Clinical Practice Guidelines for Bipolar Disorder

MANAGEMENT OF PERSONS WITH PSYCHOSES
Current Manic/Hypomanic/Mixed Episode Annotations

MODULE E

View Flow Chart for Module E.

Go to Module D.

Go to Module F.

A. Determine Appropriate Setting for Care

See Screening Algorithm for Persons with Psychosis and Bipolar Disorder, Module A.

B. Begin Psychoeducational Tasks, Evaluation for Psychotherapy and/or Psychosocial Rehabilitation

Some persons with bipolar disorder benefit from organized psychosocial rehabilitation programs. Readers are referred to the accompanying Psychosocial Rehabilitation Modules for further information.

1. Psychosocial interventionæ Psychosocial intervention should begin at the time of diagnosis. Specifically for bipolar disorder, this will consist of psychoeducation to optimize treatment (see Table 1, Goals and Process of Psychoeducation). Formal psychotherapy for bipolar disorder is usually part of prophylaxis and geared toward relapse prevention and may depend on person and resource factors. Similarly, psychosocial rehabilitation may also be necessary. Psychosocial intervention often complements but never replaces medication treatment (Frances et al. 1996).
2. Psychoeducation is the process by which the clinician and the client develop a treatment plan with mutually acceptable goals and procedures. Psychoeducational tasks are appropriate and necessary for persons in all stages of the illness, from remission to psychotic mania. Depending on the phase of the illness, content and methods may differ. A person with a short attention span or severe symptoms may need to have a brief assessment of illness management skills and brief, concrete explanations. At other times, the person may benefit from and desire more information.
   
   Psychoeducational tasks frequently involve key members of the person's social network, including spouse, other family members, and caregivers. Optimal psychoeducation involves nonmental health caregivers. The successful treatment plan requires agreement between health care providers, the client and family regarding mutually acceptable goals and procedures. The goal of psycho- education, a process of collaborative decisionmaking, and other key tasks are outlined in the table below.

Table 1: Goals and Process of Psychoeducation

1. Assessment: What factors will impact on this person's ability to manage his or her disease?

1. Which comorbid disorders are present or likely to recur?
2. What personal characteristics are likely to contribute to illness management skills? Explore the individual's views about his or her illness and its treatment:

1. What are the person’s beliefs about what causes his or her suffering?
2. What has been his or her past experience in treatment? What has helped?
3. More important, what has not helped and why? Have some treatments been helpful for symptoms, but intolerable in other ways?
4. What has the treatment setting been like for him or her in the past? By an individual? By a team? Time-limited? Open-ended? What are his or her expectations and preferences?
5. Determine the individual's past experience in collaborative decisionmaking around treatment decisions (some will have extensive experience in having a say in treatment; to others, it will be a foreign concept to have any input).

2. Implementation: Setting the stage for making treatment decisions

1. Educate regarding boundary conditions, requirements of treatment, and the practicalities of accessing caregivers between appointments. That is, what can the person expect from treatment, and how do you expect him or her to act?
2. If there are discrepancies between the person’s expectations
3. (A.2) and preferences (B.1), discuss and resolve early in treatment.
4. Discuss explicitly the how-to of access, particularly in crisis (don't underestimate the importance of this: perceived access will dictate degree of participation--i.e., will the person call you when he or she has a side effect or an exacerbation of symptoms, and how and when will you respond?).
5. Recommend adjuvant resources where appropriate (e.g., support groups).

3. Content areas to cover
   Note: These are didactic topics that convey information necessary if the individual is to make informed therapeutic decisions and participate actively in treatment. While a certain amount of information must be conveyed at a general, abstract level, the discussion also should cover more personal aspects of these topics (e.g., "What is my course?" "What are my triggers?" [see Bauer & McBride 1996])

1. Definition and classification/vocabulary issues
2. Course, morbidity, and outcome
3. Biological basis
4. Psychosocial triggers and their management
5. Biological triggers and their management
6. Substance abuse (if not covered in C.4 and C.5)
7. Treatment options from the perspective of costs and benefits to the patient

Adapted from Bauer & McBride 1996, 1997; Kahn et al. 1996.

Formal psychotherapy specific for bipolar disorder may be part of prophylaxis treatment. Psychotherapy for bipolar disorder is undertaken after the person is medically stable and able to benefit from the psychotherapeutic process. Therefore, psychotherapy for bipolar disorder is used in addition to, and not as replacement for, appropriate mood stabilizer and other medication treatment for bipolar disorder. Several types of formal psychotherapy are currently being investigated in bipolar disorder. In contrast to psychotherapies for major depressive disorder, psychotherapies for bipolar disorder do not typically target specific acute mood episodes. Rather, outcome variables have focused on specific longer-term outcomes, either by symptom reduction or functional improvement. The usual goal for psychotherapies for bipolar disorder is relapse prevention. Formal psychotherapies for bipolar disorder are distinguished from generic supportive psychotherapies and from "dynamic" psychotherapies by being highly structured and manual-driven. Psychotherapies for bipolar disorder are therefore amenable to testing in randomized controlled trials and to its translation to various clinical settings with reasonable fidelity to the original intent and methods. Several randomized controlled trials are underway involving these formal psychotherapies, and data documenting efficacy should soon become available.

Less well-defined forms of psychotherapies are often used for persons with bipolar disorder. Generally, supportive psychotherapy and psychoeducation are useful for bipolar persons. However, probing, "uncovering," and "insight-oriented" psychotherapies are not effective and may be hazardous in persons with unstable, psychotic bipolar disorder or poorly compensated persons (Kanas 1993; Frances et al. 1996; Kufferle 1988). Therefore, "dynamic" and unstructured forms of psychotherapy should not be part of the treatment plan for clients with bipolar illness. If these therapies are undertaken at all, they should be reserved for persons who have been stabilized for some time, and it should be understood by both client and therapist that the therapy is not being performed. In clients with bipolar disorder, psychotherapy is never a substitute for appropriate medical treatment.

Psychotherapy for Bipolar Depression

There have been no trials of psychotherapy for bipolar depression. For selected persons, formal psychotherapies e.g., cognitive or interpersonal, developed for unipolar (nonbipolar) depression may be helpful.

C. Ensure Normal Thyroid Function

Although an association between hypothyroidism and rapid cycling states has been shown (reviewed in Bauer 1994), it is unclear to what extent abnormal thyroid function may contribute to the pathogenesis of bipolar depression and mania. However, thyroid function appears to play a role in the modulation of mood (Bauer & Whybrow 1988). In addition, the clinical manifestations of abnormal thyroid function often mimic those of mania or depression. Thus, thyroid function testing appears warranted in clients who present with signs and symptoms of bipolar depression or mania. If the thyroid stimulating hormone (TSH) level is found to be persistently elevated, L-thyroxine may be administered with a goal of decreasing serum basal TSH level to within normal range. In cases of minor elevations in TSH level, the decision to treat should be based upon the client’s presentation and physician’s clinical judgment. Although some clients may experience resolution of symptoms with L-thyroxine replacement, most will require additional treatment (i.e., antidepressants, mood-stabilizers) for mania or depressive symptoms as well. Note that the goal of thyroid replacement is different from that of high-dose thyroid hormone administration, which is a treatment for refractory rapid cycling.

Treatment with lithium may cause iatrogenic hypothyroidism in some clients. In such cases, it is not necessary to discontinue lithium treatment; correction of thyroid hypofunction with L-thyroxine is usually sufficient. Of note, the clinician should be aware that treatment with various medications may cause asymptomatic changes in thyroid hormone levels. In particular, both carbamazepine and tricyclic antidepressants may lower levels of thyroxine and triiodothyronine.

D. Reduce/Stop Antidepressant

Antidepressants have been shown to induce a "switch" in mood to a manic or hypomanic state. If the person is in a manic or a prolonged (more than a week) hypomanic state, the antidepressant should be stopped immediately. If the hypomanic state is less than one week in duration, it is not unusual for hypomania to remit spontaneously. Sometimes titration of the antidepressant to a lower dose will lead to the disappearance of the hypomanic symptoms.

E. Response

Clinically, "response" is defined as a significant reduction in symptoms or in symptom severity. Usually, this involves decrease by more than 50 percent on rating scale scores or in symptoms.

F. Institute Prophylaxis and Consider Psychosocial Rehabilitation

See Bipolar/ Schizoaffective Disorder Core Module D, annotation F.

G. Insomnia, Agitation, or Anxiety?

Insomnia is difficulty falling asleep, staying asleep, or waking too early, accompanied by a subjective feeling of not being rested after sleep.

Agitation is defined as disruptive behavior over which the person has poor control. This may take the form of verbal or physical aggression, increased activity, excessive talking, or intrusiveness. The agitated person may respond briefly to admonitions, but cannot control his or her impulsive behavior for any extended period of time. While agitation may accompany many other psychiatric disorders, it is a hallmark of the manic state.

H. Consider Benzodiazepine Treatment

Use of benzodiazepines in conjunction with mood stabilizers is an effective treatment for the manic phase of bipolar disorder. Lorazepam is recommended because of its availability in intramuscular injectable form, its short half-life with no active metabolites, and its favorable side effect profile, and it may be more effective than clonazepam for acute mania (Bradwejn et al. 1990). The major side effects associated with benzodiazepines in high doses are ataxia, hypotension, and sedation. For hospitalized persons with severe, uncontrollable agitation, lorazepam may be given every 1 to 2 hours, PO or IM, 1 to 2 mg, until the person is mildly sedated, with cumulative doses up to 20 mg per 24 hours. Doses of lorazepam can then be decreased during the next 1 to 5 days. For moderate agitation or in ambulatory settings, lower doses should be used, between 0.5 and 2 mg, BID to QID. Clients should be cautioned about driving or using machinery and instructed to avoid alcohol. For other benzodiazepines, including diazepam, there are clinical studies documenting efficacy in psychotic agitation (Lerner et al. 1979).

Clients may be maintained on a benzodiazepine for several weeks until the mood stabilizer has been optimized. Benzodiazepines should not be stopped abruptly, since rebound anxiety may occur and risk of seizure is present with sudden discontinuation. An appropriate tapering schedule involves gradual decrease of the daily dose in steps of 20 to 25 percent over the course of 1 to 4 weeks. Adverse effects of benzodiazepines include amnesia, ataxia, sedation, and poor coordination. Sexual dysfunction has occurred in 50 percent of cases when benzodiazepines are combined with lithium (Ghadirian et al. 1992).

I. Psychotic Features Present?

Psychotic features are delusions, hallucinations (typically auditory), bizarre behavior, or thought disorganization. Usually the content of the delusions or hallucinations in mania is consistent with manic themes, such as grandiosity or paranoia, and they are mood-congruent psychotic features. Sometimes the content of the hallucinations or delusions has no apparent relationship to manic themes, and they are mood-incongruent psychotic features (DSM-IV). A client with bipolar depression may also demonstrate mood-congruent psychotic features.

J. Initiate and Optimize Mood Stabilizer; Assess in 2 to 3 Weeks?

Mood stabilizers are defined as agents that decrease the intensity and/or frequency of manic, hypomanic, depressive, or mixed episodes in clients with bipolar disorder without causing a switch to a different mood episode (modified from Bowden 1996). Lithium, valproate, and carbamazepine and electroconvulsive therapy (ECT) are the currently accepted mood stabilizers. The choice of mood stabilizer for a particular person depends on the clinical presentation, urgency of treatment, history of response, medical status, and risk of adverse effects. In general, the person should be maintained on a mood stabilizer for 3 weeks after a therapeutic serum level is achieved prior to assessment of response.

Lithium: Lithium, available as carbonate salts and citrate syrup, has been the mainstay of bipolar disorder treatment for over three decades. Lithium’s efficacy in treating acute mania is supported by 10 uncontrolled trials and four placebo-controlled studies. Lithium has also been found to be superior to neuroleptics in the treatment of acute mania (Janicak et al. 1993). In one double-blind, randomized, controlled comparison study with valproate, lithium was slightly more efficacious than valproate (Freeman et al. 1992). In a double-blind, randomized, controlled comparison with divalproex and placebo, lithium was superior to placebo and similar in efficacy to divalproex (Bowden et al. 1994). Lithium was also found to be superior to carbamazepine in one double-blind comparison (Lerer et al. 1987) and statistically equivalent to carbamazepine in the acute phase and over a longer term (Small et al. 1991). It is important to note that both the Small study comparing lithium with carbamazepine and the Bowden article comparing lithium with valproate included a large number of clients described as being unresponsive to lithium in the past. It has been suggested that lithium may be more efficacious than valproate in manic clients without concurrent depressive features and less efficacious in manic clients with concurrent depressive features (mixed mania) (Freeman et al. 1992; Swann et al. 1997). A study specifically addressing these questions is needed. First-line use of lithium as an acute antimanic is also supported by its established efficacy in the prophylaxis of episodes of mood change (Goodwin & Jamison 1990); thus, there is support for using the same agent for both acute and maintenance phase treatment. Comparable evidence is not yet available for the other mood stabilizers.

Before starting treatment with lithium, it is recommended that the following be obtained: general medical history and physical examination; complete blood count (CBC), blood urea nitrogen (BUN) and creatinine levels, a thyroid function test, electrocardiogram (EKG) for clients over 40, and a pregnancy test, if applicable. Lithium treatment should be initiated in low divided doses (e.g., 300 mg TID or 600 mg BID) and titrated upward according to response and side effects. Alternatively, nomograms are available to predict required dosages based on serum levels following a single dose test (Cooper et al. 1973). The serum trough level of lithium (obtained 12 hours after last dose) needs to be monitored regularly and frequently in treating acute manic episodes and should be maintained between 0.5 and 1.5 mmol/L with nonresponding clients needing higher levels (Stokes et al. 1976).

Valproate: The anticonvulsant valproate, available as valproic acid, sodium valproate, and divalproex sodium, has been found to be effective in treating mania. In one double-blind comparison, valproate was superior to placebo in the treatment of acute mania (Pope et al. 1991). Valproate was slightly less effective than lithium in treating acute mania in a double-blind randomized comparison (Freeman et al. 1992); in a double-blind, placebo-controlled study it was superior to placebo and similar in efficacy to lithium (Bowden et al. 1994). Valproate may be particularly useful in treating manic episodes in clients who failed to respond to lithium. All clients entering the Pope study (1991) and a subset of persons in a Bowden study (1994) who responded to valproate were previously classified as not responding to lithium. It is suggested that valproate may be more efficacious than lithium in manic persons with concurrent depressive features (mixed manics) and less efficacious in manic persons without concurrent depressive features (Freeman et al. 1992; Swann et al. 1997). There are no published data on controlled trials of valproate in the maintenance phase, but published open trials indicate efficacy.

Before starting treatment with valproate, it is recommended that hepatic and hematologic function be evaluated. There have been reports of hepatic failure in persons receiving valproate; however, all these reports are from the pediatric age group and most of them are of clients receiving multiple anticonvulsants for the treatment of epilepsy. Valproate can be initiated in low divided doses (e.g., 250 mg TID) and titrated upward according to response and side effects. An oral loading strategy of administering valproate at 20 mg/kg/day has been reported to be efficacious in the initial treatment of acute psychotic mania in a randomized, single-blind comparison with haloperidol (McElroy et al. 1996). Serum valproate levels between 45 and 125 micrograms per ml are much more likely to be efficacious and well tolerated (Bowden et al. 1996).

Carbamazepine: The anticonvulsant carbamazepine has been found to be effective in acute mania (Goodwin and Jamison 1990; Janicak et al. 1993). Carbamazepine was found to be less efficacious than lithium in one double-blind comparison (Lerer et al. 1987) and to be statistically equivalent to lithium in the acute phase with a slight advantage over lithium over a longer term (Small et al. 1991). There is some evidence that this drug is useful for manic persons with neurologic dysfunction. There is limited evidence on the efficacy of carbamazepine in the maintenance phase (Prien & Gelenberg 1989).

Before starting treatment with carbamazepine, the baseline evaluation should include a CBC with differential and platelet count, a liver profile, renal function tests, and serum electrolyte values. Carbamazepine can be started at a daily dosage of 200 to 600 mg in divided doses and titrated upward according to response and side effects up to 800 to 1,000 mg a day. Some clients may need higher doses, but doses above 1,600 mg a day are not recommended (Hirschfeld et al. 1996). Therapeutic serum levels of carbamazepine for bipolar disorder are not established, but the therapeutic serum levels established for seizure disorder (4 to 15 mcg/ml) are used in the treatment of mania. CBC, platelet counts, and hepatic function tests should be monitored every two weeks during the first two months of the treatment and, if normal, every three months thereafter. Clients should be educated regarding hepatic, hematologic, and dermatologic reactions and to report if these symptoms occur.

Serum levels of mood stabilizing medications should be monitored. For treatment of acute mania, therapeutic levels are:

To achieve optimal response, dosage may need to be adjusted to reach a higher therapeutic range. Generally, a 3 week trial at a higher serum level, tolerated without serious side effects, is needed to determine the efficacy of a mood stabilizer. Lack of significant response at the end of 2 weeks may necessitate adjustment of the dosage or consideration of adding or switching to another mood-stabilizer (see annotation O).

ECT: Electroconvulsive therapy (ECT) is a rapid, effective treatment for acute mania (Mukherjee et al. 1994). In controlled prospective studies, ECT was found to be equal to or more effective than medications (Small et al. 1988; Mukherjee et al. 1988). ECT is reserved for persons who need a rapid clinical response, are unable to tolerate medication (including pregnant clients), fail to respond to medication, or prefer this modality of treatment. ECT is considered the treatment of choice for persons with catatonic features (Gelenberg and Hopkins 1976) and manic delirium (Mann et al. 1990). ECT has a good chance of success in persons who failed with other treatments (Mukherjee et al. 1988). ECT can be used as a maintenance treatment if there is reason not to use the mood stabilizing medications.

K. Add or Change Mood Stabilizers or Consider Alternative Therapy Until Stable

If there is minimal or no response after 3 weeks of treatment with a mood stabilizer at a therapeutic serum level, a different mood stabilizer should be started and the first agent tapered (Kahn et al. 1996; Hirschfeld et al. 1996).

If there is only a partial response (reduction of symptoms by less than 50 percent) after 3 weeks of treatment at therapeutic serum levels of the mood stabilizer, a second mood stabilizer can be added. When the person is on lithium, either valproate or carbamazepine can be added. If the client is on one of the latter two medications, lithium can be added (Kahn et al. 1996; Hirschfeld et al. 1996). If there is a history of partial or nonresponse to single mood stabilizers, combination treatment may be helpful (Kahn et al. 1996; Hirschfeld et al. 1996).

Retrospective and prospective studies have suggested that clozapine has mood-stabilizing properties as well as antipsychotic effects (McElroy et al. 1991; Suppes et al. 1992). There exist preliminary reports on the efficacy of drugs used for mood stabilization in combination with the anticonvulsant drugs lamotrigine and gabapentin (Weisler et al. 1994; Calabrese et al. 1996). Other agents that have been reported to have mood-stabilizing properties in the management of bipolar disorder include thyroid hormones (Stancer and Persad 1982; Stein and Anvi 1988), benzodiazepines (Chouinard 1987; Sachs et al. 1990) and calcium channel blockers (Dubovsky et al. 1986).

View Flow Chart for Module E.

Go to Module D.

Go to Module F.