The differentiation of bipolar disorder from unipolar depression is comparatively recent, with studies from the early 1960s outlining a bipolar-unipolar distinction by showing differences in genetic risk, age at onset, gender and outcome.¹ Bipolar I disorder has a prevalence of almost 1% of the population, presents slightly more commonly in women and typically has an onset in the late teens or early 20s.^1-3 In contrast to schizophrenia, bipolar disorder is equally represented among all socioeconomic groups. As well as having a 20-30 times higher risk of suicide (6-9% lifetime risk), than the general population, cohort studies have shown that patients with bipolar disorder may have a higher risk of mortality due to cancer and cardiovascular causes.¹    

Difficulties around diagnosis

Bipolar I disorder

Diagnosis of bipolar I disorder can be difficult as a diagnosis can only be made when the person is in a mixed or manic episode. Over 50% of first episodes are with depression, which can often lead to a dilemma in psychiatric or general practice as potent antidepressants can precipitate mixed or manic episodes or rapid cycling, particularly in women.¹ Generally, a diagnosis of bipolar disorder needs to be considered when patients who have a strong family history of bipolar disorder first present with severe or psychotic depression in their late teens or early 20s. In assessing a patient with a more established diagnosis of a mood disorder it is important to consider diagnosis carefully as this can have major implications for treatment and prognosis.

Bipolar II disorder

Studies have shown that if a careful longitudinal history is taken about 15% of patients with unipolar depression will actually have previous hypomanic or manic episodes and will have been misdiagnosed.¹ Many of these patients will have bipolar II disorder which presents clinically as much more like unipolar depression than bipolar disorder. This disorder, which has a population prevalence of over 2% and like unipolar depression is much more common in women that men, is characterised by long, often very refractory depressive episodes interspersed by short mixed or hypomanic episodes with irritability, racing thoughts and insomnia.⁴ Patients with bipolar II disorder typically spend more time with depression at syndromal or subsyndromal levels and often show poor response to antidepressants and mood stabilisers.

Bipolar disorder and schizophrenia

Bipolar disorder is also seen as being on a continuum with other psychotic disorders, in particular, schizophrenia. Classification systems, allowing mood incongruent psychotic symptoms during manic episodes, are broad for bipolar disorder, meaning some patients may be diagnosed with bipolar disorder now that may have been historically considered to have schizoaffective disorder or even schizophrenia.¹

Diagnosing schizoaffective disorder has been considered to be confusing and subtle to both psychiatrists and GPs alike. However, the clearest and quite restrictive definition of schizoaffective disorder comes from the US Diagnostic and Statistical Manual-IV (DSM-IV) which requires the individual with schizoaffective disorder to have suffered from periods of mood disorder with psychosis, that mood symptoms have been present for a substantial proportion of the illness and that periods of psychosis in the absence of mood disorder are present for at least two weeks during the illness.⁵ In other words, if a patient only suffers from psychosis when they are manic or depressed they fulfil criteria for bipolar disorder. Similarly, recent treatment approaches tend to lay emphasis on psychosis or mood episode treatment rather than treatment of disorder.    

Many of the findings from large scale epidemiological studies on schizophrenia have been replicated to a lesser degree with bipolar disorder. For example, the large multi-centred Aetiology of Schizophrenia and Other Psychosis Study (AESOP) showed that rates of bipolar disorder as well as schizophrenia among ethnic minorities in the UK are six to 10 times higher than among the Caucasian population, with even higher rates among those who are ethnically isolated in the community and that pathways to care are typically through the police and judicial system rather than primary care.⁶ However, in contrast to schizophrenia, the data linking early regular cannabis abuse with increased risk of bipolar disorder are modest, with some data suggesting that cannabis may be one of the environmental factors predisposing men to an earlier onset of bipolar disorder than women in some studies.² Similarly, neuroimaging studies show, in contrast to schizophrenia, relative preservation of brain structures even late in the course of bipolar disorder with some lateral ventricular enlargement and reduction in prefrontal cortex volume and inconsistent findings from other brain structures.⁷

The issue of comorbidity

Differentiation of bipolar disorder from other psychiatric disorders, in particular personality disorders, can be difficult. A recent prevalence survey involving over 41,000 adults in the US highlighted the issue of comorbidity, showing a lifetime risk of having a comorbid anxiety disorder with bipolar disorder at over 40%, alcohol or substance misuse at over 50% and personality disorder at around 60%.⁸ A simple rule of thumb is that with bipolar disorder comorbidity is the rule and almost never occurs in isolation. It is often very difficult early in course to distinguish emotionally unstable, narcissistic and histrionic personality traits from mania and these disorders often co-exist. 

However, these personality traits often become much less florid when mania resolves. In practice, when faced with diagnostic uncertainty it is advisable to take as thorough a history as possible in terms of longitudinal course and precipitants, as well as taking a comprehensive collateral history to evaluate whether apparent personality difficulties appear to predate onset of manic episode. 

It is important to assess symptoms over time and avoid prescribing mood stabilisers, which may have lifelong implications and significant risks, especially for young women in terms of weight gain, contraception and teratogenicity, until a clearcut diagnosis is established. Similarly, symptoms of agitated depression can be hard to distinguish from mania as patients can complain of racing thoughts, insomnia, obsessional ruminations and restlessness. The age of the patient can be helpful as first-episode mania is relatively rare from mid-life onwards and agitated depression common, often with a preoccupation with somatic symptoms or nihilistic delusions. More importantly, mood is invariably low with anhedonia in agitated depression, whereas mood will be irritable or grandiose in mania.

High lifetime recurrence 

Outcome of bipolar disorder is problematic, with a lifetime recurrence risk of 90-95%, with over 50% of patients showing long-term problems in psychosocial and economic functioning.^9,10 Unipolar mania is very uncommon in Caucasian patients with most suffering from multiple depressive episodes over their life.⁹ Episodes of depression tend to be associated with worse psychosocial outcome.¹⁰ Long-term, patients with bipolar disorder spend on average half their lives with symptoms of bipolar disorder, mainly depression at subthreshold levels.⁹ Chronic symptomatology, particularly residual depression, strongly predicts full recurrence and poor psychosocial functioning.¹⁰

Recent studies have shown widespread impairment in neuropsychological functioning in patients with bipolar disorder, particularly in attention, verbal and visual memory and especially in executive functioning.¹⁰ These deficits are seen even in patients who have been asymptomatic for long periods of time and are associated with severity of illness, the number of previous episodes (especially of mania), duration of episodes and chronicity of symptoms.¹⁰

These neuropsychological deficits are associated with psychosocial impairment and like clinical symptoms are increasingly being considered as a core feature of bipolar disorder.¹⁰ From the above, it follows that early detection and appropriate treatment may improve clinical and psychosocial outcomes and even neuropsychological functioning. Sadly, delays in recognition and treatment are commonplace with on average about 8-10 years elapsing between onset of symptoms and first treatment.^1,10   

There has been increasing recognition that bipolar disorder can have an onset in childhood or adolescence. Symptoms appear to be similar to those of adult onset mania but are often misdiagnosed as being due to attention deficit hyperactivity disorder or conduct disorder.¹ Epidemiological and cohort studies have shown an association between aggression and behavioural problems in childhood and early onset of bipolar disorder especially in men which may well be early manifestations of bipolar disorder.² Regarding gender, women tend to have more complex forms of bipolar disorder and a worse prognosis. Women have more periods of depression at both syndromal and sub-syndromal levels which are associated with chronicity and a worse prognosis.¹¹ Women have a later onset of bipolar disorder in most studies as well as more rapid cycling and mixed episodes.¹¹ Finally, women are much more likely to develop bipolar II disorder and have higher rates of comorbidity both with psychiatric disorder and personality disorders except for substance misuse.¹¹

Approaches to mania and to depression

Pharmacological treatment of bipolar disorder has not changed greatly in recent years with atypical antipsychotics and valproate chrono being largely used to treat manic symptoms, and lithium, antipsychotics, lamotrigine and to some extent antidepressants being used to treat depression. Given the very high recurrence risk, maintenance is now being widely used after even one manic episode with lithium still being the gold standard, with in addition lamotrigine, and atypical antipsychotics being used in those with a predominantly depressive polarity and atypical antipsychotics, valproate chrono and carbamazepine being used in those with a predominant manic or mixed polarity.¹

There is considerable debate about when to use antidepressants in bipolar depression with the US academic view advocating never using antidepressants given the risk of inducing mania and rapid cycling, and the more pragmatic European view being that antidepressants can be used but only with other mood stabilisers.¹

Given the burden and disability associated with bipolar depression it is often justifiable to treat bipolar depression with antidepressants, generally only single action antidepressants such as SSRIs, but only if other mood stabilisers are used concomitantly and if the natural course of the illness in terms of significant depressive symptomatology justifies this approach. There are also emerging data showing antidepressant effects from atypical antipsychotics, especially for olanzapine and quetiapine.¹

Early identification and specialised pharmacological and psychological treatments around lifestyle and compliance offer hope in averting chronicity, illness burden and cognitive change in bipolar disorder.

Are psychological therapies helpful?

Outcome of randomised controlled trials of psychotherapy in bipolar disorder have been disappointing with best results coming from pragmatic individual or group therapy approaches emphasising psychoeducation, lifestyle regulation, treatment adherence and relapse prevention.^1,12 Similarly, intensive specialised care in affective disorders services as an inpatient followed by psychoeducational day patient programmes and then shared GP and psychiatric care when things have stabilised offer best medium-term outcomes.¹

Cognitive behavioural therapy (CBT), using many of the approaches described above, has shown promise in small trials in bipolar disorder.¹ However, the largest Medical Research Council multicentre randomised controlled trial of over 250 patients failed to show any benefit of CBT over treatment as usual.¹² However, some data suggest that patients early in their course of illness and with few manic episodes showed significant improvement with CBT, whereas those with a chronic course and multiple episodes actually did worse with CBT.¹²   

Specialised treatment centres in an Irish context are rare but are available for those with health insurance cover in St Patrick’s University Hospital (SPUH).                  

Case study 1

A 54-year-old married woman was admitted, by her GP and psychiatrist, for an eight week period with a recurrence of  severe depressive illness with strong suicidal ideation (ICD-10 F32.2). On admission she was severely depressed, with biological symptoms of depression including suicidal ideation, anhedonia, poor energy and motivation, sleep and appetite disturbance and total social withdrawal. She described periods of profound depression lasting for two to three weeks coming on in a cyclical fashion every six weeks or so. During these periods of depression she withdrew  from activities and spent all her time in bed, leading her to feel even more hopeless and suicidal. These episodes were typically severe for about two weeks before remitting slowly over the following week.

Her depressive illness began 15 years earlier and had led to eight psychiatric admissions with two serious suicide attempts previously. She had numerous antidepressant trials including most SSRIs, lofepramine, a tricyclic antidepressant (TCA) and lithium at therapeutic doses as well as supportive psychotherapy with little benefit. She had no symptoms of depression until her mid-30s. Her depression had led to marital strain and to her taking early retirement from the work she loved as well as hopelessness about the future. 

On taking a detailed longitudinal history and closely assessing symptoms with both the patient and her family it became evident that her periods of low mood were followed by hypomanic symptoms, namely an enhanced feeling of wellbeing, a tendency to be more giddy, chatty and outgoing than normal, a desire to overspend and irritability and anger towards her family that was out of character. This hypomanic spell would typically last for about a week after which she described feeling normal for about two weeks before becoming depressed again. This pattern had persisted for much of the previous 15 years and the cycles had become more frequent after treatment with antidepressants.

Her diagnosis was changed to bipolar affective disorder, rapid cycling (ICD-10 F32.8) and antidepressants and lithium were withdrawn slowly, as antidepressants often worsen frequency of mood cycles. She was started on sodium valproate chrono, which along with carbamazepine, are the mood stabilisers recommended for treatment of rapid cycling mood disorder.¹ This was titrated to a dose of 1400mg/day. During this time she noticed a lessening in frequency of mood cycles though she still experienced short-lived periods of depression with no hypomanic periods. Towards discharge she was started on lamotrigine, a predominantly antidepressant mood stabiliser, which was slowly titrated to 250mg/day as an outpatient. Additionally, she had frequent sessions of individual and group cognitive behavioural therapy to address her negative core beliefs, to help her with behavioural activation and maintain a structure rather than withdrawing socially when she felt depressed. Additionally, she received psychoeducation about medication adherence and lifestyle balance.  

Over six months post-discharge as lamotrigine was being titrated, she continued to experience periods of depression every two months, though these periods were shorter and less severe than before and she coped with them more effectively. Since early 2011 she has had no major recurrence of depression or hypomania and has been attending outpatients for bloods and a consultation only every six months. She describes one to two days every two to three months where she feels a little low but does not feel functionally impaired. Her relationship with her family has greatly improved. She is retraining and undertaking voluntary work regularly.   

Discussion

Where depressive episodes appear to be relatively short-lived and occur frequently it is important to consider bipolar disorder and to take a thorough longitudinal history from the patient and family members who have observed mood swings. Periods of irritability, rapid speech, overspending or an enhanced sense of wellbeing are important to enquire about as hypomanic symptoms are often subtle and easily missed. Studies have shown that if a careful history is taken, about 5-15% of those with recurrent depression will actually have bipolar disorder or bipolar spectrum disorder.⁴ These distinctions are important from a treatment and prognosis point of view. Rapid cycling mood disorders tend to worsen both in frequency and intensity, as in this case, with antidepressant treatment and mood stabilisers, particularly valproate chrono and carbamazepine, are treatments of choice, particularly for manic or mixed symptoms. Mood stabilisers with a predominantly antidepressant effect such as lamotrigine or atypical antipsychotics may also be beneficial for depressive symptomatology. Caution is advised in using mood stabilisers, particularly valproate and carbamazepine, in young women due to teratogenicity, weight gain, enzyme induction and association with polycystic ovarian syndrome, though used carefully these medications can significantly improve prognosis and functioning.¹ Cognitive behavioural therapy and psychoeducation around adherence, monitoring symptoms and lifestyle are the most evidence-based psychological treatments for bipolar disorder, particularly early in course.¹²         

Case study 2

A 45-year-old separated businesswoman with an eight-year history of refractory depression was referred for inpatient care by her GP. She described feeling depressed with diurnal variation, had poor energy and motivation, poor concentration, binge eating with weight gain and oversleeping, as well as suicidal thoughts. Over this time she had difficulty functioning to any meaningful level in her workplace and had socially withdrawn, spending most of her time on her own despite good support from her GP and local psychiatrist and adequate trials of different SSRIs, TCAs, venlafaxine and SSRI/TCA combinations. On admission she described a persistent low mood, suicidal ideation and poor concentration. However, she also complained of racing thoughts and had obvious pressure of speech. She was irritable and appeared paranoid towards other patients and nursing staff, claiming that people were making disparaging comments about her and staring at her. As she had features of both depression and mania she was considered to be in a mixed mood state and a diagnosis of bipolar II disorder was considered.

Treatment for a mixed mood is usually similar to mania and, as such, antidepressants were withdrawn, and she was commenced on mood stabilisers with an antimanic effect, namely valproate chrono to a dose of 2000mg and quetiapine XR to a dose of 400mg. She also attended psychoeducational and group psychotherapeutic programmes over her admission and was referred for individual CBT with the objective of tackling her poor self-esteem and core negative beliefs, as well as behavioural activation. Her mood settled quickly after discharge, though she had residual symptoms of depression with anxiety and some fatigue on discharge. Over the following six months on attempting to return to work she became severely depressed with social withdrawal and significant suicidal ideation. She was started on antidepressants, namely venlafaxine XL, an energising SNRI which was gradually titrated to 300mg. She reported a partial improvement after two months and was able to return to part-time work. This was later augmented with mirtazapine, a predominantly noradrenergic antidepressant with good effect. She has also continued individual CBT in addition to attending day patient programmes focusing on mindfulness, and interpersonal psychotherapy. She was in full remission and returned to full-time work by the summer of 2011 and has remained well since on combination high-dose antidepressants and mood stabilisers, though with brief breakthrough symptoms each winter which have resolved with short-term additional antidepressant augmentation.

Discussion

It is important to consider bipolar II disorder when patients present with chronic refractory depressive episodes that respond poorly to antidepressants, and to look for historical mixed moods or hypomanic symptoms. There are little evidence-based data to guide treatment of bipolar II disorder, with antidepressants being commonly prescribed though usually with concomitant use of mood stabilisers.⁴ Psychiatrists often use mood stabilisers with a predominantly antidepressant mode of action such as lamotrigine, though evidence remains weak.⁴ Venlafaxine, particularly at higher doses due to a linear dose effect, has been considered to be particularly effective in treating refractory depression, as a higher proportion treated with venlafaxine achieve full remission than with most SSRIs and meta-analyses favour venlafaxine over SSRIs.¹³ Similarly, in refractory depression, augmentation of the initial antidepressant with drugs such as lithium, low dose atypical antipsychotics, mirtazapine and bupropion are favoured particularly in those with partial response, as such approaches improve the possibility of achieving full recovery which is likely to improve long-term prognosis and reduce risk of recurrence.¹³ Cognitive behavioural therapy, focusing on cognitive distortions and negative core belief systems, has been shown to be effective in preventing recurrence in those with chronic depression for up to five years in recent well conducted RCTs but these have involved as many as 20-30 individual sessions of CBT over a six to 12 month period.¹⁴       

References

1. Goodwin FA, Jamison KR. Manic Depressive Illness: Bipolar Disorder and Recurrent Depression 2nd Edn. Oxford University Press, 2007 
2. Kennedy N, Boydell J, Fearon P, et al. Gender differences in incidence and age at onset of mania and bipolar disorder: results from a 35-year study. American Journal of Psychiatry 2005; 162:262-267  
3. Kennedy N, Everitt B, Boydell J, et al. Incidence and distribution of first-episode mania by age: results from a 35-year study. Psychological Medicine 2005; 35:855-863
4. Parker G. Bipolar II Disorder: Modelling, Measuring and Managing. Cambridge University Press, 2008
5. Diagnotic and Statistical Manual IV (DSM-IV). American Psychiatric Asssociation (APA), 2000
6. Lloyd T, Kennedy N, Fearon P, et al, on behalf of the AESOP study team. Incidence of bipolar affective disorder in three UK cities. Results from the AESOP study. British Journal of Psychiatry 2005; 186:126-131  
7. McDonald C, Zanelli J, Rabe-Hesketh S, et al. MRI meta-analysis of regional brain volumes in bipolar affective disorder. Biological Psychiatry 2004; 56:411-417
8. Grant BF, Hasin DS, Stinson FS, et al. Co-occurence of 12-month mood and anxiety disorders and personality disorders in the US: results from the national epidemiologic survey on alcohol and related conditions. Journal of Psychiatric Research 2005; 39:1-9
9. Judd LL, Schletter PJ, Akiskal HS, et al. Long-term symptomatic status of bipolar I vs. bipolar II disorder. International Journal of Neuropsychopharmacology 2003; 6:127-137 
10. Kennedy N, Foy K, Sherazi R, et al. Long-term psychosocial functioning after severe depression. Bipolar Disorders 2007; 4:25-37
11. Arnold LM. Gender differences in bipolar disorder. Psychiatric Clinics of North America 2003; 26:595-620
12. Scott J, Paykel E, Morriss R, et al. Cognitive behavioural therapy for severe and recurrent bipolar disorder: randomized controlled trial. British Journal of Psychiatry 2006; 188:313-320
13. Kennedy SH, Lam RW, Nutt DJ, Thase ME. Treating Depression Effectively. Applying Clinical Guidelines 2nd Edn. Informa Healthcare, 2007 
14. Paykel ES, Scott J, Cornwall PL, et al. Duration of relapse preventation after cognitive therapy in residual depression: follow-up of controlled trial. Psychological Medicine 2005; 35:59-68