2.1 How common is bipolar disorder?

The simplest answer to this is that it occurs in 1% of people. This seems to be true around the world in any population investigated. However, there is a spectrum from severe mania through milder forms of manic depression to recurrent depression and on to normal mood variation. It is always difficult when you have a spectrum to know exactly where to draw the lines within this to define the different disorders. Tight definitions of manic depression would give a figure of less than 1%; broader definitions which include hypomania would give figures from 2 up to 10% (Fig. 2.1).

Fig. 2.1 The prevalence of affective disorders. BP, bipolar.

(Reprinted from Clin Appr Bipolar Disord 2002;1:10-14 by permission of Cambridge Medical Publications.)

2.2 What causes manic depression?

‘Nature or nurture, genes or environment?’ is a common question in the cause of any illness that is rarely answered definitively. For bipolar illness (particularly the more severe end of the spectrum–mania rather than hypomania) the genes/nature answer seems more likely. Bipolar illness tends to run in families and the more closely an individual is related by blood to a manic depressive the more likely that person is to develop the illness. If one parent suffers from manic depression and a child is adopted away and brought up by another family, there is still the same risk of developing manic depression as if the child was brought up by the biological parents. In the case of twins, if one twin suffers from manic depression then the chance of the other twin becoming ill is higher in cases of identical rather than fraternal twins (Smoller & Finn 2003). All of these observations indicate that bipolar illness is strongly influenced by genetic factors. However, it is not the whole story as even identical twins are not 100% concordant: if one identical twin suffers from manic depression the other will have a bipolar illness in about 50% of cases and recurrent depression in about 30%. However, there are still 20% of identical twins who are discordant, i.e. one is illness-free.

2.3 What family history am I likely to elicit from a patient with manic depression?

About 70% of bipolar patients will have a close relative (a parent or sibling) who suffers either from manic depression or from serious (usually recurrent) depression. Only 20% will actually have a bipolar relative (Case vignette 2.1). The equivalent figures for the general population are about 10% (depression) and 1% (bipolar). First-degree relatives (sibling, parent or child) of someone who is manic depressive have about a 10-15% chance of having or developing manic depression (Gershon et al 1982, Gottesman 1991, Smoller & Finn 2003) (Fig. 2.2).

Fig. 2.2 The possible relationship between the closeness of the blood relationship to someone with manic depression and the likelihood of developing depression and bipolar disorder. DZ, dizygotic; MZ, monozygotic.

Relatives of bipolar patients do however have a higher chance of being successful academically and occupationally. It is only a small effect but may indicate that having a small dose of what leads to manic depression can actually be an advantage.

2.4 How many genes are involved?

It is possible that some families with several members with manic depression have single genes having a major effect, although in most cases it is likely to be several genes that increase the liability to manic depression. These genes are likely to overlap with the genes that predispose to severe and recurrent depression and also with those predisposing to psychosis including schizophrenia. It is probable that we all have some of these genes and that those who suffer the illness have more of the susceptibility genes and in particular combinations. However, despite a major research investment in this area we are not yet at the point of predicting who is particularly susceptible to manic depression from examination of their genes. The other hope from genetic research is that we might be able to identify which patients are likely to respond to which medicines from an analysis of their genes.

The questions that are likely to emerge include: ‘What are the normal functions of these genes?’ and ‘To what extent do we need to be liable to depression and elation?’

2.5 Why are there genes that cause extreme emotions prevalent in the ‘normal’ population?

One theory is that the ability to develop a depressed mood has been selected for in our recent (last few million years) evolutionary development and is vital to our social competence. The facts that mood disorders tend to be more common in women and the most vulnerable time is in the postnatal period suggest that the ability for a mother to become depressed has some reproductive advantage. How could this be when it would seem a major disadvantage to get depressed when there is a baby to look after? Perhaps depression is best seen as the mental equivalent of pain. How would we expect a mother to react if her baby died? We would of course expect her to become depressed and would be very surprised if she did not. In fact we would not really regard this depression as a psychiatric problem because it would be so understandable. Our evolutionary tactic has been to have a small number of offspring but to take very good care of them, rather than have many and rely on a few surviving. A mechanism that meant mothers would know that their baby dying would lead to extreme mental distress would seem to be a powerful additional factor in ensuring that mothers take very good care of their offspring. Of course, like other pain mechanisms, it can go wrong and activate when not beneficial and this may be what constitutes the depression end of manic depression.

The advantages of hypomania are probably easier to recognise, when increased energy, attention and optimism can still bring results even in (or perhaps particularly in) this modern era. But the extreme of mania could be this mechanism going wrong.

If the genes that are responsible for manic depression are identified, then the next question to be answered is: ‘What kind of world would we be living in if we identified and eliminated the genes responsible for manic depression?’

2.6 Which genes cause the illness?

There has been a huge investment in trying to discover the particular genes that relate to manic depression. Many genes have been looked at and several have initially shown promise, but at this point no clear linkage has been established which can be replicated consistently. It seems likely that the problem will prove much more complicated than we had hoped and differences in how genes are expressed may be as important as differences between variations (alleles) of a particular gene.

There is another problem when looking at the genetics of manic depression, i.e. what constitutes manic depression?

We would all include those with a bipolar I disorder, but would we also include bipolar II patients?

How about those who suffer recurrent depression but have family members with bipolar illness?

Would we include all those with recurrent severe depression?

The question of where we draw the lines around the phenotype is still one of the basic questions that need to be answered (see Ch. 1).

2.7 Who is likely to develop manic depression?

It is not possible to predict who will develop manic depression and it is a continual surprise when those people who have been treated for depression for several years suddenly become manic (see Case vignette 4.1). However, those whose illness started at a young age (particularly in the teenage years or the twenties) and those who have frequent episodes and rapid shifts between depression and euthymia are particularly at risk (Fig. 2.3).

Fig. 2.3 Recurrent depression can precede bipolar disorder.

2.8 Does adversity bring on manic depression or cause relapses?

There is a general assumption when looking at psychiatric disorders that they have been set off by something that has happened in the patient’s life. The same assumptions are sometimes made about physical illnesses such as heart attacks and strokes. Working out to what extent this is true is difficult because bad things happen quite often (e.g. we all suffer bereavements) and if an illness occurs it may coincide with an adverse event.

The situation is made more complicated by the fact that we can make bad things happen to us–for example if a husband is abusive to his wife then she may leave. Depression and mania can lead to changes in behaviour which trigger off adverse events. Getting a court summons for speeding may be related to the early symptoms of mania.

There have been several studies (e.g. Hunt et al 1992) that have tried to tease out these various factors which have led to the conclusion that adverse events do play a part. The first episode of manic depression is likely to be at a time of other difficulties in the person’s life, and recurrences are also likely to coincide with adverse life events (Fig. 2.4). However, the contribution from life events is small and it is best to look at them as triggers rather than causes.

Fig. 2.4 Life events prior to relapses in manic depression.

2.9 Does upbringing affect the chance of developing manic depression?

Those who have had a difficult and abusive childhood are more likely to suffer from depression and personality problems. An unhappy childhood does not make anyone more liable to develop manic depression and the childhoods of manic depressives are usually normal. Those who suffer from schizophrenia are more likely to have had adverse physical problems in childhood, difficult births with obstetric complications and illnesses such as meningitis, encephalitis and head injury. However, this does not seem to be true for manic depressives–or at least not to the same extent. It is unlikely that adverse childhood experiences offer an explanation for the development of manic depression.

2.10 Are there physical illnesses that can lead to manic depression?

Several organic diseases have been linked with manic depression, particularly in those whose illness begins in older age (over 65). Cerebrovascular disease is the most common; non-dominant hemisphere cerebrovascular accidents in particular appear to predispose to the development of mania but this is still a rare outcome and often there is either a previous history of depression or a family history of affective disorder.

Other brain disorders (or systemic diseases with cerebral involvement) can present with mania or severe (often psychotic) depression. This includes infection with HIV, usually when it reaches the immunodeficiency stage, autoimmune disorders such as systemic lupus erythematosus (SLE) or primary neurological diseases such as multiple sclerosis. Accompanying the affective syndrome there is usually evidence of disorientation and other features of confusion along with visual hallucinations, both of which are uncommon in primary bipolar illness.

2.11 Does the use of illicit drugs cause manic depression?

Stimulant illicit drugs such as amfetamines, cocaine and ecstasy can certainly mimic mania. People who are acutely intoxicated with stimulants usually appear overcheerful, excited, overactive and talkative. Sometimes they will also have paranoid ideas and they usually have difficulty sleeping (or don’t want to sleep). The distinguishing features are the history of ingestion, very acute onset and rapid recovery, usually within 24-48 hours.

Stimulant drugs can also precipitate mania among manic depressives. Illicit drugs seriously worsen the course of the illness and its treatment, and invariably the frequency of recurrence, complications and compliance are affected. Whether taking stimulants can actually set off a lifelong propensity to mania and depression is much less certain. If this were beyond doubt, the incidence of manic depression would have increased considerably in recent years and that probably is not the case.

Stimulant drugs can also lead to more persistent depression (up to a few weeks) in the period after prolonged drug use.

Cannabis is associated with the development of psychosis, particularly among heavy users, and is an important causal factor in schizophrenia. Whether it can also induce manic depression is less clear, but it is likely to lead to psychosis, particularly during periods of mania.

2.12 Do any medicines cause mania or depression?

The drugs most often associated with mania are the corticosteroids which ties in with the link with Cushing’s disease (see Q 2.10). Minor elevations of mood with euphoria are felt by many people taking steroids but it is rare that this extends to the full manic syndrome. However, depression is a more common and debilitating outcome of this treatment.

Those with Parkinson’s disease who are having treatment with L-dopa can also experience elevation of mood and other manic symptoms. The use of stimulants including methylphenidate can cause mania (see Q 2.11) and this can cause significant problems when considering the diagnosis in adolescents who already have deficits in attention. However, the most common drugs to induce mania are antidepressants (see Q 3.15).

There is a long list of drugs that are associated with depression apart from steroids (Box 2.1). The induction of depression by the antihypertensive reserpine–which depletes monoamines including noradrenaline (norepinephrine) and serotonin–was the original basis for the theory that monoamines were the important neurotransmitters in depression.

BOX 2.1 Drugs commonly associated with depression

Steroids, including prednisolone and dexamethasone

Beta-blockers, particularly those that cross the blood-brain barrier, including propranolol

Reserpine

Oral contraceptives

Mefloquine

Depression is commonly precipitated in those who are predisposed (especially those with previous episodes).

2.13 Is bipolar illness more common among some ethnic groups?

Manic depression is an illness that is found throughout the world in all countries and cultures. The incidence of the severe end of bipolar illness is similar in different countries, but it is not clear if this is true at the bipolar II end of the spectrum. Surveys looking at rates of hypomania have been undertaken only in developed countries, and the reliability of this diagnosis in surveys is not good.

Several studies show that minority ethnic groups are more likely to develop manic depression. This may be because migrants are particularly vulnerable to this illness, with an increased rate among those who have moved between countries. The explanation for this could be that those who do move have a particular type of personality (energetic, highly motivated, ‘hyperthymic’) or that the stress of migrating and adapting to a new culture is causal.

2.14 Does a particular personality predispose to bipolar illness?

Personality is a notoriously difficult concept to tie down and the relationship between personality type and manic depression is even more complicated. Traditionally it is the cyclothymic temperament that is associated with bipolar illness. Cyclothymia shows itself as mood swings, both up and down, usually with other symptoms of mania and depression but at a low and subclinical level. We are probably all familiar with someone who shows this, though often they are not very aware of it themselves. Some weeks they are cheerful and sociable and others rather grumpy and miserable–this is true of many of us but particularly true of cyclothymics!

Some people who develop clear manic depression do report cyclothymia as the type of personality that they have had since their teenage years. However, this is uncommon and most people with manic depression would have been considered to have a personality within the realm considered normal prior to their illness developing.

Between episodes of illness, manic depressives revert to their usual personality. This is in quite marked contrast to those who suffer from schizophrenia who often have change in their personality with a blunting of emotional responses.

2.15 Is manic depression becoming more common?

The answer is probably not but we are getting better at recognising the range of manic depressive illnesses (although some would say we have got worse at drawing limits around a serious illness). The closer that epidemiological studies come to looking at the community, rather than judging the prevalence of manic depression from hospital samples, the larger the number of cases that are found. More patients who had previously been treated as suffering from recurrent depression are now recognised as bipolar because, when examined over a long period of time, it is clear they have periods of elevated mood between their episodes of depression.

Some believe that manic depression really is becoming more common, not just better recognised, particularly among those that misuse drugs such as cocaine and anabolic steroids. It may be that this increase is not really a rise in the prevalence of manic depression but an increase in drug-induced manias, as the long-term outcome for these people is not clear.

2.16 At what age does manic depression start?

The first episode can occur at any age after puberty. (Whether episodes of mania occur before puberty is controversial; if it does happen, it is very rare.) The average age is about 30 years but this covers a very broad range (see Fig. 2.5). Most patients will have had their first major episode of affective illness in their twenties.

Fig. 2.5 Age of first episode of bipolar disorder.

Sometimes patients will recall earlier but more minor episodes going back to their teens. Later onsets are, however, not uncommon, particularly in people who have already experienced depressions. In the elderly mania is much more likely to be preceded in earlier life by periods of depression, so it may not be apparent that this is a bipolar illness until old age. A first episode of mania can certainly occur in the elderly but if there have been no previous periods of depression then physical illness should be excluded as a trigger factor (see Q 2.10).

What is striking is that there is usually a gap of 5-10 years between the onset of symptoms and the diagnosis being made (see Q 1.16). This is partly because it is impossible to make a bipolar diagnosis when only depression has occurred so far and also because the symptoms of both mania and depression may initially be mild and difficult to recognise. It is also hard to disentangle bipolar symptoms from personality and situational factors in the early years. It can be difficult to make this diagnosis, but it should be borne in mind in everyone who presents with depression. However, this needs to be balanced with falling into the other trap of making the diagnosis when it is not appropriate.

The earlier the diagnosis can be made, the earlier effective treatment can be provided, thus reducing the impact on the patient’s life and improving the long-term outcome.

2.17 Does an early onset mean a more severe condition?

Many people who suffer from manic depression actually date the onset to their teenage years. However they report that the illness is mild during that time and they manage to continue their lives without major disruption. It is only later in life that the episodes become more severe and disabling and they require treatment.

Severe and psychotic episodes can begin at any age including adolescence. Likewise illnesses that do not respond well to treatment and become either chronic or with frequent relapses can occur at any age. Overall the range of severity of the illness and the frequency of relapse are not higher in those with an earlier onset.

The treatment of adolescents with bipolar illness can be very challenging, not intrinsically because of the nature of the illness but because of the nature of the teenager. Teenagers are more likely to have complications such as drug misuse and poor compliance with treatment (see Q 6.20)

2.18 How common is recurrence after a single episode of mania?

Recurrence of bipolar disorder is almost inevitable. Single episodes of mania with no further illnesses occur in only 5% of patients. However the gap between illnesses can be many years (Fig. 2.6). Recurrence of mania and occurrence of depression are equally likely after recovery from mania.

Fig. 2.6 The course of manic depression can include lengthy well intervals. This graph demonstrates a long well interval after four episodes.

In the year after a first manic episode about 50% will have an episode of depression or mania. If there have been previous depressions then the risk is even higher. There are really no other factors at first episode that can help predict whether any patient will be particularly likely to experience an early recurrence.

2.19 How long do episodes of mania and depression last?

Looking back at the data collected a hundred years ago it is apparent that most patients did recover in the long run but mania and depression regularly lasted for many months. Modern treatments have meant that most patients will be recovering within a few weeks, with episodes lasting longer being considered ‘treatment resistant’.

2.20 Do patients make a full recovery from each episode?

If this is the first episode of mania or an early episode of depression you can be fairly confident that the patient will recover in that the symptoms will go away entirely. In the early years of suffering from manic depression symptomatic recovery is the norm; however getting back to full function is much harder and there is a fallout in terms of occupational capacity from early on, partly because it is difficult for patients to get back to highly responsible jobs if they have acted very erratically either in mania or in depression (e.g. losing their license to drive a heavy goods vehicle). It is also hard for patients to return to high stress jobs with long hours if they are trying to take good care of their mental and physical health.

2.21 Is it mania or depression which predominates in the long run?

Almost all bipolar patients experience both depression and mania. Women tend to have an illness course that is predominantly depressive, men tend to have more manic spells and fewer depressions. Unipolar mania is sometimes defined as having three episodes of mania and none of depression, and this is more commonly found among men.

For any individual patient the ratio of manias to depressions tends to be fairly constant throughout the illness course. If your patient has shown a particular pattern of illness then they will usually continue along the same path. Overall, treatments for mania are probably more effective than those for depression so that in the long term the problem you will be faced with as a doctor treating bipolar patients is the management of chronic depression.

2.22 How often does mania switch into depression?

Up to a third of manias end in depression and a similar proportion of manias are preceded by depression. These are not always severe depressions and some level of low mood can be expected after mania–often an understandable reaction when patients recognise how they have been behaving and the consequences of this. It can be difficult to judge when this low mood tips into depression.

Some patients have a regular pattern of switching in their illness so that, for example, they can expect to experience depression after mania (Fig. 2.7). Men are particularly likely to have this pattern of illness.

Fig. 2.7 The commonest time for mania and depression to occur is immediately following an episode of the opposite pole.

Because of this pattern of switching it is always advisable to be looking at long-term treatment in whichever pole you are treating. The best way of preventing one pole of the illness is to prevent the other!

2.23 How quickly do episodes of mania and depression come on?

‘I can go from zero to supernova in 48 hours’ is the classic comment about the onset of mania but sometimes the onset is over several weeks and it is only as behaviour becomes more extreme that boorish irritable behaviour can be clearly seen to be part of a manic swing. Depression is more commonly insidious in its onset over a few weeks rather than a few days and tends to move from mild to more severe states gradually.

2.24 How many people become chronically depressed?

2.25 How can the future course of the illness be predicted?

Past performance is the best predictor of future prospects. It can be very useful to draw up a life chart with a patient (see Fig. 6.1) which shows when their illness started, what episodes they have had and what treatment has been taken. Looking at the frequency, severity and type of previous episodes in relation to what medications have been used can indicate a pattern and give some indication of the future. However, if you look very hard it is easy to see patterns that are not really there.

When trying to help someone to predict the future course, look particularly at the well interval for some guidance as to when the next episode is likely to occur. For instance, if someone suffered a period of mania as a student in their twenties and then a period of depression in their forties (Case vignette 2.2) it is reasonable to be confident that there will be a further lengthy well interval (see Q 2.18). However, strong predictions in manic depression should be avoided as the illness often surprises, both with patients with frequent relapses reaching lengthy stability and others who have been free of the illness for many years experiencing multiple episodes.

The other major factor in prediction is insight–how well has the illness been recognised and what action has been taken to avoid relapse or extra treatment undertaken at an early stage?

2.26 Does bipolar II illness develop into bipolar I?

Generally the pattern of bipolar illness for any particular person tends to be fairly constant, both in its severity in each episode and the frequency of episodes. So you would expect that if only hypomanic episodes have occurred that this will be the pattern for the future. However, just as there is a small number of people each year that ‘convert’ from recurrent unipolar depression to bipolar illness there is also a small number that convert from bipolar II (recurrent depression and hypomania, Fig. 2.8) to bipolar I (recurrent depression and mania). This is particularly likely to occur if they are taking high doses of antidepressants (especially if they are not taking lithium) or if the depressions themselves are becoming more severe and so require higher levels of treatment. Stopping treatment such as lithium suddenly can precipitate mania for the first time and that is why it is important to change from treatments only slowly (see Q 5.27).

Fig. 2.8 The course of bipolar II illness with hypomania and depression.

2.27 How common is chronic illness?

About a quarter of those with manic depression will be experiencing symptoms which are so persistent that they are rarely free of them. Most commonly these are chronic depressive symptoms which are at a relatively mild but still distressing and disabling level. There are other patients who become chronically manic although this is rarer. A similarly small proportion continues a relentless course of switching between mania and depression over days/weeks or months in the longer term (see Q 1.12). Although there are not many of these patients about, if you look at any long-term facility for chronically ill psychiatric patients there will be a number of people with manic depression among the larger group with chronic schizophrenia. These patients are commonly women who have started their illness at a rather later age and who also have prominent cognitive impairment.

2.28 Is bipolar disorder a progressive condition?

There is a small increase in the frequency of relapse with time, but this effect is small. It would be expected that each of the first few episodes has a shorter well interval but then usually a more stable pattern emerges (Fig. 2.9). However the course of manic depression is endlessly variable!

Fig. 2.9 The relationship of the number of episodes to the length of illness.

Those people who have the most severe illnesses will usually suffer frequent relapses and poor recovery right from the start. It is usually apparent from early on who has a better prognosis as they will experience shorter, less severe episodes with a long well interval. However the social toll of the illness can grow with time even if the frequency and severity of the episodes do not. Losing close relationships (including divorce, which is particularly likely for manic depressives) and a lack of consistency at work often lead to cumulative personal damage.

2.29 Is manic depression a permanent condition?

The vulnerability to episodes of mania and depression probably lasts a lifetime and having experienced one or more manias and depressions then there is a lifelong vulnerability to recurrence.

There is a small group of between 10 and 20% of people who have suffered several episodes of illness who then have a prolonged period of many years free of bipolar disorder (see Fig. 2.6). However it is not uncommon to see people who have been well for many years (see Case vignette 2.2) who experience a recurrence 20 years after their last episode. It would appear that the vulnerability to manic depression never goes completely away even when people have been well for many years.

2.30 What is the long term outcome for bipolar disorders?

Professor Angst of Zurich has contributed an enormous amount of information about bipolar disorders and their long term course and outcome. His studies have indicated that at 20 years after the start of their illness, more than two-thirds of manic depressives will still be experiencing disabling problems from their illness (Table 2.1) (Angst & Sellaro 2000).

TABLE 2.1 Long term outcome in bipolar disorder

Adapted from Angst & Sellaro (2000).

There is no doubt that manic depression is one of the most serious and disabling illnesses to which mankind is subject. The World Health Organization ranks it in the top ten conditions that lead to death and disability. This would seem unlikely at first sight because it is a relatively uncommon illness. However, as it afflicts people from an early age, it has a major impact on working lives and so a considerable economic impact.

2.31 Does bipolar illness develop into dementia/schizophrenia?

There is no evidence that manic depression leads on to dementia. It is true that sometimes the early stages of dementia present with prominent mood symptoms–usually depression but sometimes also of elation. Agitation and restlessness are themselves common symptoms in dementia and can be the presenting complaints early in the illness. There is some evidence of cognitive impairment in manic depression. If this does progress it is a very slow progression and certainly not the rapid progression usually seen in dementia (see Q 6.3).

There is a small number of patients who are clearly diagnosed as having manic depressive illness who then develop symptoms that are typical of schizophrenia. After a few years it then becomes clear that they are suffering from chronic schizophrenia rather than manic depression. Some people say the diagnosis was wrong to start with and that the manic illness was an early stage of schizophrenia, and if you had searched more carefully you would have found features of schizophrenia. It is very common for those with schizophrenia to suffer prominent symptoms of depression and up to a third experience excited and elated periods but the symptoms typical of schizophrenia continue throughout.

The border between manic depression and schizophrenia is a blurred one, and you will find some psychiatrists who seem to diagnose almost everybody with a psychotic illness as having a schizoaffective disorder as they can find elements of schizophrenia and also of affective illness in every severe mental illness. This doesn’t seem to be a very helpful way of proceeding as most trials of treatment are based on one diagnosis or the other and treatment trials in schizoaffective disorder are few and far between. The point of diagnosis is to guide treatment and estimate prognosis and that is why it is worth making the split between the diagnoses. Having said that, there are patients who clearly do stand on the border and merit the diagnosis of schizoaffective disorder (see Q 1.15).

2.32 What is likely to trigger off a recurrence?

When you look carefully at stressful life events there is an excess before both depression and mania (see Fig. 2.4). However, adverse life events are probably only responsible for about 20% of episodes of mania or depression. There is some evidence that early episodes of an illness are more likely to be triggered by adverse events. It is difficult to tease out the effects of life events from the natural course of the illness. Unpleasant and distressing events happen to all of us (on average three serious events a year) and if you become ill (in any way but particularly psychiatric illness) you will tend to try to find a link to something that has happened recently, in a search for meaning.

It is usually adversity that triggers affective illness although rarely you come across a patient whose illness started after a great success or a very positive event in their life.

2.33 Does the menstrual cycle affect bipolar illness?

Most women who are depressed will notice that their depressive symptoms are worst in the premenstrual phase; this is true in both unipolar and bipolar illness. Likewise some women report improvements in their affective illness when they take the oral contraceptive, although others find that it makes their symptoms worse. Without any good clinical trial data available it is difficult to give any general advice but each case needs to be judged individually. My approach is to focus first on the usual treatments for manic depression and only to consider hormonal approaches if the standard treatments are not effective.

In mania there are no known effects of the menstrual cycle.

It is uncommon for bipolar illness to be diagnosed before menarche in girls or before puberty in boys, suggesting that there is some alteration, which may relate to the hormonal changes occurring at this time that increases the vulnerability to affective disorders.

The potency of childbirth to precipitate both mania and depression is well known (see Q 6.17) and this is a common time for the illness to appear. It has been reported that manic depression can arise for the first time at the menopause, but this is rare. There is no clear connection between the illness and the menopause which would allow you to predict for a woman what is likely to occur to her manic depression at this time. Some find that their illness actually improves and they achieve more stability; others report it to be a stormy time for their illness. It can be tempting to try using hormone replacement therapy to alleviate affective symptoms but it is better to focus on the usual treatments at least initially and only experiment with hormonal treatment if this is not successful.

2.34 Does manic depression occur more often at a particular time of year?

There is a condition called seasonal affective disorder where people experience depression starting in the autumn or winter with a recovery by the spring, often accompanied by a hypomanic episode.

However, among those with bipolar I disorder (i.e. depression with mania rather than hypomania) a predictable seasonal relapse is uncommon. In fact even those with seasonal affective disorder often develop depressions or hypomanias at the wrong time of year.

Bright artificial light is an effective treatment for recurrent winter depression (see Q 3.34). This approach does require a considerable time commitment: at least an hour in the morning sitting very close to and looking repeatedly at a light. Some people are able to stick with this but most find it difficult to sustain. There are now some ‘light hats’–baseball-type caps with lights in the visor–which are easier to use, but look slightly ridiculous and patients should probably be discouraged from going out in them unless they are trying to gain a reputation for eccentricity. Alternatives to the light treatment are walking outside or antidepressant drugs.

2.35 How important is a regular sleep pattern?

It has been reported that air travel can precipitate mania and most doctors who cover airports will have seen more than their share of manic patients arriving. It has been suggested that those travelling from west to east are more likely to become manic but it is difficult to understand why this should be so. The sleep loss and jet lag are probably both relevant factors in this as the body clocks of bipolars are sensitive to these changes. This needs to be considered by the jet-setting bipolar patient and judicious use of hypnotics to help the adjustment is a good idea.

Most of those who are depressed will be having difficulty sleeping at night but will often resort to sleeping in the day. This can lead to the sleep pattern becoming very erratic. It is worth encouraging the depressed to try to get as much of their sleep as possible at night and not sleep in the day (see Q. 3.30).

A proportion of bipolars who are depressed are sleeping excessively and this pattern also exacerbates depression. Again it is very difficult to change this but encouraging them to get up at a reasonable time and not going to bed too early is worthwhile. A more extreme approach of sleep deprivation can be an effective antidepressant treatment both in unipolar and bipolar depression. This involves keeping patients awake all night, usually for 3 or 4 nights (and days). It can lead to a beneficial change in mood but is usually only a short-term solution. Sometimes it is helpful for those with an intractable depression when a shift out of their depression even for a short period can give you and them hope that a change can be made.