To Name :
To Email :
From Name :
From Email :
Comments :

Evidence-Based Reviews

Antipsychotics and mood disorders: A complicated alliance

Neuroleptics have played an important but controversial role in mood disorder treatment. Novel antipsychotic agents may represent a significant departure from their earlier counterparts, providing clinicians with an alternative strategy for mood stabilization.

Vol. 1, No. 1 / January 2002

Major mood disorders are challenging to diagnose and often difficult to treat. They entail unipolar depression; bipolar disorder, which includes manic, depressed, or mixed episodes; and schizoaffective disorder, which includes both depressed and bipolar subtypes. Antidepressants and mood stabilizers are the primary pharmacological treatments. They may be insufficient, however, for patients with more severe episodes, often characterized by psychosis and treatment resistance.

In these patients, antipsychotics have played an important but controversial part in management, primarily as oral or parenteral adjuncts. Literature and clinical experience now support another, unique role for the current generation of novel agents.

Compared to earlier antipsychotics, these agents produce substantially fewer neurological adverse effects, including acute extrapyramidal and tardive syndromes, and can augment antidepressants and mood stabilizers. In addition, they may:

  • Possess a better antipsychotic profile, with enhanced therapeutic effects on positive, negative, cognitive, and mood symptoms
  • Have a role in the acute and long-term management of these disorders when anticipated parenteral formulations become available (e.g., acute intramuscular olanzapine and ziprasidone—and long-acting intramuscular risperidone)
  • Possess inherent thymoleptic properties (see “Unresolved issues with antipsychotics,” below).


  1. Defining what constitutes a mood stabilizer.1 Proposed definitions suggest that the drug must entail the following:
  2. Clarifying the mechanisms underlying the apparent mood-regulating effects of novel agents
  3. Ascertaining both acute and maintenance efficacy
  4. Clarifying the propensity of some agents to switch depressed patients into mania
  5. Increasing the number of well-designed studies with sufficient sample sizes, including comparison trials assessing the relative efficacy of different novel agents
  6. Reducing the tendency to publish only positive reports when new drugs are first available
  7. Introducing parenteral formulations of novel agents
  8. Resolving concerns about weight gain, new-onset diabetes, QT c prolongation, and sedation
  9. Rectifying the current level of substantially greater costs

Management of unipolar depression

Neuroleptics Delusions and hallucinations indicate a more severe form of depressive disorder, with poor short- and long-term outcomes in comparison to those without psychosis. To illustrate, Table 1 lists a summary of response rates in psychotic and nonpsychotic depressed patients given a tricyclic antidepressant (TCA). The data indicate that patients suffering from psychotic depression typically do not benefit from antidepressant monotherapy and usually require a combination of antidepressant and antipsychotic or, alternatively, electroconvulsive therapy (ECT).

There is, however, some limited clinical and neuroimaging evidence that amoxapine can be used as an effective monotherapy in this group. Amoxapine is an antidepressant whose primary active metabolite, 8-hydroxy amoxapine, may have antipsychotic properties.1 With the possible exception of amoxapine, combined antipsychotic-antidepressant treatment is the rule.

Table 1

Psychotic and nonpsychotic depressed patients’ response to monotherapy with a tricyclic antidepressant






Responders (%) (n=127)

Nonresponders (%) (n=236)

Responders (%) (n=464)

Nonresponders (%) (n=227)


13 studies






Adapted from Chan CH, Janicak PG, Davis JM, et al. Response of psychotic and nonpsychotic depressed patients to tricyclic antidepressants. J Clin Psychiatry. 1987;48:197-200.

Historically, studies have also evaluated neuroleptic monotherapy for depressed patients. While some reported superiority over a placebo, none found conventional antipsychotics superior to imipramine. Indeed, patients with schizophrenia who are treated with a neuroleptic often develop symptoms that are difficult to distinguish from depression (e.g., secondary negative symptoms). These often improve when the neuroleptic is discontinued or the patients are switched to a novel antipsychotic such as risperidone, olanzapine, or ziprasidone, all of which have putative antidepressant effects.

When employing an antipsychotic in depressed patients, the dosage and duration of treatment are two critical considerations. To minimize neuromotor adverse effects, use low doses of a neuroleptic (e.g., haloperidol, 1 to 5 mg/d) in conjunction with the primary antidepressant therapy. The neuroleptic should then be tapered gradually after psychotic symptoms have been controlled, usually during the acute phase of treatment. Ideally, patients would then take antidepressant monotherapy through the continuation phase and, if necessary, the maintenance phase of treatment. If psychosis recurs, re-introduce the antipsychotic intermittently.

Novel antipsychotics In contrast to neuroleptics, novel antipsychotics have been reported to improve depression in various psychotic and mood disorders.

For example, ziprasidone has serotonin and noradrenergic reuptake blocking effects comparable to such classic TCAs as imipramine and amitriptyline, as well as high binding affinity at the 5-HT1A, 5-HT1D, and 5-HT2C receptors. This neuroreceptor profile indicates possible antidepressant effects.

While randomized, controlled trials with mood-disordered patients are few, there have been promising preliminary reports of augmentation of antidepressants with risperidone and olanzapine in both psychotic and nonpsychotic depressed patients.

Ostroff and Nelson2 reported the results of an open-label study of eight SSRI-nonresponsive patients (mean treatment 7.3 weeks). These patients had no psychotic features and had a dramatic reduction in depressive symptoms, as well as some improvement in sexual dysfunction, with the addition of 0.5 mg to 1.0 mg risperidone. The clinicians suggested that risperidone’s 5-HT2A antagonism might explain its augmentation of the partial SSRI response.

Olanzapine alone (n=3) or combined with an antidepressant (n=12) has also been reported to improve both depression and psychosis.3 In a double-blind, amitriptyline-controlled trial, Svestka and Synek4 found that olanzapine demonstrated antidepressant efficacy in 33 unipolar and seven bipolar depressed patients. Thirteen of these patients also had psychotic symptoms.

Shelton et al5 reported the results of a two-center, 8-week, double-blind comparison of olanzapine alone, fluoxetine alone, or their combination in 28 patients suffering from treatment-resistant, non-bipolar disorder without psychosis. They found that the combination was superior to either drug alone based on improvement in the Hamilton Depression Rating Scale (HDRS) total score. From their preliminary data, it also appears that the doses required were relatively low, reducing the risk of side effects.

Their findings, however, need to be replicated in more controlled studies with combinations, addressing possible adverse effects, the potential for clinically relevant drug interactions, decreased compliance rates, and increased cost of treatment. Earlier reports raised concern about the potential of these agents to increase switching to hypomania or mania. But in more recent reports, this has not emerged as a significant problem. 7

Finally, several case reports and case series indicate that agents such as clozapine and risperidone may augment ECT in particularly severe, treatment-resistant depressive episodes.7

Management of bipolar and schizoaffective depressed episodes

Neuroleptics Antipsychotics are frequently used to manage more severe, usually psychotic episodes of bipolar and schizoaffective depression. Reports indicate that affectively ill patients receiving neuroleptics may be more prone to develop neuromotor adverse effects than are those suffering from schizophrenia. Thus, their use for such patients must be well justified, limited in dosage and duration, and carefully monitored for the emergence of acute and tardive neurological events.

Novel antipsychotics Novel antipsychotics have demonstrated fewer propensities than have neuroleptics in worsening depression or negative symptoms in schizophrenic patients, and have possible antidepressant effects. In support of this hypothesis, and reminiscent of data from earlier risperidone and olanzapine trials, ziprasidone was observed to improve the Montgomery Asberg Rating Scale (MADRS) and Brief Psychotic Rating Scale (BPRS) depressive cluster scores in three clinical trials with schizophrenic and schizoaffective patients.8,9

Vieta et al reported the efficacy and safety of risperidone add-on therapy for treating various episodes of bipolar (n=358) and schizoaffective (n=183) disorders.6 In this multicenter, open study, 33 patients (6.1%) suffered a depressed episode and received a mean risperidone dose of 1.6 (± 2.3) mg/d added to their ongoing but ineffective drug regimen. Mean HDRS declined significantly over the 6-month course. Further, switch rates were low and in the expected range for spontaneous fluctuations seen in these disorders.

The results of a 6-week, double-blind, controlled trial of risperidone versus haloperidol in 62 patients with schizoaffective disorder, bipolar or depressed subtype, were published.10 Risperidone (average dose of 5.5 mg/d) was comparable to haloperidol (average dose of 10.8 mg/d) in reducing the mean in the Positive and Negative Syndrome Scale and Clinician-Administered Rating Scale for Mania change scores.

In those patients with baseline HDRS scores ≥ 20, risperidone produced a significantly greater reduction in mean change scores than did haloperidol. In addition, patients had no mood switches with risperidone or haloperidol; there was a significantly higher incidence of patients who had extra-pyramidal symptoms with haloperidol than among those taking risperidone; and six patients in the group taking haloperidol dropped out after experiencing adverse effects. None of the patients taking risperidone dropped out.

Table 2

Lithium versus antipsychotics for acute mania






Responders (%) (n=64)

Nonresponders (%) (n=10)

Responders (%) (n=38)

Nonresponders (%) (n=33)


5 studies






Adapted from Janicak PG, Newman RH, Davis JM. Advances in the treatment of mania and related disorders: a reappraisal. Psychiatric Ann. 1992;22(2):94.

Management of bipolar manic or mixed episodes

Up to 80% of all bipolar patients receive an antipsychotic drug during the acute and/or maintenance phase of their illness, even though loading doses of valproate and benzodiazepines may also be used during an exacerbation and pose much less risk, especially in terms of adverse neurological effects.

Neuroleptics Shortly after their introduction, neuroleptics were found to reduce mortality secondary to dehydration and exhaustion in many highly agitated patients during an acute manic episode such as lethal catatonia.7

While earlier controlled studies found these agents to be effective in the treatment of acute mania, they are clearly less efficacious than lithium for core manic symptoms.11 Table 2 demonstrates a meta-analysis of five well-controlled, double-blind studies documenting the statistical superiority of lithium over neuroleptics. These agents, however, offer the advantage of a more rapid onset of action, particularly when given in the acute parenteral formulation, and are superior to lithium in the initial control of agitation. Further, long-acting depot formulations of neuroleptics may be the only viable strategy for chronic, recurrent, noncompliant patients.

As with psychotic depression, dosing and duration of neuroleptic treatment are important concerns. In this context, Rifkin et al demonstrated that 10 mg of haloperidol per day had comparable efficacy but fewer adverse effects than did 30 or 80 mg per day in a group of acutely manic patients.12 Despite such data, high chlorpromazine-equivalent doses are often administered acutely and maintained for sustained periods. This can be a significant problem given the apparent great sensitivity of bipolar patients to the neurological sequelae of these antipsychotic agents.

Novel antipsychotics Early case series reports indicated that clozapine may benefit treatment-refractory bipolar patients. Given the inherent drawbacks of clozapine (e.g., agranulocytosis and seizure induction), attention now focuses on other novel agents with more benign adverse effect profiles than clopazine. Controlled trials with olanzapine and risperidone serve to reinforce the usefulness of these as well as other novel agents.

Tohen et al published the results of a 3-week, double-blind, placebo-controlled trial of olanzapine in 139 patients experiencing an acute bipolar manic or mixed episode.13 Olanzapine produced a statistically greater mean improvement than did the placebo on the Young Mania Rating Scale (YMRS) change scores. Further, 49% of the olanzapine-treated group (n=70) met the a priori criteria for response versus only 24% of the placebo-treated group (n=69). A second study using a higher starting dose of olanzapine, less rescue medication, and longer treatment duration than the first study resulted in a similar outcome.14

Sachs et al reported on the results of a 3-week, double-blind, placebo-controlled trial involving 156 patients with bipolar manic or mixed subtype who received a mood stabilizer (lithium or valproate) plus a placebo, risperidone (1 to 6 mg/d), or haloperidol (2 to 12 mg/d).15 The clinicians concluded that risperidone plus a mood stabilizer was statistically superior to a placebo plus a mood stabilizer, and produced more rapid reduction in manic symptoms, regardless of whether psychosis was present.

Sajatovic et al16 published the results of a prospective, open trial with quetiapine (mean dose = 203 ± 124 mg/d) as add-on therapy in 20 patients (10 bipolar, 10 schizoaffective; 19 male, 1 female) insufficiently responsive to their mood stabilizer or antipsychotic. Pre-post assessments indicated significant improvement in the BPRS, Mania Rating Scale (MRS), and HDRS scores. While the combination was generally well tolerated, there was a mean weight gain of 4.9 kg (10.8 lb). This raises the specter of complications associated with substantial weight gain produced by several of the novel antipsychotics.

Did you miss this content?
Prescriber’s guide to using 3 new antidepressants: Vilazodone, levomilnacipran, vortioxetine