Treatment Options in Bipolar Disorder: Mood Stabilizers

Charles L. Bowden, MD

Medscape Psychiatry & Mental Health eJournal. 1997;2(4) 

In This Article

Mood-Stabilizing Agents: Li

The term "mood-stabilizing agent" was originally used when referring to Li because of its well-known effects in stabilizing the mania in patients with bipolar disorder.[5] The mood-stabilizing properties of divalproex and CBZ have also since been recognized. The term mood-stabilizer distinguishes the aforementioned drugs from those that have been used to treat bipolar disorder but have the potential to worsen certain aspects of the disease, such as antidepressants.[5] In general, the term applies to drugs that alleviate the frequency and/or intensity of manic, hypomanic, depressive, or mixed episodes in patients with bipolar disorder.[5,6]

Although the mechanism of action of Li has not been definitively identified, it seems to attenuate diverse circadian rhythms that may be part of the rhythmicity inherent in manic-depressive illness.[3,12,13,14,15] It is theorized that Li exerts a common effect on the overactive processes that mediate the opposite mood phases of mania and depression. An example of this is the increased dopaminergic or noradrenergic tone thought to be associated with mania and the depression-related increases in acetylcholinergic tone. Hence, the paradox of Li's usefulness in both phases of the illness.

An alternative thesis is that Li interferes with overactive phosphatidylinositol metabolism, an intracellular biochemical signaling system that may be disturbed in bipolar disease. Some of Li's psychotropic actions may be caused by its reported ability to inhibit a variety of intracellular second-messenger systems, such as neurotransmitter-stimulated adenylate cyclase. These intracellular transduction mechanisms can play critical roles in cell cycling and the generation of circadian and other rhythms.[3,12,13,14,15]

Different mechanisms inherent in Li's action may mediate specific components of its effect on the manic-depressive cycle, which would explain the timing of its effects and suggest potential links to its long-term prophylactic and anticycling effects. For example, data suggest that it blocks the rebound phenomenon associated with recovery sleep, which can lead to a depressive relapse following mania-associated sleep deprivation. The onset of Li's effect on acute mania and depression is slow, but its potentiation of other agents' effects--especially those of depression--can be very rapid.[3,12,13,14,15]

A variety of controlled and uncontrolled studies have been done to assess the response to Li in acute mania.[16,17,18,19,20,21,22,23] However, the studies must be evaluated with attention to several methodologic limitations: Diagnostic criteria for mania were nonspecific; manic states were not distinguished from schizoaffective states; and rating scales were not used to measure clinical response. Also, many studies were single-blind.[1,24]

The controlled studies used small samples and defined response differently.[16,17,18,19] Nevertheless, Li was superior to placebo and comparable in efficacy to antipsychotic and anticonvulsant agents. Pooled response rates from double-blind, placebo-controlled studies of Li in the treatment of acute mania showed significant improvement in 70% of patients.[19,20,21,22,23] Yet, despite its antimanic effect, it is also clear that in a large percentage of patients (as many as 50%) the response was inadequate. Thus, the overall response rate in acute mania should actually be considered to be between 50% and 70%.[24]

Few controlled clinical studies have been conducted in patients with acute bipolar depression.[24] In most studies, Li was found to be superior to antidepressants.[25] However, many clinicians treating bipolar depression have found that many patients do not respond to Li alone.[24]

Early reports (from the 1960s and 1970s) demonstrated Li to be superior to placebo in preventing affective episodes in placebo-controlled maintenance studies in which patients were monitored for up to 1 year. Relapse rates for patients taking Li averaged 37% compared with 79% for those on placebo.[26] Most studies suggested that Li had a better prophylactic effect on manic rather than depressive episodes. Further support for the use of Li maintenance therapy came from Suppes and colleagues,[14] who found that more than 50% of patients relapsed within 6 months of Li discontinuation.

However, more recent naturalistic studies for longer follow-up periods suggest that a substantial number of patients do not respond adequately to Li maintenance.[24] Rates for those who remained episode-free for a period ranging from 18 months to 5 years were 33%, 36%, and 37% in certain studies.[27,28,29]

Bowden[4] found that a family history of bipolar disorder and a previous episode sequence of manic-depressive illness-euthymia were associated with a favorable response to Li. Conversely, patients with rapid cycling, mixed mania, episode sequence of depression-mania-euthymia, multiple prior episodes, comorbid substance or alcohol abuse or dependence, and familial negative affective styles were less likely to respond to Li maintenance.[24]

Li is a safe, efficacious, inexpensive mood stabilizer that revolutionized the treatment of bipolar disorder (Table I). Until recently, it was considered the initial treatment of choice for first episodes of manic-depressive illness because it has high efficacy in euphoric manic states.[8] While it is highly efficacious in patients with euphoric mania, it is not equally useful for all categories of bipolar disorder. Use of alternative mood stabilizers (eg, VPA, CBZ) should be considered for those who are likely to be Li-resistant, as outlined below. Some data indicate that Li lowers mortality of bipolar disorder by reducing the rate of suicide.[12,30,31,32,33]

Published and anecdotal reports indicate that 30% to 50% (or up to 80%, depending on subtype) of patients do not respond to Li. People with severe mania, psychosis, and especially mixed states (ie, depressive mania) respond relatively poorly.[4,14,24,30,34,35]

The response is better when cycles are long than with rapid cycling. As mentioned, response to Li also varies according to episodic sequencing: Li is more effective in preventing biphasic episodes that begin with a mania and switch to a depression followed by an interval of wellness (M-D-I) than those that begin with a depression and switch to mania and then to an interval of wellness (D-M-I).[24] A history of substance abuse (including alcoholism), neurologic insult, or developmental disorders predicts a poor response, perhaps indirectly through the association of these factors with mixed-states disorder.[4,5,30]

Li's narrow therapeutic range is well recognized: 0.8 to 1.4mEq/L measured at trough (12 hours postdose). Therefore, effective use requires careful dose titration and mandatory serum monitoring with dosage adjustments as needed. Subsyndromal affective symptoms are more likely to occur with low levels (0.4 to 0.6mEq/L) and usually will be followed by a full-blown episode. Full-blown manic episodes are more common after hypomanic symptoms.[30,31] By contrast, subsyndromal depression does not presage full-blown episodes of either mania or depression.

Li also has a narrow therapeutic-to-toxic index--that is, toxic effects occur at serum levels that are within or close to the therapeutic range. Another practical disadvantage of Li is the time lag before it takes effect in the treatment of acute mania: A beneficial response is not typically observed until after 7 to 14 days of therapy.

Li is efficacious as initial therapy for mild manic symptoms, (ie, stage I mania and hypomania). When used in conjunction with benzodiazepines to restore the normal sleep pattern, it can often avert escalation to more severe mania. In cases of severe mania, however, particularly when psychotic symptoms and psychomotor agitation predominate, anticonvulsants (ie, VPA, CBZ) and neuroleptics are superior to Li during the first weeks of an episode. After the first 2 weeks, Li and, probably, the anticonvulsants, are more effective than neuroleptics. In addition, mood stabilizers are less sedating than neuroleptics and, unlike the neuroleptics, do not cause tardive dyskinesia.[1,36]

The choice of formulation (tablet or solution) is usually based on expense and patient preference, with no therapeutic benefit of any 1 over the others. The citrate solution, of course, is indicated for a person who cannot swallow capsules or tablets. Li may be given in a single daily dose, but bid dosing usually produces fewer side effects (eg, nausea and diarrhea). Three or 4 doses may be tried for those who experience gastrointestinal (GI) or central nervous system (CNS) effects.[1]

Li is readily absorbed from the GI tract, and taking it with food has no apparent effect on its bioavailability. The citrate solution of Li (lithium citrate syrup) is the most rapidly absorbed, with peak concentrations reached within 15 to 60 minutes. Absorption from the conventional tablets and capsules takes up to 6 hours, with peak serum levels occurring within 4 to 12 hours. Absorption from extended-release tablets is both delayed and prolonged; they take nearly twice as long as conventional preparations to reach peak serum levels and are not as completely absorbed (60% to 90%, compared with 95% to 100%). There is a biphasic decline in serum concentration, with an initial half-life of less than 1.5 hours and a terminal half-life of 20 to 27 hours. However, the half-life can be more than twice as long in patients who have been taking Li for 1 year or longer.[1]

Li therapy is usually contraindicated in patients with a history of renal function impairment, acute myocardial infarction, myasthenia gravis, or compromised fluid or salt homeostasis, and in women who are pregnant (first trimester) or nursing. Other conditions in which the risks versus benefits of Li should be carefully considered before the drug is prescribed include advanced age, cardiac disease, Parkinson's disease, pregnancy (second to third trimesters), epilepsy, thyroid disease, cerebellar disorders, dementia and other CNS disorders, diabetes mellitus, ulcerative colitis, psoriasis, and osteoporosis.[1]

Li can affect virtually any organ system, but side effects most frequently involve the CNS, GI tract, and kidneys. The most commonly reported adverse effects are excessive thirst, polyuria, CNS impairment (ie, memory problems, drowsiness, and fatigue), tremors, weight gain, and GI reactions (ie, diarrhea, stomach upset).[1]

Weight gain is among the most disturbing side effects reported with Li; approximately one quarter of all patients experience a gain of 4.5kg (10 lbs) or more. In the majority of patients, the weight gain is not caused by water retention. Li is known to alter carbohydrate metabolism and cause hypothyroidism. Thus, patients must be counseled to restrict their carbohydrate intake and to increase their exercise regimen. Some patients may benefit from thyroid-hormone supplements.[1]

Most of the side effects of Li are dose-related and occur when the trough serum level is greater than 1.0 to 1.3mEq/L; dose reduction may resolve these side effects.[1] However, for many patients, particularly children and the elderly, CNS effects can occur at any drug level. For some adverse effects, intervention with another drug therapy may be of use. For example, Li-induced tremor can sometimes be controlled with the use of low-dose beta-blockers (propranolol or metoprolol).[1]

Although many of the common Li-associated adverse reactions are not serious or life-threatening, they are unpleasant and often intolerable, even when the Li dose is lowered. Patients who are unwilling to tolerate the adverse reactions or to take another medication to offset them generally refuse to comply with the prescribed Li regimen. For these individuals, alternative mood-stabilizer therapy should be considered.[1]

A number of clinically important drug interactions occur with Li (Table II). Some of these can result in elevation of the serum Li level and a subsequent Li-related adverse event, or may even precipitate acute Li intoxication (eg, diuretics). Concomitant use of Li and interacting drugs should be avoided. Although VPA may decrease the serum Li level in some patients, there appears to be no mutual enhancement of drug toxicity with Li-VPA combinations. In contrast, concomitant use of Li-CBZ may result in additive CNS adverse effects and neurotoxicity unless the doses of each drug are modified.[1]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.