The treatment of bipolar disorder is fraught with controversy.  The two most controversial issues include the definition of a "mood stabilizer" and the use of antidepressant medications for the bipolar depressed patient.  A true mood stabilizer should ideally treat acute mania, acure depression, and prevent both relapse and recurrence of either mania or depression.  By that definition there is really no true mood stabilizer outside of electroconvulsive therapy (ECT), which is successful in treating all categories of bipolar illness.  Typically, therefore, most clinicians equate the term mood stabilizer with a medication that treats acute mania and prevents its recurrence.  More appropriately, these should be termed antimanic agents, as they are parallel to antidepressant agents and treat the other end of the mood spectrum.  All mood stabilizers fail in the treatment of acute depression.  the jury is out as to whether they prevent mania.  For this reason antidepressants are still commonly used.  But because antidepressant medications may switch an individual from depression into mania and actually worsen one's overall condition, the issue of antidepressant use remains controviersial.  Complicating this issue is the fact that clinicians define a mood stabilizer as any agent that treats one arm of the bipolar spectrum without causing switching to the other arm.  By this definition, some antidepressants may meet this standard, though the jury is out in this regard as well.  It does appear that some classes of antidepressants cause less switching than others.  These classes include the SSRI's and buproprion.  Also by this definition the anticonvulsant Lamictal (lamotrigine) has been called a mood stabilizer even though it is clearly more effective in preventing a depressive relapse or recurrence than a manic relapse or recurrence.  Thus, the term mood stabilizer is generally used very loosely to describe any medication that treats "mood swings," an equally vague term that can mean just about any type of emotional change, even those associated with personality disorders such as borderline personality.  One final caveat: most antimanic agents stabilize neuronal cell membranes.  As a result, any type of overstimulation of the central nervous system, whether it is seizure activity, mania, a panic attack, or explosive rage, can respond to an antimanic agent.  Therefore, the fact that one's mood is "stabilized" by an antimanic agent does not necessarily mean one is bipolar.

With that introduction let us now proceed with the different classes of medications that are used in the treatment of bipolar disorder.  The classes more specifically break down into several of the following categories in order of understanding and importance in treating the condition:

• Lithium earbonate and its different formulations
• Anticonvulsant medications
• Atypical antipsychotic medications
• Typical antipsychotic medications
• Benzodiazepines
• Antidepressant medications (specifically the selective serotonin reuptake inhibitors, or SSRIs)
• Calcium channel blockers
• Mood stabilizers
• Others under investigation

Note that the antimanic effects of anticonvulsants do not exhibit a class effect.  That is, just because a medication is considered an anticonvulsant one cannot immediately assume it has antimanic properties.  This was prominently demonstrated after the medication Neurontin (gabapentin) was touted for its potential antimanic effects prior to any clinical trials, which later demonstrated that it was not superior to placebo and led to a series of lawsuits against the manufacturer for false promotion.  Alternatively, both the typical and atypical antipsychotics all demonstrate antimanic effectiveness and therefore clearly exhibit a class effect.

The majority of medications used in the treatment for bipolar are used off-label as they are not FDA approved.  First, the FDA is slow to approve medications.  Second, pharmaceutical companies are slow to perform clinical trials.  Trials are expensive, and once a drug is approved for acute mania, the need to seek approval for the other aspects of the condition diminishes considerable unless increase in market share can be anticipated with FDA approval.  Third, once a drug becomes generic, only the government will spend money on a clinical trial.  Finally, just because there are negative clinical trials does not mean that a medication is no longer used.  There is enough individual variability in bipolar disorder to not immediately discount any medications, particularly when one is refractory to those medications that have proven beneficial.  This is one reason why Neurontin (gabapentin) continues to be used, although in a very limited, circumscribed manner today than prior to the uproar around it.  There are individuals who may idiosyncratically respondto one medication just as there are individuals who have idosyncratic paradoxical responses to another, neither of which condition can be predicted.

There also appear to be many differences between bipolar I and bipolar II disorder.  All clinical trials leading to FDA approval are based on bipolar I disorder, and it is not clear if these medications have the same effect on bipolar II disorder.  It is also not clear if bipolar II disorder is truly a "spectrum" disorder--that is, lying on a continuum with bipolar I disorder rather than a unique entity itself.  This uncertainty is partly based on genetic studies that suggest that bipolar I disorder shares more genes in common with schizophrenia and schizoaffective disorder than with bipolar II disorder.  It is also partly based on at least some clinical reports that bipolar II patients may respond differently to the listed medications than bipolar I patients.

That being said, the general guideline makes the following recommendations.  If a patient with acute mania, use of one medication is indicated; the choice between lithium and Depakote (valproate) is based partly on each medication's side-effect profiles as well as the presence or absence of various symptoms.  For example, for patients who have suffered from depression but are currently manic, lithium is recommended.  Lithium appears to be the closest to meeting the definition of a mood stabilizer in that it appears to prevent both depression and mania.  For those patients who have recurrent manic episodes without depression, anxiety/agitation, or a substance abuse disorder, valproate is recommended.  Patients with rapid cycling symptoms (four or more episodes annually) may respond better to either Equetro (carbamzepine) (if the episodes include more depression) or Depakote (valproate) (if the episodes include more mania).  For manic patients with psychotic features (delusions, hallucinations, and/or grossly disorganized thinking and behavior) the addition of an atypical antipsychotic medication is recommended, but most atypical antipsychotics have FDA approval for monotherapy treatment of an acute manic episode.  If symptoms are not adequately controlled within 10 to 14 days, addition of a second first-line agent is indicated (e.g., adding an antipsychotic if not already prescribed).  Clozaril (clozapine) may be effective in refractory cases.

For bipolar depression, either Lamictal (lamotrigine) alone (or with an antimanic agent) or Symbyax should be initiated.  If one fails to respond to these two strategies the addition of Seroquel (quetiapine) or lithium may be recommended.  Failing that, an additional antimanic agent plus an SSRI, buproprion, or Effexor (venlafaxine) may be added.  Electroconvulsive therapy (ECT) is considered for more refractory cases (as well as for refractory mania).

Trials of nontraditional medications such as calcium channel blockers, stimulants, or thyroid hormone should be considered in conjunction with all of the above when symptoms cannot be adequately controlled.  This is known as rational polypharmacy, as medications from different classes with different actions are added that may have a synergistic effect to improve mood stability.