to find other evidence of result worsening, such as increase of mortality associated to a model of awareness of progressive shortening of cycle length, though there was an interesting association between kindling model and an increase dementia risk (21). We remind that kindling model predicts that results will be worse after several episodes. We would expect that the results of evolution of those treated and untreated should reflect this model. This prediction was supported by some but not all studies. Gitlin and colleagues (22) found that more previous episodes predict the earlier relapse after 4.3 years of follow-up (2.2 as compared to 3.5 years, p = 0.007), a similar finding to that shown previously by Goldberg and Harrow (11)). On the other hand, in a study of 4.6 years on 360 bipolar-I patients (n = 220) and bipolar-II patients (n = 140) under lithium treatment, Baldessarini and collaborators (19) do not confirm the prediction of kindling model of worsening of results for each successive episode.
The relationship between the frequency of episodes and the age at onset remains somehow unclear. 29 older studies found an increase of relapse frequency for the higher age at onset, but two studies did not demonstrate this association (23, 4), and they found the contrary (24,
25). Grof and colleagues (26) showed that patient’s age and the onset age each contributed independently to the prediction of relapse (3). Onset at the age of 20 was associated to a probability of 20 percent of relapse within 24 months, the onset at the age of 30 to a probability of 50 percent and the tardy onset (the age of 50 or more) to a very high probability (80 percent). Only the length of first cycle (the time interval between the first and the second episode) was linked to the age at onset. 31 Another question about the cycling model in bipolar illness is whether burnout may appear. Kraepelin (2) noticed that the disease tends to slow down after the fourth decade, though he did not give details about this observation. In a prospective follow-up study on 215 bipolar patients (150 bipolar-I, 65 bipolar-II) for a period of 17- 21 years, Angst (27) did not find a decrease of episode frequency related to age; 26 percent of bipolar patients (as compared to 42 percent of unipolar patients) had no relapse for more than
5 years, despite the fact that most patients were still actively ill, at the age of 60, when the follow-up ended. At the same time, the results of recent studies do not generally support the concept of burnout (28), though long-term follow-up since maturity until old age is rare.
Often, the onset of manic episodes is abrupt, developing over a few days. Depressive episodes develop more gradually, over weeks (29, 30, 31). Some patients experience a “hypomanic alert” (21) a period of days or even weeks of hypomanic symptoms before the switch into mania. The most common prodrome is sleep disturbance, which had a prevalence of 77 percent in one study (32). Family members are twice more likely to observe early behavioral symptoms of mania (31).
Many studies suggested that episode polarity at onset of bipolar disorder may have a predictive value of certain aspects related to the clinical features of life. Authors examine this possibility on a large well chosen sample of bipolar I patients. Knowing the onset pole may help the clinician supply prognosis information and management consultancy for a bipolar person. (33)
Estimates of the proportion of bipolar patients who begin the illness with a manic episode (34-79 %) present an average just over 50 percent. It is important to recognize that the other half of patients with bipolar form of manic- depressive illness will go undiagnosed as bipolar at onset because they initially experience only major depressive episodes (21); this condition is sometimes called “false” unipolar depression.
One study of false unipolar depression examined 17,447 patients with mood disorders who were hospitalized in Denmark for a 22-year period (1971-1993) (34). Among those whose first episodes were depressive, if a manic episode was to develop, it usually did so within 5 years. Patients with false unipolar depression experienced more recurrent depressive episodes than did patients with true unipolar depression. 34 It is also important to recognize that there is a relationship between false unipolar depression and age at onset. An often-cited figure is that 1 to 2 percent of patients with depression may experience a hypomanic/manic episode in every year of follow-up. This figure is based on the CDS group (35), in which 10.2 percent of a depressed subgroup (mean age 36.8 years) had a manic or hypomanic episode during 10 years of follow-up. A younger age at onset within this group was associated with increased risk of manic/hypomanic switch. Other younger groups have confirmed higher natural switch rates to mania.
In a group of 74 depressed young adults (mean age 23.0 years), Goldberg and colleagues (36) found that 46 percent had experienced a manic or hypomanic episode at 15-year follow-up. Thus the risk of switching from false unipolar depression is highest in childhood and young adulthood, occurring at a rate of about 3 to 5 percent a year; it then decreases by the late 30s, at which point it flattens to about 1 percent per year.
Studies showing the highest estimates of manic onset use first hospitalization as the onset criterion. It is possible that these high estimates result from underestimation of depressive onsets because many depressions do not require hospitalization. On the other hand, hypomanic onsets are probably underestimated compared with depressive onsets, because patients are more likely to experience (and report) depressive symptoms as illness. Angst (37) reported manic onset of illness in the majority of his bipolar sample (N = 95), where onset was defined as the first occurrence of symptoms requiring treatment. On the other hand, Roy-Byrne and colleagues (4), who defined onset as the first symptoms meeting the criteria for an affective episode, found that 60 percent of 71 bipolar patients had a depressive first episode. Perris and d’Elia (38) reported that among patients with a manic first episode 62 percent went on to have a predominantly manic course and only 25 percent a predominantly depressive course. It is interesting that a predominantly manic course may announce a better outcome.
This report is supported by another study of 320 patients with bipolar-I disorder (39), in whom depressive episode onset was associated with higher levels of later rapid- cycling course as well as suicide attempts. In general, these data confirm the existence of some distinct longitudinal models within bipolar I disorder which, in their turn, seem to correlate with polarity at onset. Rapid c y c l i n g a n d m i x e d s t a t e s a p p e a r a s d i s t i n c t psychopathological processes. (40)
Later psychotic features were more likely to be associated with manic onset. In light of these findings of a relationship between polarity at onset and subsequent course, a recent report that the polarity of onset appears to be familial is especially interesting (21). It has generally been observed that women tend to be more likely to have a predominantly depressive pattern to their bipolar illness, and men a predominantly manic pattern. More recent data have not uniformly supported these findings. In one chart review of 131 patients (63 women and 68 men) (40), there were no statistically significant gender differences in age at onset, number of depressive or manic episodes, and number of hospitalizations for depression. However, women had been hospitalized for mania significantly more often than men, whereas substance abuse, although high in both groups, was more prevalent among men.
Also, in a cross-sectional analysis of 500 subjects in the NIMH Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD) study group, there were no gender differences in number of past depressive versus manic episodes (87.4 percent of women and 86.8 percent of men had experienced three or more manic episodes; 92.1 percent of women and 92.3 of men had experienced three or more depressive episodes) (41). As noted earlier in relation to age at onset of depression and false unipolar depression, polarity distinctions are not stable until approximately the age of 40.37 years. Before that age, depressed individuals remain at a relatively high risk of eventual occurrence of manic or hypomanic episodes.
The characteristics of a manic episode my also relate to the illness evolution. In a retrospective evaluation, Swann and colleagues (42) found that mania with depressive features (mixed state) was associated to an early illness onset and more previous episodes as compared to mania with irritable features (but without concomitance of depressive symptoms).
During the 10-year follow-up in a prospective study, clinical data and systematic results were collected from 300 bipolar I and II patients. The sample was divided into two groups depending on the polarity of onset episode (depressive onset [DO] 67% vs. manic/hypomanic onset [MO]) and we compared the clinical features and social functioning between the two patient groups. The patients with a depressive onset DO were more chronic than MO patients, with a higher number of total episodes and longer illness duration. Patients with a depressive onset had a higher number of depressive episodes than MO patients who, in their turn, had more manic episodes. Patients with a depressive onset had more suicide attempts, a later illness onset, were les frequently hospitalized and were less susceptible to develop psychotic symptoms. The depressive onset was more frequent among bipolar II patients. Bipolar I patients with DO had more comorbidities on axis II and were more susceptible to have a history of psychotic symptoms as compared to bipolar II patients with DO. Polarity at onset is a good predictor of polarity of subsequent episodes over time. A depressive onset is twice more frequent as MO and more loaded by chronicity and cyclicity (43).
The topic of episode pattern is of historical importance. Early investigators demonstrated that the three basic models were obvious: mania followed by depression and then a wellness interval (euthymia) (MDI), depression followed by mania and a wellness interval (DMI), and continuous cycling (MDMD). (44) There is evidence in these studies that MDI model may have the best result (70) whereas MDMD model the worst result. DMI model seems to be associated to mania which appears after the treatment of depressive phase by antidepressants (44). Contemporary studies of episode model do not entirely concord with this older literature. In a study of all hospitalized cases of bipolar disorder and unipolar disorder in Denmark over a 22-year period (1971-1993; N = 17447), DMI and MDI evolutions were associated to a similar, number of relapses (34). On the other hand, in a recent reanalysis of CDS data in which 165 bipolar I patients followed for 15 years, Turvey and colleagues (10) reported that mood episodes starting in depression (depression or depression followed by mania only) were associated to poorer results after 15 years than those starting in mania. Also, polarity sequences remained stable over time: 75-80 percent of the individuals whose mood episodes started in mania continued to have mood episodes starting in mania and 55-60 percent of those whose mood episodes started in depression followed the same model. This study is the longest one on polarity. Its findings suggest that patterns of mood episodes are supported over time and that the worse prognosis of depression being at the opposite end of mania diagnosis persists on a long term.
Taken together, more studies (45) considerably support the importance of stressful events in the onset of episodes in bipolar patients. Although some theories give a primary importance of causality to environmental psychosocial factors, now it is generally accepted that environmental psychosocial or physical conditions contribute more to the evolution over time of the mood episodes than to the underlying vulnerability which is genetic to a large extent. Thus, the modern biological theory reaffirms the classic position of Kraepelin (2): “We must consider all potential lesions as possible unloading sparks of individual attacks, but the true cause of the malady must be sought in the permanent internal changes which very often, maybe always, are inborn … Unfortunately, the lack of power of our efforts to cure must convince us too many times of the fact that manic-depressive attacks may be astonishingly independent of external influences”. Early precipitating events influence the “calendar” of an episode, they may in fact activate preexisting vulnerability thus making the person more vulnerable to subsequent episodes as it has been demonstrated recently (21). More recent evidence supports the conclusion that the influence of life events representing the trigger in the onset of mood episodes is much more prominent in the early phases than in the later phases of bipolar disorder.
Precipitants of bipolar disorder episodes:
Conclusions (21):
-All studies have found significantly more stress before manic episodes as compared to different control groups.
-Only one study (46) involved a comparison between patients (e.g. control of inborn vulnerability to stress). This found that more stressful events preceded a manic episode.
-The onset of initial (or early) episodes is much more profitable to be associated to stress than the onset of later episodes (five out of six studies), a thing that concords with kindling hypothesis.
-Work-related events may be very important precipitators of manic or hypomanic relapse.
-Cyclothymia is associated to a high response to stress, in accordance with the initial part of bipolar spectrum (47).
-Manic patients frequently create stressful life events.
-Stressful events often result in loss of sleep which, in its turn, may precipitate mania.
-Stress may result in the increase of alcohol and drug use, both leading afterwards to the loss of sleep and/or mania.
Older literature in terms of the association between life events and relapse is supported by a study on 62 bipolar disorder patients (48), in which a more frequent rate of severe life events (19 percent) was found in the month preceding relapse, as compared to the monthly rate of such events (5 percent), during 2 years of follow-up. In another study, the recovery from mood episodes of 67 patients hospitalized for bipolar I disorder was delayed three times more in the presence of a negative precipitating life event (49). As the association of external life events to mood episodes has been established, recent studies have focused on other aspects of psychosocial circumstances of bipolar disorder persons. For example, Staner and colleagues (50) reported that among 27 patients with bipolar disorder in remission for 1 year, social support and self-esteem were not associated to a high risk of occurrence of subsequent episodes of bipolar illness, but social inadaptation was. Specific life events worth mentioning separately are pregnancy and childbirth. In older literature, it was suggested a significant risk of postpartum (puerperal) depression or mania. More recent data show a high risk of depression during pregnancy or postpartum period in bipolar women (30 to 50 percent) as well as a certain risk of mood instability even during pregnancy. (21) Comorbidity with personality disorder may be another factor interacting with psychosocial stressors to increase the risk of relapse in bipolar disorder. In a study on 52 US military veterans euthymic with bipolar disorder 38 percent met the DSM-III-R criteria for personality disorder which was associated to lower work commitment and high substance abuse (51).
Poor result is associated to anxiety (52) and concurrent personality disorders (52). There is an overlapping with ADHD, anxiety and personality disorders, for example labile effect, irritability, mood instability, stress, low mood/dysphoria, and many have warned about overdiagnosis and underdiagnosis of bipolar disorder (52).
Medicines and physical illnesses, unlike life events, are not sometimes recognized as significant precipitants for the manic or depressive episodes, but they should be. Medicines and alcohol may “in their own right” precipitate manic episodes and they may also affect the quantity and quality of sleep, thus increasing the probability of mania onset.
Medical conditions and drugs reported as being precipitants of manic episodes (21): Endocrine illnesses a n d t r e a t m e n t s ( C u s h i n g ( i n c r e a s e d s t e r o i d s ) , hyperthyroidism, androgens, steroid treatment and steroid interruption), Drugs (isoniazide, procarbazine, levodopa, methyldopa interruption, hallucinogens (e.g. LSD, PCP a n d m e s c a l i n e ) , c o c a i n e , a l c o h o l , b r o m i d e , sympathomimetic amines, cimetidine, amantadine, tolmetin, iproniazid, methylphenidate, triazolam, zidovudine, buspirone, alprazolam, lorazepam, guanfacine, lisinopril), Metabolic conditions (post- surgery status, hemodialysis), Infections (Influenza, Q fever, cryptococcal meningitis), Pathology of central nervous system, neoplasm (diencephalic suprasellar tumor, cranial traumatism, parasagittal meningioma, AIDS, Creutzfeldt-Jakob disease, corea Syndenham, multiple sclerosis, Chondroma), Vascular lesions (aneurism, infarct), Others( right temporary epilepsy, B12 vitamin deficit, L-glutamine, aspartame, metrizamide as a contrast agent in myelography).
A special attention must be paid to the evaluation of all psychical and psychic comorbidities, substance abuse and the complete therapeutic diagram before instituting treatment, due to the risk of appearance of drug interactions. “Even if they cannot be seen, it does not mean that they do not occur”. They may influence treatment and the subsequent evolution of illness. Polypragmasy with all its faces “the bad, the good and the ugly” must be considered very carefully in bipolar patients that are known as having a risk for comorbidities and substance consumption.
Risk factors for mania chronicization: (52) physical illnesses, substance abuse, positive family history, poor pre-morbid social functioning, cycles of depression or mixed states, rapid cycling.
Wehr and colleagues (53) said that reduction of sleep may be the common denominator of more separate events and stressors that precipitate mania. Indeed, this suggestion is well supported by the clinical experience of many authors. Loss of sleep is common in manic episodes reported after:
1. diverse stressful events such as death (54), 2. postpartum status and 3. “Jet lag” associated to flight to areas having a different time zone.
Another potential important trigger for mood episodes is the season. In the context of illness course, patients with bipolar disorder experience a seasonal pattern many times. In a study, 49 percent of 146 patients that met the DSM-III-R criteria for seasonal affective disorders (55) were diagnosed with bipolar disorder mainly of type I (30 percent of the entire sample; 19 percent were type II) what means that these patients had the tendency to experience complete manic episodes in spring or summer. While winter depression is frequently mentioned in the diagnosis and treatment of seasonal affective disorders, we should also consider the high likelihood of mania in summer (21). A 4-year follow-up study on 75 patients was carried out to investigate the result after recovery from a manic episode. Predictors of unfavorable results included: dissatisfactory occupational status before the index episode, history of previous episodes, alcoholism antecedents, psychotic characteristics and depression symptoms during the index manic episode, male gender and affective inter-episodic symptoms after 6 months of follow-up. The mortality risk in the follow-up period was 4%. Author declared that the identification of specific risk factors depends on the defining of results and the follow-up length (56).
Taking into account the negative consequences of recurrences on illness and psychosocial functioning of patients, early diagnosis owed to the appearance of a first manic episode is an important step towards prescribing a mood stabilizer ever since the beginning of bipolar disorder. Exact knowledge on the clinical features and illness evolution in the patients with a manic episode at the illness onset may help clinicians in developing more specific and relevant therapeutic interventions for these patients (57). Since the improvement of prediction of early evolution of the bipolar disorder is necessary to guide the planning of treatment, authors have assessed recovery, the first relapse and the new onset of illness after the first hospitalization for mania (17).
166 bipolar patients (17) were followed-up for 2-4 year since the first hospitalization for a manic or mixed episode so as to assess the calendar of illness events and result predictors. There aspects of recovery were analysed: syndromal (DSM-IV criteria for disorder are no longer met), symptomatic (Young Mania assessment scale score </ = 5 and Hamilton scale for depression score </ = 8), and functional (reacquiring the social and occupational pre- morbid state). Remission rates (sustained syndromal recovery > / = 8 weeks), the switch (onset of a new episode different as polarity before recovery), relapse (a new manic episode within 8 weeks of syndromal recovery) and recurrence (a new post-remission episode) were also assessed.
After 2 years, most subjects attain syndromal recovery (98%, with 50% attaining recovery in 5.4 weeks); 72% obtained symptomatic recovery. The factors associated to a shorter time until syndromal recovery for 50% of subjects were female gender, short index hospitalizations and lower initial scores for depression. Only 43% obtained functional recovery; these subjects were most of the times elderly and had shorter index hospitalizations. Within 2 years of syndromal recovery, 40% experienced a new manic episode (20%) or depression (20%) and 19% had a switch without recovery. Predictors of manic recurrence were initially psychosis congruent with mood state, lower pre-morbid occupational status and manic initial presentation. Predictors for depression onset were higher professional status, mixed initial presentation and any comorbidity. Anti-depression treatment was marginally linked to the longer time until recovery and early relapses.
Within 2-4 years since the first hospitalization for mania, all, except 2% of patients, had a syndromal recovery, but 28% remained symptomatic, only 43% registered a functional recovery and 57% had a switch or new illness episodes. The risks of new manic or depressive episodes were similar, but they were predicted by contrasting actors.
Many studies showed that the evolution of bipolar disorder tends to worsen over time underlining the importance of early intervention. Despite the recognized need for adjuvant psychosocial treatments during the first manic episode, very few studies assessed psychological interventions for this significant risk period. While pharmacologic treatment helps patients attain recovery from an acute episode, it does not succeed in brining patients to sustained remission. Early psychosocial interventions may be imperative in reducing residual symptoms, preventing the recurrence of mood episodes and improving psychosocial functioning. Despite all these, very few researches on psychosocial interventions for the first manic episode were studied systematically. Studies of the first manic episode indicate a gap between syndromal/symptomatic recovery and functional recovery. New psychosocial interventions for the first manic episode “may build a bridge” over this gap, but they need a controlled study (58).
224 patients were enrolled for another study to determine the clinical and therapeutic relevance of predominant longitudinal polarity for the bipolar disorder. Collecting the data regarding predominant polarity took ten years. Patients were divided according to depressive or manic/hypomanic preponderance. The two groups were compared in terms of clinical and socio-demographic variables.
135 patients (60.3%) were classified as having depressive polarity, whereas 89 (39.7%) were considered as having manic polarity. Manic polarity was more spread among bipolar I patients than bipolar II patients. Depressive polarity was strongly associated to the depressive onset of bipolar disorder. The history of suicide attempts was strongly associated to depressive polarity. As for the therapeutic aspects, the acute use and maintenance of atypical anti-psychotic drugs and conventional neuroleptic drugs was more frequent among those with manic polarity, whereas the use of antidepressants and lamotrigine was extremely spread among those with depressive polarity. Prevention of depression is crucial for the maintenance treatment of bipolar II patients, whereas prevention of mania and depression would be very important for the bipolar I patients. Predominant polarity is a valid prognosis parameter with therapeutic implications (59).
Some clinical differences between men and women in terms of the evolution ad results of bipolar disorder have already been described and others are still controversial. It is important to explore the differences of gender in terms of the clinical and socio-demographic features among the bipolar patients, with a focus on predominant polarity and depressive symptoms. Data were collected from 604 bipolar patients during a 10-year follow-up period. Socio- demographic and clinic variables were collected so as to identify the differences related to gender. Bipolar women stand more chances than bipolar men to have a predominance of depressive polarity as well as a depressive onset whereas men would be more susceptible to have substance abuse as comorbidity. Women have significantly a higher prevalence for psychotic depressions and axis II comorbidities. Bipolar women are also much more susceptible to have a family history of suicide and a personal history of suicide attempts. Suicide attempts are much more often violent among bipolar men. In a logistic regression model, two variables were responsible for most clinical differences related to gender: type of predominant polarity – women being more susceptible of depression, and cocaine consumption – more spread among men (60).
Unfortunately, many suicide attempts take place in this latency period before initiation of lithium which is the only stabilizer of mood known to reduce suicide. A longer delay also results in poor social functioning, more annual hospitalizations and a higher likelihood of suicide attempts, regardless of the polarity of the first episode. The excess of mortality was demonstrated in bipolar disorder in Sweden both for suicide and natural causes, but suicide rates are significantly reduced by long-term medication with an antidepressant, neuroleptic drug or lithium or combinations (52).
The main feature of bipolar women is depression both at onset of illness and as predominant polarity along the course of illness. This thing might have important diagnosis and therapeutic implications (60).

CONCLUSIONS
The natural course and results of manic-depressive illness contribute to its defining and differentiation from schizophrenia. If the clinician understands the natural course, it may help them answer the questions of a patient regarding the most important estimation of all, namely prognosis: will it come and when? Despite the methodological issues of the previous studies and those of recent ones and the limited data from prospective follow- up studies, the literature on evolution and results of manic- depressive illness accepts the following conclusions:
-Bipolar subtype has the earliest age at onset followed by the most recurrent forms of unipolar depression and finally by the less recurrent forms. For bipolar subgroup, representative value varies depending on the definition of onset, from the appearance of first symptoms (affective lability) before the age of 20 until the first hospitalization (in the 25-30 year age interval).
-Almost all bipolar patients experience a relapse. The results of contemporary studies indicate higher relapse rates than those in previous studies; the extended use of antidepressants may have contributed to this increase. Extremely recurrent unipolar depression is a risk factor for the development of bipolar illness.
-Cycle length does not change predictably over time, though it may shorten progressively in initial stages of illness in a subgroup of individuals. A significant percent of bipolar patients develop rapid cycling. Rapid cycling may represent an extreme of bipolar spectrum, rather than a completely autonomous group. The nosological status of rapid cycling remains unclear in spite of all these.
-Literature constantly shows that manic episodes are shorter than depressive or mixed ones. The average duration of episodes remains stable throughout the illness.
-About 1-2% of unipolar patients experience a first manic or hypomanic episode per year suggesting that over a long follow-up period a large minority of patients previously diagnosed as unipolar will shift to diagnosed bipolar disorder. The model of initial depression followed then by mania is associated to poor results in the long term. Antidepressants may play a role in the acceleration of poor results in the long term.
-Psychosocial and physical stressors continue to be strong predictors of the relapse moment, despite the fact that these severe life events most probably interact with a patient with existing vulnerability, in a complex way. Life events seem to be associated to relapse more in early phases than in the subsequent phases of bipolar illness.
-Long-term data suggest that up to a third of bipolar patients attain full remission and a similar number attain complete functional recovery. Although syndromal recovery is two times or even more times frequent, few patients recover the pre-morbid functioning level. Chronic persistence of symptoms may be encountered in about 20% of cases and social incapacity in about 30%. Poor results in recent studies may be influenced by the selection bias for the inclusion of sicker patients in tertiary care centers. In the community, results might be better than the ones mentioned here. The early age at onset, depression, mixed episodes, psychosis, substance abuse, and medicine non-compliance and probably the long-term use of antidepressants are all associated to poor results. Mortality and suicide rates are high in bipolar illness, but they may be substantially reduced by an adequate lithium treatment.
-There are several interesting theories on relapse mechanisms, including chronobiological explanations and kindling theory. Kindling may represent a subgroup effect being relevant for a more sever or atypical subgroup of bipolar disorder individuals.

REFERENCES
1.Gershon ES, Hamovit J, Guroff JJ et al. A family study of schizoaffective, bipolar I, bipolar II, unipolar and normal control probands. Arch Gen Psychiatry 1982;39: 1157–1167.
2.Kraepelin E. Manic-Depressive Insanity and Paranoia. Edinburgh: E. and S. Livingstone, 1921. Originally published as Psychiatrie. Ein Lehrbuch fur Studierende und Arzte. ed. 2. Klinische Psychiatrie. II. Leipzig: Johann Ambrosius Barth, 1899.
3.Zis AP, Goodwin FK. Major affective disorder as a recurrent illness: A critical review. Arch Gen Psychiatry 1979;36: 835–839.
4.Roy-Byrne P, Post R, Uhde T, Davis D. The longitudinal course of recurrent affective illness: Life chart data from research patients at the NIMH. Acta Psychiatr Scand 1985a;71: 1–34.
5.Angst J. (1981b). Course of affective disorders. In:. Van Praag H.M, Lader M.H, Rafaelsen O.J, Sachar E.J. (Eds.). Handbook of Biological Psychiatry. New York: Marcel Dekker,1981b, 225-242.
6.Taschev T. The course and prognosis of depression on the basis of 652 patients deceased. In: Angst J (Ed.). Classification and Prediction of Outcome of Depression. Symposium Schloss Reinhartshausen/Rhein. Symposia Medica Hoeschst 8. Stuttgart: FK Schattauer,1974, 157-172.
7.Keller MB, Shapiro RW, Lavori PW, Wolfe N. Relapse in major depressive disorder: Analysis with the life table. Arch Gen Psychiatry 1982;39: 911–915.
8.Angst J, Sellaro R. Historical perspectives and natural history of bipolar disorder. Biological Psychiatry 2000;48: 445–457.
9.Winokur G, Coryell W, Endicott J, Akiskal H. Further distinctions between manic-depressive illness (bipolar disorder) and primary depressive disorder (unipolar depression). Am J Psychiatry 1993;150: 1176–1181.
10.Turvey CL, Coryell WH, Solomon DA et al. Long-term prognosis of bipolar I disorder. Acta Psychiatr Scand 1999;99: 110–119.
11.Goldberg JF, Harrow M. Kindling in bipolar disorders: A longitudinal follow-up study. Biol Psychiatry 1994;35: 70–72.
12.Kessing LV, Andersen PK, Mortensen PB, Bolwig TG. Recurrence in affective disorder: I. Case register study. Br J Psychiatry 1998a;172: 23–28.
13.Oepen G, Baldessarini RJ, Salvatore P, Slater E. On the periodicity of manic-depressive insanity, by Eliot Slater (1938): Translated excerpts and commentary. J Affect Disord 2004;78(1): 1–9.
14.Bratfos O, Haug JO. The course of manic-depressive psychosis: A follow up investigation of 215 patients. Acta Psychiatr Scand 1968;44: 89–112.
15.Angst J, Baastrup P, Grof P et al. The course of monopolar depression and bipolar psychoses. Psychiatr Neurol Neurochir 1973;76: 489–500.
16.Zis AP, Grof P, Webster M, Goodwin FK. Prediction of relapse in recurrent affective disorder. Psychopharmacol Bull 1980;16: 47–49.
17.Tohen M, Zarate C, Hennen J et al. The McLean-Harvard First- Episode Mania Study: Prediction of recovery and first recurrence. Am J Psychiatry 2003;160(12): 2099–2107.
18.Koukopoulos A, Sni G, Girardi P. Cyclicity and manic-depressive illness. In: Marneros A, Angst J (Eds.). London, England: Kluwer Academic Publishers. Bipolar Disorders: 100 Years after Manic- depressive Insanity, 2000, 315–334.
19.Baldessarini RJ, Tondo L, Hennen J, Floris G. Latency and episodes before treatment: Response to lithium maintenance in bipolar I and II disorders. Bipolar Disorders 1999;2: 91–97.
20.Kessing LV, Mortensen PB, Bolwig TG. Clinical consequences of sensitization in affective disorder: A case register study. J Affect Disord
1998b;47: 41–47.
21.Goodwin FK, Jamison KR. Manic-Depressive Illness. Bipolar Disorder and Recurrent Depression (second edition). New York: Oxford University Press, 2007.
22.Gitlin MJ, Swendsen J, Heller TL, Hammen C. Relapse and impairment in bipolar disorder. Am J Psychiatry 1995;152(11):
1635–1640.
23.Dunner DL. Unipolar and bipolar depression: Recent findings from clinical and biologic studies. In The Psychobiology of Affective Disorders, 1980, 11-24. Pfizer Symposium Depression. Basel, Switzerland: Karger.
24.Okuma T, Shimoyama N. Course of endogenous manic-depressive psychosis, precipitating factors and premorbid personality: A statistical study. Folia Psychiatr Neurol Japonica 1972;26: 19–33.
25.Winokur G, Kadrmas A. A polyepisodic course in bipolar illness: Possible clinical relationships. Compr Psychiatry 1989;30: 121–127.
26.Grof P, Alda M, Ahrens B. Clinical course of affective disorders: Were Emil Kraepelin and Jules Angst wrong? Psychopathology 1995;28: 73–80.
27.Angst J. The course of affective disorders. Psychopathology 1986d;19(Suppl. 2: 47–52.
28.Goldberg JF, Harrow M. Bipolar disorders: Clinical course and outcome. Washington, DC: American Psychiatric Press, 1999.
29.Winokur G. Duration of illness prior to hospitalization (onset) in the affective disorders. Neuropsychobiology 1976;2: 87–93.
30.Molnar GJ, Feeney MG, Fava GA. Duration and symptoms of bipolar prodromes. Am J Psychiatry 1988;145: 1576–1578.
31.Keitner GI, Solomon DA, Ryan CE et al. Prodromal and residual symptoms in bipolar I disorder. Compr Psychiatry 1996;37(5): 362–367.
32.Jackson A, Cavanagh J, and Scott J. A systematic review of manic and depressive prodromes. J Affect Disord 2003;74(3): 209–217.
33.Forty L , Jones L, Jones I, Smith DJ, Caesar S, Fraser C, Gordon- Smith K, Hyde S, Craddock N. Polarity at illness onset in bipolar I disorder and clinical course of illness. Bipolar Disord 2009;11(1): 82-8.
34.Kessing, LV. The effect of the first manic episode in affectivedisorder: A case register study of hospitalised episodes. J Affect Disord 1999;53: 233–239.
35.Coryell W, Endicott J, Maser JD et al. Long-term stability of polarity distinctions in the affective disorders. Am J Psychiatry 1995;152: 385–390.
36.Goldberg JF, Harrow M, Whiteside JE. Risk for bipolar illness in patients initially hospitalized for unipolar depression. Am J Psychiatry
2001;158: 1265–1270.
37.Angst J. The course of affective disorders: II. Typology of bipolar manic-depressive illness. Arch Psychiatr Nervenkr 1978;226: 65–73.
38.Perris C, d’Elia G. A study of bipolar (manicdepressive) and unipolar recurrent depressive psychoses: X. Mortality, suicide, and life cycles. Acta Psychiatr Scand 1966;42(Suppl. 194): 172–183.
39.Perugi G, Micheli C, Akiskal HS et al. Polarity and course of manicdepressive illness: A retrospective analysis of 320 bipolar I patients. Compr Psychiatry 2000;41: 13–18.
40.Hendrick V, Altshuler LL, Gitlin MJ, Delrahim S, Hammen C. Gender and bipolar illness. J Clin Psychiatry 2000;61(5): 393–396.
41.Baldassano et al. Gender differences in bipolar disorder. Systematic Treatment Enhancement Program First 500 (STEPBD), 155th annual meeting of the American Psychiatric Association, May 18–23, 2002,Philadelphia, PA.
42.Swann A, Janicak P, Calabrese J et al. Structure of mania: Depressive, irritable and psychotic clusters with different retrospectively assessed course patterns of illness in randomized clinical trial participants. J Affect Disord 2001;67: 123–132.
43.Daban C, Colom F, Sanchez-Moreno J et al. Clinical correlates of first-episode polarity in bipolar disorder. Compr Psychiatry 2006;47(6): 433-7.
44.Maj M, Pirozzi R, Starace F. Previous pattern of course of the illness as a predictor of response to lithium prophylaxis in bipolar patients. J Affect Disord 1989;17: 237–241.
45.Ellicott AG. A prospective study of stressful life events and bipolar illness. Unpublished doctoral dissertation.
University of California, Los Angeles, 1988.
46.Kennedy S, Thompson R, Stancer HC, Roy A, Persad E. Life events precipitating mania. Br J Psychiatry 1983;142: 398–403.
47.Depue RA, Kleiman RM, Davis P, Hutchinson M, Kraussm SO. The behavioral high risk paradigm and bipolar affective disorder: A conceptual framework with five validation studies. J Abnorm Psychol Monograph 1981;90: 381–437.
48.Hunt N, Bruce-Jones W, Silverstone T. Life events and relapse in bipolar affective disorder. J Affect Disord 1992;25(1): 13–20.
49.Johnson SL, Miller I. Negative life events and time to recovery from episodes of bipolar disorder. J Abnorm Psychol 1997;106: 449–457.
50.Staner L, Tracy A, Dramaix M et al. Clinical and psychosocial predictors of recurrence in recovered bipolar and unipolar depressives: A one-year controlled prospective study. Psychiatry Res 1997;69: 39–51.
51.Kay JH, Altshuler LL, Ventura J, Mintz J. Impact of axis II comorbidity on the course of bipolar illness in men: A retrospective chart review. Bipolar Disord 2002;4(4): 237–242.
52.Bazire S. Psychotropic Drug Directory, 2009, 38-45.
53.Wehr TA, Sack DA, Rosenthal NE. Sleep reduction as a final common pathway in the genesis of mania. Am J Psychiatry 1987a;144: 201–204.
54.Krishnan KRR, Swartz MS, Larson MJ, Santoliquido G. Funeral mania in recurrent bipolar affective disorders: Reports on three cases. J Clin Psychiatry 1984;45: 310–311.
55.Faedda GL, Tondo L, Baldessarini RJ, Suppes T, Tohen M. Outcome after rapid vs. gradual discontinuation of lithium treatment in bipolar disorders. Arch Gen Psychiatry 1993;50(6): 448–455.
56.Tohen M, Waternaux CM, Tsuang MT. Outcome in mania. A 4-year prospective follow-up of 75 patients utilizing survival analysis. Arch Gen Psychiatry 1990;47: 1106–1111.
57.Kaladjian A, Fakra E, Adida M et al. Manic polarity of the onset episode of bipolar disorder: clinical and prognostic considerations. Encephale 2010;36 Suppl 1: S13-7.
58.McMurrich S, Sylvia LG, Dupuy JM et al. Course, outcomes, and psychosocial interventions for first-episode mania. Bipolar Disord 2012;14: 797–808.
59.Colom F, Vieta E, Daban C, Pacchiarotti I, Sánchez-Moreno J. Clinical and therapeutic implications of predominant polarity in bipolar disorder. J Affect Disord 2006;93(1-3): 13-7.
60.Nivoli AM, Pacchiarotti I, Rosa AR et al. Gender differences in a cohort study of 604 bipolar patients. J Affect Disord 2011;133(3): 443-9.

***