Treatment resistance is extremely common. Even under optimal maintenance conditions, almost half of bipolar patients with symptom remission will have a recurrence in 2 years under standard care (including medication combinations).2
Optimizing phase-specific evidence-based treatments is crucial. This may include raising the dosage of an initial treatment according to response unless limited by adverse effects. Acute treatments are often continued into maintenance. Combinations are commonplace, and the number of potential combinations is large. Decisions about which medications to use first or in what combinations, as well as dosing issues, require good clinical judgment on the part of each clinician because there is little evidence to drive such decisions.
Traditionally, lithium(Drug information on lithium) and anticonvulsants have been used as first-line treatments for acute mania; antipsychotics were reserved for resistant, severely ill, or psychotic patients. This practice is based on the long history of lithium and divalproex use rather than on comparative efficacy trials. The range of FDA-approved uses for atypical antipsychotics in bipolar disorder is growing; these agents can now be considered first-line treatments even for moderately ill manic patients.
To dramatically reduce treatment resistance and enhance stability, comorbid conditions, such as substance abuse, need to be addressed. And, destabilizing medications, such as antidepressants, need to be discontinued. Evidence-based intensive psychotherapies are clearly useful in maintenance. For truly resistant conditions, clozapine(Drug information on clozapine) and electroconvulsive therapy (ECT) are also recommended for depression and mania in both acute and maintenance treatment. If these fail, a number of novel treatments have been suggested, and some have been studied in placebo-controlled trials.
Sachs3 suggested that the term “treatment-resistant bipolar disorder” should be reserved for patients who do not respond to a combination of 2 standard medications in a specific period, such as 6 weeks for mania, and 6 months or 3 cycle-lengths for maintenance. Others have required multiple trials of combinations or that patients fail to respond to nonstandard treatments, such as antidepressants.4,5
Treatment resistance in unipolar depression and schizophrenia is usually defined as failure to respond during an acute episode to 2 adequate monotherapy trials of agents with established efficacy and implies that a novel treatment should be considered. In schizophrenic patients, this makes sense because clozapine has shown efficacy for treatment-resistant schizophrenia. In unipolar depression, this conceptualization of treatment resistance may be less useful because failure to respond to one SSRI is not associated with an increased risk of not responding to another.6
Treatment resistance in bipolar disorder is even harder to define. Standard care in any phase often includes more than 1 medication, which implies some degree of resistance in most patients.7 Furthermore, treatments lack any unifying mechanism of action, an evidence-based rationale for combining is lacking, and dosing of combinations has not been standardized.
The process of selecting medication combinations is often simplified by grouping treatments into classes, such as anticonvulsants and atypical antipsychotics—distinctions based on their uses in epilepsy and schizophrenia. Unfortunately, specific medications within these classes have considerably different efficacy.
Phase of illness
The basic approach to managing treatment-resistant bipolar disorder should be to first clarify the phase of treatment and then to optimize evidence-based treatments for that phase.8 This includes titrating the dosage of a standard medication as tolerated and, possibly, considering combinations. For acute depression and maintenance, psychotherapy should also always be part of the treatment plan.
Standard treatments for bipolar disorder are listed in the Table. Some are more effective for treating or preventing mania (eg, lithium, divalproex, carbamazepine(Drug information on carbamazepine), aripiprazole(Drug information on aripiprazole), olanzapine(Drug information on olanzapine)). Others are more effective for treating or preventing depression (eg, lamotrigine(Drug information on lamotrigine), quetiapine(Drug information on quetiapine)).
In acute mania, antidepressants should be discontinued immediately. The focus should be on using evidence-based treatments for mania. Although lithium is not recommended for mixed episodes or for patients with many previous episodes, lithium and divalproex are often used before an atypical antipsychotic because they are thought to be safer with long-term use.
An atypical antipsychotic may have greater efficacy in severe mania and can be added as needed. In most studies of mania, antipsychotic dosing has approximated commonly used dosages for schizophrenia. Monotherapy with each of the atypical antipsychotics has been shown to have evidence-based efficacy in acute mania studies that included participants with psychosis (about 40%).9,10
Should therapy fail, there are many reasonable and creative possibilities for combining standard treatments, although they have not been supported by controlled studies. These include higher doses of an atypical antipsychotic, lithium with divalproex, 2 anticonvulsants, and other combinations.
The most commonly recommended nonstandard treatments for treatment-resistant mania are clozapine and ECT, which have been shown to have efficacy.8,11,12 A combination of clozapine and ECT has also been suggested.13
Several novel treatments have been studied using an augmentation approach in combination with standard treatments for treatment-resistant mania. These include donepezil(Drug information on donepezil), gabapentin(Drug information on gabapentin), topiramate(Drug information on topiramate), mexiletine(Drug information on mexiletine), and intravenous magnesium sulphate.14-20 The reported efficacy in these uncontrolled reports is confounded by the continuation of the previous treatments. One exception is tamoxifen(Drug information on tamoxifen), which, like lithium and valproate(Drug information on valproate), inhibits protein kinase C and was found to have antimanic efficacy superior to placebo.21
