Bipolar spectrum disorders in primary care: optimising diagnosis and treatment

Depression is an extremely common presentation in primary care and the public health importance of depressive disorders are now very well established.1^,2 Patients with bipolar affective disorder (who experience episodes of depression alternating with episodes of mania or hypomania), frequently present to their GPs with difficult-to-treat depressive episodes. Indeed, for most of these patients, depressive symptoms (rather than manic symptoms) dominate the long-term clinical course of their illness.

Bipolar disorder type I (BD-I; depression alternating with mania) affects around 1% of the population and bipolar disorder type II (BD-II; depression alternating with hypomania) affects a further 2–3%.3 Although the clinical features of BD-I and BD-II are widely known, it is less well recognised that the boundary between bipolar disorder and recurrent unipolar depression is far from clear-cut.4 A significant proportion of patients with unipolar depression experience mild or brief episodes of hypomania which fall below the threshold for a formal diagnosis of BD-I or BD-II.5^–7 These ‘bipolar spectrum’ patients often have patterns of depressive episodes, comorbidities, and treatment responses that differ from those with more straightforward unipolar depression and which, therefore, require a different approach to diagnosis and management. This generally under-recognised issue has far-reaching implications for the way in which clinicians, particularly GPs, approach the assessment and management of all of their depressed patients.

Within the research community, there is an emerging consensus that the diagnostic criteria for hypomania are overly restrictive and result in many patients with significant bipolar symptoms being placed within the broad diagnostic category of unipolar depression.7^–9 Converging evidence from a number of studies from around the world suggests that at least 25% of patients with recurrent unipolar depression may be better classified as having a broadly-defined bipolar spectrum disorder.5^–7^,10^–12 Importantly, these patients have similar rates of bipolar family history and comparable levels of health service use, long-term illness severity, and psychosocial morbidity as patients with bipolar disorder diagnosed according to the formal DSM-IV criteria.6^,7^,10 Furthermore, certain clinical subgroups of depressed patients, such as those with an especially severe or early-onset of depression13^,14 or those with treatment resistance to antidepressants,15 appear to have particularly high rates of unrecognised bipolar disorder.

For a number of reasons, BD-I and BD-II disorder (and, by extension, bipolar spectrum disorder) can be very difficult to diagnose.16 Most BD-I and BD-II patients will experience a delay of several years between their first onset of significant manic symptoms and receiving the correct diagnosis, and misdiagnosis in the interim as unipolar depression, personality disorder, a primary drug or alcohol problem, or even schizophrenia is relatively common.16^–19 Given that depression (rather than mania) dominates the natural history of bipolar disorder, patients are much more likely to seek help, usually from their GP, during their low periods.17^,18 Therefore, consultations are inevitably ‘depression centric’ and it is not currently usual practice to systematically assess all depressed patients for a past history of manic symptoms. This problem is compounded by poor patient insight into their prior experiences of hypomania and mania and by infrequent use of corroborative histories taken from close relatives, which can often be extremely helpful for identifying previous manic symptoms. Similarly, although the use of screening questionnaires for depression, such as the Patient Health Questionnaire (PHQ), the Hospital Anxiety and Depression Scale (HADS), and the Beck Depression Inventory (BDI), is now routine in primary care, currently very few clinicians make use of available screening instruments for hypomania19 such as the Mood Disorder Questionnaire (MDQ),20 the Bipolar Spectrum Diagnostic Scale (BSDS),21 and the Hypomania Checklist (HCL-32).22 All of these instruments are relatively brief and have been validated for use in a range of clinical settings. The BSDS may be particularly useful for identifying hypomania in primary care patients with depression.21

Perhaps the most important practical issue for patients with an unrecognised bipolar spectrum disorder concerns the use of antidepressants. Recent evidence suggests that antidepressants may be of limited therapeutic benefit in the treatment of bipolar depression23^,24 and may even be unhelpful by causing destabilisation of mood,25 more frequent mood episodes,26 treatment resistance,27 and possibly also (especially in young bipolar patients) an increase in suicidal behaviour.28^–30 Although caution is required when extrapolating findings from studies of BD-I and BD-II patients to groups of patients with broadly-defined bipolar spectrum conditions, the principle of ‘first do no harm’ should guide decisions on the use of antidepressants in this group. Unfortunately (largely due to the reluctance of drug companies and non-industry research funders to investigate diagnoses which go beyond formally accepted categories) there have been very few studies, to date, which have assessed the risks and benefits of antidepressants in patients with broadly-defined bipolar depression.31

A pragmatic approach to this problem is to consider avoiding antidepressants in depressed individuals where they have previously been unhelpful on several occasions (either because of lack of response or adverse effects) and instead consider treatments that are known to be effective for bipolar depression, such as lamotrigine or quetiapine.32 This is not to say that these medications should be prescribed without careful consideration. Lamotrigine has been (rarely) associated with serious skin reactions and quetiapine, as an antipsychotic, can carry quite a high side-effect burden for some patients. Where there is no clear history of poor or adverse response to antidepressants these could be prescribed, but this should always be with mood stabiliser ‘cover’ and the antidepressant should be withdrawn within 3 months of recovery from the depressive episode (to avoid inducing hypomania, mania, or more rapid cycling of mood episodes).32^,33

In the UK, the current focus of secondary care psychiatric services is on the management of individuals with severe and enduring mental illnesses such as schizophrenia or BD-I. Depressive disorders, including those which prove to be relatively treatment-refractory, are increasingly managed solely by GPs. Given that many of these patients may have an unrecognised bipolar spectrum disorder, diagnostic and clinical management input from secondary care would seem to be appropriate. However, in many parts of the country it is difficult for GPs to obtain comprehensive psychiatric assessments on their difficult-to-treat depressed patients and this seems unlikely to change within the near future. Therefore, it is likely to be important for the future that GPs develop competencies in the assessment, diagnosis, and treatment of bipolar depression.

In summary, broadly-defined bipolar disorders are relatively common in both primary and secondary care settings but are often not recognised or diagnosed. This issue is not simply one of academic interest but has important implications for the way that clinicians approach the assessment and management of all recurrently depressed patients, especially those patients who have complex presentations or have not responded well to at least two courses of antidepressant therapy. Changes in the organisation and delivery of psychiatric services in recent years have meant that GPs are managing more and more patients with complex depressive disorders, many of whom may have an undiagnosed bipolar disorder. GPs, as the first port of call for most of these patients, should be aware of the possibility that a significant proportion (especially those with early-onset or treatment resistant depression) may have a primary bipolar disorder. In the future, guidance on screening for bipolar disorders and appropriate management strategies for bipolar depression needs to be made more widely available. In the longer term, it is hoped that a systematic and tailored approach to this under-recognised problem may yield significant benefits in terms of early diagnosis and improved long-term management of recurrent depression for large numbers of patients.

• © British Journal of General Practice, 2010.