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

Evidence-Based Reviews

Treatment-resistant schizophrenia: What can we do about it?

Use pharmacotherapy and other interventions to target the symptoms that matter most to your patient

Vol. 10, No. 06 / June 2011

Dr. Citrome: How to best help patients with residual schizophrenia symptoms

Discuss this article at

Patients with treatment-resistant schizophrenia can be broadly defined to include any persons with residual symptoms that cause distress or impairment despite several treatment attempts. Unfortunately, this definition may include most of our patients with schizophrenia.

Clinical trial data on treatment-resistant schizophrenia can be contradictory, leaving “N of 1” empirical treatment trials for individual patients as the current state of the art. This article presents data from clinical trials for pharmacologic and nonpharmacologic options and offers recommendations to try to help our treatment-resistant patients.

Defining treatment resistance

Research reports regarding treatment-resistant or treatment-refractory schizophrenia have relied on operational criteria such as that found in the pivotal study for clozapine1:

  • at least 3 periods of treatment in the preceding 5 years with neuroleptic agents from at least 2 different chemical classes at dosages equivalent to ≥1000 mg/d of chlorpromazine for 6 weeks, each without significant symptomatic relief, and
  • no period of good functioning within the preceding 5 years.1 In that study, patients also underwent a prospective treatment trial with what we now know are high doses of haloperidol (up to 60 mg/d or higher) and benztropine mesylate (6 mg/d) for a period of 6 weeks to confirm lack of drug responsiveness. Other studies have more relaxed criteria, such as:
  • persistent positive symptoms—hallucinations, delusions, or marked thought disorder—after at least 6 contiguous weeks of past or present treatment, with ≥1 typical antipsychotics at doses of ≥600 mg/d in chlorpromazine equivalents
  • a poor level of functioning over the past 2 years, as defined by the lack of competitive employment or enrollment in an academic or vocational program and not having age-expected interpersonal relations with someone outside the biologic family with whom ongoing regular contacts were maintained.2

In this study, no prospective period of treatment to confirm lack of drug responsiveness was required.

The most clinically relevant definition of treatment resistance depends on the patient’s individual circumstances. For some patients, targeting positive symptoms is a high priority; for others it may be negative and cognitive symptoms; for others, it may be excitement. Moreover, families may complain of symptoms or behavior that are of little or no concern to your patient.

Although we desire treatment response and remission for our patients, definitions for remission and functional recovery are in flux. Proposed criteria define symptomatic remission as 6-month maintenance of simultaneous ratings of mild or less on delusions, hallucinations, disorganized speech, grossly disorganized or catatonic behavior, and negative symptoms.3,4 Emsley et al4 note that reported remission rates vary widely across studies (17% to 88%) and that patients in remission do better than their non-remitted counterparts in several other outcome domains. Also, patients move in and out of remission over time. Predictors of remission include:

  • early treatment response
  • baseline symptom severity
  • subjective well-being.4

Recovery is a more complex construct than remission and includes social outcomes. Although recovery lacks a standard definition, it is the implied goal of treatment. Anything short of recovery can be viewed as inadequate. If we set the bar at this height, many or most of the patients we treat for schizophrenia could be considered treatment-resistant.

Confounding factors

Before concluding that a patient is treatment-resistant, address medication adherence and possible substance use. Partial or nonadherence with antipsychotic treatment is common—approximately one-half of patients are nonadherent5—and associated with relapse and re-hospitalization.6 In addition, an estimated one-half of all individuals with schizophrenia also use substances.7

Be aware of the optimal dose for any particular antipsychotic and factors that can interfere with achieving adequate plasma levels. This means acknowledging that dosing ranges established during registration studies may not reflect the needs of day-to-day clinical practice.8 Pharmacokinetic interactions with other medications, such as carbamazepine or rifampin, can induce liver enzymes and result in subtherapeutic antipsychotic levels. Cigarette smoking also may have this effect. Lowered clozapine or olanzapine plasma levels have been observed in patients who resume smoking after being discharged from a non-smoking inpatient environment. Some antipsychotics, such as ziprasidone and lurasidone, must be taken with food in order to have sufficient bioavailability.9

What does a patient want?

Patients with schizophrenia often have limited insight into their psychotic symptoms.10 Savvy clinicians will attempt to leverage a patient’s insight into ancillary symptoms—such as impaired sleep, anxiety, and dysphoria—to encourage a therapeutic alliance and therefore adherence. If patients feel their concerns are not addressed, they may consider treatment inadequate even though the intensity of their hallucinations and delusions may have decreased.

Which antipsychotic is best?

Meta-analyses of randomized controlled trials (RCTs) of antipsychotic treatment for schizophrenia found that, although individual response will vary, clozapine generally has better efficacy that other antipsychotics.11-13 Olanzapine, risperidone, and amisulpride (which is not available in the United States) appear to be more efficacious than first-generation antipsychotics. Other second-generation antipsychotics do not consistently show greater efficacy than first-generation antipsychotics, although their tolerability profiles vary greatly.11-13

Antipsychotic monotherapy. More than 25 RCTs have focused on antipsychotic monotherapy for treatment-resistant patients; for a bibliography of these studies, click here. For the most part, clozapine has consistently demonstrated superiority over comparators. Because not all patients with schizophrenia can tolerate clozapine or are willing to have their blood monitored as required, other second-generation antipsychotics have been suggested as possible substitutes. Olanzapine has established superior efficacy to first-generation antipsychotics11-13 and perhaps comparable efficacy to clozapine in some studies.2,14-17 Risperidone appeared to be comparable to clozapine in some studies,18,19 whereas clozapine’s superiority was evident in others.14,20,21 Although an RCT found comparable efficacy for ziprasidone vs clozapine,22 patients enrolled in this study may not have been treatment-resistant regarding efficacy but instead could not tolerate prior treatments. Enrolling patients on the basis of poor efficacy and/or poor tolerability to their prior antipsychotic regimen also has complicated the interpretation of studies comparing olanzapine with clozapine16 and risperidone with clozapine.18

Antipsychotic combinations. Combinations of antipsychotics are used commonly when treating chronic schizophrenia.23 Of the approximately 20 RCTs of antipsychotic combination therapy, most tested clozapine combined with other second-generation antipsychotics, such as risperidone. For a bibliography of these studies, click here. Only 5 studies support a combination approach (Table 1).

Table 1

Antipsychotic combinations: Few studies support efficacy





Shiloh et al, 1997a

10-week, double-blind, placebo-controlled

28 patients nonresponsive to typical antipsychotics and partially responsive to clozapine received add-on sulpiride,* 600 mg/d, or placebo

The sulpiride group showed improvements in positive and negative symptoms

Josiassen et al, 2005b

12-week, randomized, double-blind, placebo-controlled

40 schizophrenia patients unresponsive or partially responsive to clozapine randomized to clozapine + placebo or clozapine + risperidone, 6 mg/d

Mean BPRS total and positive symptom subscale scores reduced in both groups but reductions were greater in the clozapine/risperidone group; reduction in SANS also was observed in the clozapine/risperidone group

Genç et al, 2007c

8-week, randomized, single-blind

56 treatment-resistant schizophrenia patients randomly assigned to clozapine + amisulpride* or clozapine + quetiapine

Both groups improved at week 8 as measured by BPRS, SANS, SAPS, and CGI; however, patients receiving amisulpride showed greater improvement

Muscatello et al, 2011d

24-week, randomized, double-blind, placebo-controlled

31 treatment-resistant schizophrenia patients receiving clozapine randomized to receive adjunctive aripiprazole or placebo

Aripiprazole showed beneficial effect on positive and general psychopathologic symptomatology, but no significant effects on executive cognitive function

Takahashi et al, 1999e

8-week, randomized, single-blind, crossover

10 neuroleptic-treated patients received add-on risperidone and mosapramine*

Both additions resulted in significant, yet modest, improvement; no significant difference in PANSS between risperidone and mosapramine

*Not available in the United States
BPRS: Brief Psychiatric Rating Scale; CGI: Clinical Global Impression; PANSS: Positive and Negative Syndrome Scale; SANS: Scale for the Assessment of Negative Symptoms; SAPS: Scale for the Assessment of Positive Symptoms

a. Shiloh R, Zemishlany Z, Aizenberg D, et al. Sulpiride augmentation in people with schizophrenia partially responsive to clozapine. A double-blind, placebo-controlled study. Br J Psychiatry. 1997;171:569-573.
b. Josiassen RC, Joseph A, Kohegyi E, et al. Clozapine augmented with risperidone in the treatment of schizophrenia: a randomized, double-blind, placebo-controlled trial. Am J Psychiatry. 2005;162(1):130-136.
c. Genç Y, Taner E, Candansayar S. Comparison of clozapine-amisulpride and clozapine-quetiapine combinations for patients with schizophrenia who are partially responsive to clozapine: a single-blind randomized study. Adv Ther. 2007;24(1):1-13.
d. Muscatello MR, Bruno A, Pandolfo G, et al. Effect of aripiprazole augmentation of clozapine in schizophrenia: a double-blind, placebo-controlled study. Schizophr Res. 2011;127(1-3):93-99.
e. Takahashi N, Terao T, Oga T, et al. Comparison of risperidone and mosapramine addition to neuroleptic treatment in chronic schizophrenia. Neuropsychobiology. 1999;39(2):81-85.

What about augmentation?

Adjunctive non-antipsychotics also are commonly used when treating patients with chronic schizophrenia. For example, lithium and anticonvulsants are used in approximately one-half of all inpatients with schizophrenia in facilities operated by the State of New York Office of Mental Health.24,25 The evidence base for these agents as adjuncts to antipsychotics generally is weak.26 Specifically, early reports of benefit with adjunctive lithium have been negated by later studies. Similarly, large trials of adjunctive valproate and lamotrigine have failed to replicate early and promising efficacy signals from smaller trials, although the larger studies did not specifically target treatment-resistant schizophrenia.

Among mood stabilizers, lamotrigine may be the most promising for treatment-resistant schizophrenia. In a meta-analysis of clinical trials examining schizophrenia patients receiving clozapine (N=161) who were randomized to receive adjunctive lamotrigine or adjunctive placebo, lamotrigine was superior to placebo in total score for psychosis symptoms and scores for positive and negative symptoms.27

More than 125 published RCTs have studied a wide variety of adjunctive agents other than lithium or anticonvulsants for treating persistent schizophrenia symptoms (Table 2).

Only some of the approximately 40 RCTs regarding adjunctive antidepressants in patients with chronic schizophrenia focused on patients with ongoing depressive symptoms. For a bibliography of these studies, click here. In a meta-analysis measuring improvement of negative symptoms from 23 trials (N=819),28 the effect size was moderate in favor of antidepressants. Subgroup analysis revealed significant responses for fluoxetine, trazodone, and ritanserin.

More than 50 RCTs have focused on augmenting medications for cognitive dysfunction in chronic schizophrenia. Unfortunately, agents used to treat Alzheimer’s disease have shown disappointing results when tested in patients with schizophrenia, as have agents prescribed for attention-deficit/hyperactivity disorder (methylphenidate, guanfacine, atomoxetine) or agents used to promote alertness (modafinil and armodafinil).

Medications that act on glutamate receptors may offer another potential solution, although not in combination with clozapine.29

Other agents that require further study where ≥2 positive studies have been reported (with ≤2 negative studies) include celecoxib, neurosteroids and hormones, purinergic agents, serotonin 5-HT1A receptor agonists, and serotonin 5-HT3 receptor antagonists.

Table 2

Agents studied as adjuncts to antipsychotics

Acetylsalicylic acid and nonsteroidal anti-inflammatory agents

Anticonvulsants and lithium



Agents used to treat attention-deficit/hyperactivity disorder

Beta blockers

Cholinesterase inhibitors and other agents used to treat Alzheimer’s disease

Experimental agents that act on glutamate receptors

GABAA receptor drugs

Neurosteroids and hormones

Omega-3 fatty acids

Opioid system agents


Purinergic agents

Serotonin 5-HT1A receptor agonists

Serotonin 5-HT3 receptor antagonists

Wakefulness promoting agents

Therapeutic neuromodulation

More than 10 RCTs of repetitive transcranial magnetic stimulation (rTMS) in patients with refractory symptoms of schizophrenia have been published; the results were mixed. For a bibliography of these studies, click here. In a meta-analysis of 9 trials (n=213),30 prefrontal rTMS for treating negative symptoms demonstrated a small-to-medium effect size. In another meta-analysis31 of all prospective studies of rTMS for negative symptoms and for auditory hallucinations and overall positive symptoms in refractory schizophrenia, the effect sizes showed moderate effects.

Fewer controlled trials are available for electroconvulsive therapy,32,33 but its use with clozapine appears encouraging.34

Psychological and behavioral intervention. Cognitive-behavioral therapy, although labor-intensive, can be helpful even in patients considered treatment-resistant (Table 3). These interventions generally are provided together with pharmacotherapy.

Did you miss this content?
Psychosis and catatonia after dancing with a dangerous partner