Bipolar Disorders 2014 © 2014 John Wiley & Sons A/S

Published by John Wiley & Sons Ltd.

BIPOLAR DISORDERS

Review Article

Clozapine for treatment-resistant bipolar

disorder: a systematic review
Li X-B, Tang Y-L, Wang C-Y, de Leon J. Clozapine for treatment- Xian-Bin Lia,b, Yi-Lang
resistant bipolar disorder: a systematic review. Tanga,c, Chuan-Yue Wanga,b
Bipolar Disord 2014: 00: 000–000. © 2014 John Wiley & Sons A/S. and Jose de Leond,e,f
Published by John Wiley & Sons Ltd. aBeijing Key Laboratory of Mental Disorders,
Department of Psychiatry, Beijing Anding
Objective: To evaluate the efficacy and safety of clozapine for treatment- Hospital, Capital Medical University, bCenter of
resistant bipolar disorder (TRBD). Schizophrenia, Beijing Institute for Brain
Disorders, Laboratory of Brain Disorders (Capital
Methods: A systematic review of randomized controlled studies, open-label Medical University), Ministry of Science and
prospective studies, and retrospective studies of patients with TRBD was Technology, Beijing, China, cDepartment of
carried out. Interventions included clozapine monotherapy or clozapine Psychiatry and Behavioral Sciences, Emory
combined with other medications. Outcome measures were efficacy and University School of Medicine, Atlanta, GA,
adverse drug reactions (ADRs). dMental Health Research Center at Eastern State

Hospital, University of Kentucky, Lexington, KY,

Results: Fifteen clinical trials with a total sample of 1,044 patients met the USA, ePsychiatry and Neurosciences Research
inclusion criteria. Clozapine monotherapy or clozapine combined with other Group (CTS-549), Institute of Neurosciences,
treatments for TRBD was associated with improvement in: (i) symptoms of University of Granada, Granada, fBiomedical
mania, depression, rapid cycling, and psychotic symptoms, with many Research Centre in Mental Health Net
patients with TRBD achieving a remission or response; (ii) the number and (CIBERSAM), Santiago Apostol Hospital,
duration of hospitalizations, the number of psychotropic co-medications, and University of the Basque Country, Vitoria, Spain
the number of hospital visits for somatic reasons for intentional self-
harm/overdose; (iii) suicidal ideation and aggressive behavior; and (iv) doi: 10.1111/bdi.12272
social functioning. In addition, patients with TRBD showed greater clinical
improvement in long-term follow-up when compared with published Key words: bipolar disorder –
schizophrenia data. Sedation (12%), constipation (5.0%), sialorrhea (5.2%), clozapine – treatment-resistant
weight gain (4%), and body ache/pain (2%) were the commonly reported
ADRs; however, these symptoms but did not usually require drug Received 11 February 2014, revised and
discontinuation. The percentage of severe ADRs reported, such as accepted for publication 11 August 2014
leukopenia (2%), agranulocytosis (0.3%), and seizure (0.5%), appeared to be
lower than those reported in the published schizophrenia literature. Corresponding author:
Chuan-Yue Wang, M.D., Ph.D.
Beijing Anding Hospital,
Conclusion: The limited current evidence supports the concept that Capital Medical University
clozapine may be both an effective and a relatively safe medication for
No. 5 Ankang Lane
TRBD.
Dewai Avenue, Xicheng District
Beijing 100088
China
Fax: +86-10-58303195
E-mail: [email protected]

Clozapine, an atypical antipsychotic, is primarily randomized clinical trials (RCTs) (5, 6) and a recent
used for the treatment of treatment-resistant review of effectiveness trials (7) supported the greater
schizophrenia in most parts of the world (1, 2). Long- efficacy of clozapine among antipsy-chotics in
term use of clozapine is associated with improvement schizophrenia.
in clinical symptoms, measurable social and A growing number of reports, however, suggest
functional gains, and decreased hospi-talization as that clozapine may also have a role in other treat-
compared with typical antipsychotic agents (3, 4). ment-resistant psychotic conditions (8–10), such as
Furthermore, meta-analyses of schizoaffective disorder and psychotic mood
Li et al.

disorders (11–13). Furthermore, case reports and with clozapine are commonly a factor discouraging
retrospective studies suggest that clozapine may be clinicians from prescribing it.
particularly effective in the treatment of medica-tion-
resistant unipolar depression and bipolar dis-order
Methods
(BD); some even suggested it is more effective than it
is for schizophrenia (12, 14–16). Before we conducted this systematic review, our
Compared with unipolar depression, BD is a more protocol of reviewing clozapine use for TRBD was
serious type of mood disorder. BD is a recur-rent, published online (http://www.crd.york.ac.uk/
potentially disabling, sometimes even fatal prospero/); the registration number was
psychiatric illness (17–19), and the estimated life- CRD42013004322 at the Preferred Reporting Items
time prevalence of various types of BD is over 2.0% for Systematic Reviews and Meta-Analyses
(20, 21). BD is often associated with high lev-els of (PRISMA). PRISMA provides an evidence-based
unfavorable outcomes or treatment resis-tance (22– minimum set of items for reporting in systematic
24). In contrast to schizophrenia, definitions of reviews and meta-analyses (50).
treatment-resistant bipolar disorder (TRBD) vary
greatly (17, 25–27). However, a fail-ure to respond to Types of studies
at least two trials of dissimilar treatments, involving
an adequate dose and dura-tion, could serve as a All types of trials evaluating the efficacy and safety
conservative definition (28–31). of clozapine for TRBD were eligible for inclusion.
We included RCTs (Table 1), open-label retrospec-
Although mood disorder was traditionally con- tive studies (Table 2), and prospective trials (Table
sidered a rather rare condition in China, recently 3). We excluded meta-analyses and system-atic
conducted epidemiological studies in the country reviews. We also excluded from the compre-hensive
showed that it is one of the common mental disor- review case series and reports, since they offer a
ders (32, 33), with a one-month prevalence of 6.1% lower level of evidence, and are associated with a
(32). Unlike other countries, clozapine has been high suspicion of publication biases. How-ever, we
widely used for BD in China despite not having been have included them in Table 4 and pro-vided a brief
approved for mood disorders (34–36), and it is statement on them for the sake of entirety. The
indeed one of the most commonly used antipsy- retrospective open study by Nielsen et al. (51) was
chotics in the treatment of BD (34, 37, 38). Some included in this review (Table 2), although the
psychiatrists even preferred it as a first-line treat- sample also included patients with non-TRBD; it was
ment for mania (38–40). Similar to findings from not possible to exclude them. All tables provided
studies in Western countries, RCTs showed that details of the contamination of the studies by other
clozapine was an effective add-on treatment to an- diagnoses when it was not pos-sible to separate the
tidepressants for treatment-resistant depression (41). patients.
Clozapine was also effective for treatment-resistant
mania in a case report (42) and an RCT (43). Study selection

Clozapine is a drug of choice for TRBD in China We searched PubMed, Embase, and Cochrane
but the evidence for its use in Western coun-tries Library databases and the Cochrane Controlled Trials
remains sparse, and the studies are limited to case Register of clozapine for TRBD. We also searched
reports (44), open-label trials (11), and only one RCT the Chinese databases [the Chinese Bio-medical
with fewer than 20 patients in each group (45). As Literature and China National Knowledge
China has the largest population on cloza-pine (46– Infrastructure databases] using the same keywords.
48), the Chinese experience and studies may be of The search included all studies published between
keen interest to Western psychiatrists (49). So far, no January 1979 and June 2014, regardless of lan-guage.
exhaustive systematic review on clozapine for TRBD The keywords used for the searches included:
has been published. clozapine, bipolar disorder, manic, depres-sion,
The primary aim of this review was to evaluate the resistant/resistance/refractory, drug therapy, and trial.
efficacy and safety of clozapine for TRBD. As The keywords were used in combination with the
previously mentioned, in addition to international Boolean operators AND, OR, and NOT. We
databases, we also included Chinese databases that supplemented the search by using the ‘related article’
are not usually reviewed in articles written by function. We also manually searched bibli-ographies
Western psychiatrists. Particular attention was paid to of RCTs, meta-analyses, and systematic reviews for
safety and tolerability, as the potentially severe studies that were missed in the initial electronic
adverse drug reactions (ADRs) associated search (52).
 Clozapine for treatment-resistant bipolar disorder

Table 1. Clozapine randomized controlled trials for treatment-resistant bipolar disorder (TRBD)
Suppes et al. 1999 (45)
Population: 38 patients meeting the DSM-IV criteria for BD (n = 26) or SAD (n = 12) who were deemed treatment-resistant [failure
of adequate treatment with two mood stabilizers (lithium, valproate, or carbamazepine) at standard therapeutic levels]. Subjects
were randomly assigned to clozapine add-on treatment (n = 19) or TAU (no clozapine) (n = 19)
Intervention: Clozapine (355 mg/day) add-on therapy
Comparison: TAU
Measures: Patients received monthly ratings on the BPRS, CGI, BRMS, HDRS, SAPS, SANS and AIMS, and a 40-item side-
effect checklist
Study design: Randomized, TAU-controlled study with follow-up of one year
Resultsa: Significant between-group differences were found in scores on all rating scales except the HDRS. Total medication use over
one year significantly decreased in the clozapine group. No significant differences in physical complaints between groups were noted
Tan 2010 (43)
Population: 71 patients with DSM-IV BD who were classified as having TRBD were randomly assigned to clozapine added to
lithium treatment (n = 35) or clozapine added to valproate treatment (n = 36). Treatment resistance was defined as failure of
adequate treatment with two different antidepressants
Intervention: Clozapine (100–300 mg/day) added to lithium (500–1,500 mg/day)
Comparisons: Clozapine (100–300 mg/day) added to valproate (600–1,800 mg/day)
Measures: Patients received ratings on BPRS, HDRS, and TESS at Weeks 0, 1, 2, 4, and 6
Study design: Randomized, open-controlled study
Resultsa: In the study group, 89% of patients were responders (based on BPRS and HDRS) to clozapine added to lithium compared
with 64% of patients receiving clozapine added to valproate (p < 0.05). No significant differences in adverse drug reactions between
the groups were found

AIMS = Abnormal Involuntary Movement Scale; BPRS = Brief Psychiatric Rating Scale; BRMS = Bech–Rafaelsen Mania Scale; CGI =
Clinical Global Impression; DSM-IV = Diagnostic and Statistical Manual of Mental Disorders, 4th edition; HDRS = Hamilton Depression
Rating Scale; SAD = schizoaffective disorder; SANS = Scale for the Assessment of Negative Symptoms; SAPS = Scale for the
Assessment of Positive Symptoms; TAU = treatment-as-usual; TESS = Treatment Emergent Symptom Scale.
aClinical remission and response were defined differently in each study.

One author (X-BL) independently inspected full report was acquired for more detailed scrutiny.
citations from the searches and identified relevant Full reports of the abstracts meeting the review
abstracts. A random 20% of the samples were criteria were obtained and inspected by X-BL. Again,
independently re-inspected by author Y-LT to ensure a random 20% of reports were re-inspected by Y-LT
reliability. When disagreements arose, the in order to ensure reliable selection.

Table 2. Clozapine retrospective studies for treatment-resistant bipolar disorder (TRBD)

McElroy et al. 1991 (12)
Sample: All patients were either inadequately responsive to or unable to tolerate standard biological therapies
Methods: Survey of treating clinicians and chart data for all 85 consecutive patients, including 39 with schizophrenia, 25 with SAD,
and 14 with psychotic BD, who received clozapine for at least six weeks at one center
Results: Compared to patients with schizophrenia, patients with SAD and psychotic BD had significantly higher response rates to
clozapine (10% for schizophrenia versus 15–20% for SAD and 43% for psychotic BD)
Chang et al. 2006 (58)
Sample: Patients with BD resistant to conventional treatment
Methods: Analysis of clinical data from medical records of 51 patients with DSM-IV BD treated with add-on clozapine for >6 months
Results: The number of hospital days/year was reduced in 90% of patients after clozapine add-on treatment. The total number and
duration of hospitalizations/year also decreased. Significant reductions were found in the number and duration of hospitalizations
associated with manic, depressive, and hypomanic episodes. Long-term efficacy of clozapine add-on was supported by continuous
decreases in hospital days/year in the 27 selected patients
Nielsen et al. 2012 (51)
Sample: A total of 21,473 patients with a lifetime diagnosis of ICD-10 BD, of whom only 326 (1.5%) were treated with clozapine
and were included in a mirror-image analysis
Methods: A pharmacy-epidemiologic database study was carried out in Denmark, investigating the effectiveness of clozapine in
patients with BD (without a schizophrenia-spectrum disorder), between 1996 and 2007, using a two-year mirror-image design
Results: Clozapine appeared to be an appropriate choice for TRBD. Compared to the pre-clozapine period, during clozapine
treatment, the mean number of bed-days decreased from 179 to 35. The mean number of admissions was reduced from 3.2 to 2.0.
Overall, 240 patients (74%) had reduced bed-days and 130 (40%) were not admitted while treated with clozapine. Moreover, the
number of psychotropic co-medications was reduced from 4.5 DDD (25–75 percentiles: 2.4–8.2) to 3.9 DDD (25–75 percentiles:
2.4–6.1). The percentage of patients with hospital visits for intentional self-harm/overdose was reduced significantly from 8% to 3%

BD = bipolar disorder; DDD = defined daily doses; DSM-IV = Diagnostic and Statistical Manual of Mental Disorders, 4th edition; ICD-10
= International Classification of Diseases, 10th edition; SAD = schizoaffective disorder.
4

Li et al.
Table 3. Clozapine open-label trials for treatment-resistant bipolar disorder

Resistance definition Treatments Duration

Study Subjects (failure of) (mg/day) (months) Main findings

Suppes et al. 7 with dysphoric Standard CLZ (50–500) + 36–60 Symptomatic and functional improvement was assessed. Most of the
1992 (16) mania treatments ACs patients sustained substantial gains in psychosocial function in follow-
including ACs up over 3 years. No further hospitalizations were needed in 6 of
7 patients
Banov et al. 52 BD, 81 SAD, Undefined CLZ 18.7 BD manic and SAD bipolar patients had significantly better outcomes
1994 (11) 14 UD, 40 SCH than UD, BD, and SAD depressed patients. BD and SAD patients had
significantly greater improvement in social functioning than SCH patients.
One or more episodes of depression prior to CLZ predicted CLZ
Discontinuation
Kimmel et al. 25 manic with Li, ACs, and CLZ 13 18 of 25 patients demonstrated a > 50% decrease in the YMRS
1994 (62) BD or SAD ≥2 APs, or intolerant
Kowatch et al. 5 children or Multiple trials of APs CLZ (75–225) + 2 There was a 42% decrease in the CGI. Treatment was for aggressive
1995 (65) adolescents and ACs, or intolerance Li behavior and psychotic symptoms
with BD
Zarate et al. 17 mood Combinations of Li, CLZ 16.1 65% (11/17) had no subsequent re-hospitalization or mood episode.
1995 (61) disorders ACs, APs, and ECT; Significant improvement in CGI scores
or had tardive dyskinesia
Calabrese et al. 25 acutely manic Li, ACs, and APs, CLZ (494) 4 72% (18/25) improved on the YMRS and 32% (8/25) improved on the BPRS.
1996 (44) intolerable ADRs, or both The patients with BD as compared to the patients with SAD, and the non-
rapid cycling patients as compared to rapid cyclers, had significantly
greater improvement in total BPRS score
Green et al. 22 active manic 500 mg/day of chlorpromazine CLZ 3 57% (13/22) improved on the BPRS, 57% (12/22) on the YMRS, and 39%
2000 (63) or its equivalent and (8/22) on the CGI, and 77% (17/22) experienced at least a 20%
Li of at least 6 weeks reduction on all three scales
Ciapparelli et al. 34 psychotic BD, Adequate treatment with CLZ flexible 24 All patients showed significant improvement 24 months from intake (based
2000 (59) 31 SCH, 26 SAD, 3 different classes of APs doses on BPRS and CGI). The presence of suicidal ideation at intake predicted
bipolar type greater improvement at endpoint
Ciapparelli et al. 37 psychotic BD, Adequate treatment with CLZ flexible 48 Patients with SAD and BD show greater clinical improvement than those
2003 (60) 34 SCH, 30 SAD, 3 different classes of APs doses with SCH. Patients with BD had the shortest time to response and the
bipolar type highest psychosocial and occupational functioning levels (based
on BPRS, CGI, and GAF)
Fehr et al. 9 BD 2 ACs + APs CLZ (156 77) 12 Three patients demonstrated striking mood stabilization and returned to
2005 (64) previous levels of functioning (based on BPRS and HDRS). Five patients
had moderate improvement in mood stabilization and functioning (based
on BPRS and HDRS), and one patient showed a minimal response

AC = anticonvulsant; ADR = adverse drug reaction; AP = antipsychotic; BD = bipolar disorder; BPRS = Brief Psychiatric Rating Scale; CGI = Clinical Global Impression; CLZ = clozapine;
ECT = electroconvulsive therapy; GAF = Global Assessment of Functioning; HDRS = Hamilton Depression Rating Scale; Li = lithium; SAD = schizoaffective disorder; SCH = schizophrenia;
UD = unipolar depression; YMRS = Young Mania Rating Scale.
 Clozapine for treatment-resistant bipolar disorder

Table 4. Clozapine case series and reports for treatment-resistant bipolar disorder

Resistance
definition Treatments Duration
Study Subjects Age, gender (failure of) (mg/day) (months) Main findings

Calabrese et al. 2 RC BD 47 years, F AC + AP CLZ (250–350) 1.5–3.5 Remission

1991 (74) 48 years, F
Suppes et al. 3 RC BD 43 years, F AP + AC CLZ (150–400) + Li 12–20 2 remission,
1994 (15) 25 years, F 1 response
45 years, F
Antonacci & 4 euphoric Unavailable Standard CLZ _ Enhanced functioning
Swartz mania treatments + AC and insight
1995 (70)
Poyurovsky & 2 mania 24 years, M AP + AC CLZ (250–350) _ Remission
Weizman 41 years, F + ECT
1996 (71)
Lancon & Llorca 1 RC BD 42 years, F Conventional CLZ _ Successfully treated
1996 (75) therapy
Mahmood et al. 3 mania Unavailable AP + AC CLZ _ Successfully treated
1997 (72)
Chanpattana 1 mania 26 years, M Conventional CLZ (200) + ECT 18 Complete remission
2000 (73) treatment
Xu 2003 (42) 1 mania 40 years, F AP + AC CLZ (600) + 1 Remarkably effective
Li (1,500) +
CBZ (600)
Chen et al. 1 RC BD 38 years, F Various biological CLZ (350) + 36 Complete remission
2005 (76) therapies TPR (300)
Vijay Sagar 1 juvenile- 18 years, F AC combinations CLZ (200) 24 Remission
2005 (79) onset BD
Quante et al. 3 BD, 2 UD Not provided Medications + CLZ (125 and 375) 12 4/5 patients showed
2007 (119) ECT steady improvement
Gupta 1 BD 28 years, M Standard AC CLZ (350) 66 No hospitalization and
2009 (78) no more episodes
Bastiampillai 1 RC BD 52 years, F Standard APs + CLZ (150) + 60 Sustained remission
et al. 2010 (77) ACs LTG (100)
Bennedetti et al. 7 SAD and Three patients, mean Treatment Aripiprazole (6.8) + – Remission
2010 (29) psychotic BD 36 years, M resistant CLZ (293)
Four patients, mean
40 years, F

AC = anticonvulsant; AP = antipsychotic; BD = bipolar disorder; CBZ = carbamazepine; CLZ = clozapine; ECT = electroconvulsive
therapy; F = female; Li = lithium; LTG = lamotrigine; M = male; RC = rapid cycling; SAD = schizoaffective disorder; TPR = topiramate;
UD = unipolar depression.

Where it was not possible to resolve disagreement by the total. Again, any disagreement was discussed,
discussion, a third author (C-YW) mediated the decisions were documented, and, if necessary,
decision. If the matter was unresolved, an attempt authors of studies were contacted for clarification.
was made to contact the authors of the original study Data presented only in graphs and figures were
for clarification (53). extracted whenever possible, but included only if two
authors independently had the same result. We also
Data extraction attempted to contact authors through an open-ended
request in order to obtain missing information or for
Review authors X-BL and Y-LT considered all clarification whenever deemed necessary. If studies
included studies initially, without seeing compari-son were multi-center, we extracted data relevant to each
data, to judge clinical, methodological and sta-tistical component center separately (53).
heterogeneity and thereby decide whether each study
would be included for meta-analysis or other data
synthesis. We then extracted data into standard,
Assessment of reporting biases
simple forms. X-BL extracted data from all included
studies. In addition, to ensure reliabil-ity, Y-LT Reporting biases arise when the dissemination of
independently extracted data from a ran-dom sample research findings is influenced by the nature and
of these studies, comprising 30% of direction of results. We tried to locate the research
Li et al.

protocols of included RCTs. If the protocol was rospective studies (Table 2), and 10 open-label
available, outcomes in the protocol and in the prospective trials (Table 3). These studies were
published report were compared. If the protocol was equally distributed across the years between 1991 and
not available, outcomes listed in the methods section 2012, which indicates that clozapine for TRBD has
of the trial report were compared with the actually been a rather long-lasting, clinically important topic
reported results (54). for the last 25 years.
It was not possible to conduct a meta-analysis
Grading recommendations
because of the study’s heterogeneity, including dif-
ferences in illness phase (mania, depression, or rapid
We used the grading of recommendations assess- cycling BD), methodology (open-label trial or RCT)
ment, development, and evaluation (GRADE) sys- and outcome definition (response or remission).
tem to rate the quality of evidence and strength of Although meta-analysis is a powerful tool for
recommendations of this systematic review follow- analyzing data (55), confounding inter-study
ing the guidelines of the Cochrane Collaboration. variables that cannot be controlled may vio-late basic
GRADE included systematic assessments of all statistical assumptions, making this type of analysis
included trials across six main domains for each error-prone (56, 57). Therefore, we only extracted
outcome: limitations of the study design and exe- data onto standard, simple forms on a case-by-case
cution, inconsistency, indirectness, imprecision of basis and reported the efficacy of clozapine for TRBD
results, publication bias, and large treatment effect. when available, as well as other descriptive statistics.
Accordingly, we graded the recommendation for the Compared with effi-cacy, there was less
outcome measure of clozapine for BD as very low, heterogeneity in ADRs. There-fore, we conducted
low, moderate, or high (Table 5). data synthesis using this term, and all trials with ADR
details were included; the percentage of each ADR
was computed in this review and is presented in
Results
Table 6.
The various combinations of the search ‘clozapine, We were unable to locate the protocols for three
bipolar disorder, manic, depression, resistant or RCTs; therefore, we assessed the reporting bias by
resistance, refractory’ yielded 342 articles, of which means of comparing outcomes listed in the meth-ods
15 studies met the criteria. In total, 1,044 patients with the results, which indicated that the reported
with TRBD had received clozapine treatment (Fig. results were approximately consistent with outcomes
1). There were two RCTs (Table 1); three ret- listed in the methods.

Table 5. Grading of recommendations assessment, development, and evaluation system (GRADE) analysis: quality assessment of
clozapine for treatment-resistant bipolar disorder

Overall
Participants Risk of Large quality
Critical outcome (studies) bias Inconsistency Indirectness Imprecision Public bias effect of evidenced

Open studies
CGI score 236 (5) Seriousa No No No Undetected No Very low
BPRS score 248 (5) No No No No Undetected No Low
YMRS score 47 (2) No No No No Undetected No Low
Social functioning 304 (4) Seriousb No No No Undetected No Very low
Hospital days/year 377 (2) No No No No Undetected No Low
Mean no. of admissions 377 (2) No No No No Undetected No Low
Randomized clinical trials
BPRS score 109 (2) Noc No No No Undetected No Moderate
HDRS score 109 (2) Noc No No No Undetected No Moderate

BPRS = Brief Psychiatric Rating Scale; CGI = Clinical Global Impression; HDRS = Hamilton Depression Rating Scale; YMRS = Young
Mania Rating Scale.
aIncomplete accounting of patients and outcome events.
b
Relatively few patients (n ≤ 10).
cLack of allocation concealment.
dThe quality of evidence was rated using the GRADE Working Group system. High quality indicates that further research is very
unlikely to change our confidence in the estimate of effect but none of the studies reached that level. Moderate quality indicates that
further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low
quality indi-cates that further research is very likely to have an important impact on our confidence in the estimate of effect and is likely
to change the estimate. Very low quality indicates that we are very uncertain about the estimate.
 Clozapine for treatment-resistant bipolar disorder

Table 6. Clozapine adverse drug reactions in treatment-

resistant bipolar disorder

n (% of
Adverse drug reaction 797 total)

Blood cells Leukopenia 14 (1.7)

Decreases in white 6 (0.8)
blood cell count
Agranulocytosis 2 (0.3)
Metabolic system Weight gain 31 (4.0)
Weight loss 1 (0.1)
Hyperlipidemia 1 (0.1)
Increased appetite 1 (0.1)
Diabetes type 2 1 (0.1)
Endocrine system Sialorrhea 42 (5.2)
Sweating 4 (0.5)
Dry mouth 1 (0.1)
Influenza-like syndrome 1 (0.1)
Cardiovascular Abnormal EEG 6 (0.8)
system Orthostatic hypertension 6 (0.8)
Tachycardia 6 (0.8)
Orthostatic hypotension 2 (0.2)
Digestive system Constipation 40 (5.0)
Diarrhea 8 (1.1)
Nausea/vomiting 5 (0.6)
Postprandial regurgitation 1 (0.1)
Ileus 1 (0.1)
Nervous system Sedation 98 (12.2)
Body ache and pain 15 (1.8)
Dizziness 11 (1.4)
Sleep cycle inversion 7 (1.1)
Fig. 1. Preferred Reporting Items for Systematic Reviews and Transient fever 8 (0.9)
Meta-Analyses (PRISMA) flow diagram. Urinary incontinence 6 (0.8)
Seizure 4 (0.5)
Quality assessment of the included studies based Tremors 2 (0.2)
on the GRADE approach showed many limitations of Fatigue 2 (0.2)
Neuroleptic malignant 1 (0.1)
the study designs, no obvious indi-rectness,
syndrome
imprecision in result reporting and large treatment Bradykinesia 1 (0.1)
effect. Based on the above assessments, the quality of Enuresis 1 (0.1)
evidence presented for each outcome ranged from Mental confusion 1 (0.1)
‘very low’ to ‘moderate’ (Table 5).

Clozapine RCTs for TRBD with schizophrenia and psychotic BD indicated that
the latter had significantly higher response rates to
Our literature search yielded two clozapine RCTs for clozapine (12).
TRBD (Table 1). In these RCTs, adjunctive clozapine
treatment was superior to treatment as usual for
TRBD (45). In addition, clozapine with added lithium Clozapine open-label prospective trials for TRBD
was better than clozapine aug-mented with valproate The 10 clozapine open-label prospective studies for
in rapid cycling BD (43). TRBD (Table 3) included five long-term follow-up
studies (11, 16, 59–61), four focused on mania (44,
Clozapine retrospective trials for TRBD 62–64) and one focused on adolescent patients (65).
The studies found that patients on clozapine
Three clozapine retrospective trials for TRBD were demonstrated a significant decrease in the Young
identified (Table 2). Two trials described the num-ber Mania Rating Scale (YMRS), Hamilton Depres-sion
and duration of hospitalizations, the number of Rating Scale (HDRS), Brief Psychiatric Rat-ing Scale
psychotropic co-medications and the number of (BPRS), and Clinical Global Impression Scale (CGI)
hospital visits for medical reasons and for inten-tional scores (44, 66); the presence of suicidal ideation and
self-harm/overdose as significantly reduced during aggressive behavior at intake pre-dicted greater
clozapine treatment (12, 29, 51, 58). Another improvement at endpoint (59, 65) and improvement
retrospective study comparing patients in social functioning (16). In
Li et al.

addition, they also found that BD patients showed HDRS, BPRS, and CGI scores, evidenced
greater clinical improvement than those with improvement in social functioning, suicidal idea-tion
schizophrenia in the long-term follow-up (11, 60). and aggressive behavior, and had fewer subse-quent
affective episodes; furthermore, patients with TRBD
showed greater clinical improvement than those with
Clozapine ADRs in TRBD
schizophrenia in the long-term follow-up in these
ADRs are summarized in Table 6. The prevalences of trials. The current review also suggests that clozapine
the most serious ADRs were leukopenia, 2%; may have anti-manic properties in some children and
agranulocytosis, 0.3%; and seizure, 0.5%. There were adolescents with TRBD.
no cases of myocarditis. The most frequent clinically In general, this review found that clozapine for
significant ADRs were sedation (12%), constipation TRBD was safe and well tolerated (Table 6). Seda-
(5%), sialorrhea (5%), weight gain (4%), and any tion, constipation, sialorrhea, weight gain, and pain
kind of pain (2%). Other ADRs with a frequency of were the common ADRs, which is consistent with
0.5–1.0% were dizziness, diar-rhea (1%), transient schizophrenia studies (80, 81), and they were rather
fever, urinary incontinence, abnormal EEG, mild and tolerable to most patients. Moder-ate ADRs
tachycardia, orthostatic hyperten-sion, and nausea. included dizziness, diarrhea, transient fever, urinary
Other ADRs are described in Table 6. incontinence, abnormal EEG, tachy-cardia,
orthostatic hypertension, and nausea. Rare ADRs
were sweating, hyperlipidemia, diabetes type 2,
influenza-like syndrome, postprandial regurgitation,
Discussion
ileus, and bradykinesia, which is also comparable
This is the first systematic review of clozapine for with schizophrenia studies (80, 82). These ADRs
TRBD summarizing its efficacy and safety. Our were not severe enough to result in drug
comprehensive systematic review included 15 stud- discontinuation. The ADRs in the metabolic system
ies with a total of 1,044 patients and suggests that were obviously low; there is the possibility of a major
clozapine may be an effective therapy, safe and well underreport in the included trials.
tolerated. Although we excluded all of the case series Among all the reports, 17 patients had leukope-nia
and case reports in this comprehensive review, the (2%), two had agranulocytosis (0.2%), and five had
literature search provided 13 case reports/series in seizures (0.5%). These figures tend to be lower than
which clozapine was used for TRBD (Table 4). The averages reported in schizophrenia reviews (83–89).
13 articles included five on mania (42, 70–73), five We are not sure whether the lower ADR frequency in
on rapid cycling BD (15, 74–77), and three on other BD versus schizophrenia trials is real or an artefact.
TRBDs (29, 78, 79). Overall, almost all cases were Greater underreport and different methodologies may
treatment-resistant and had a remission after contribute to an artificially low ADR frequency.
switching to clozapine monotherapy or adding Some clozapine ADRs are dose-related; others are
clozapine to other drugs. not. Doses in BD trials appear lower than doses in
Western schizophrenia studies, but it was not possible
Strengths of the study to control doses for con-founders such as smoking
[which induced cloza-pine metabolism probably by
While many patients with BD respond well to con- inducing the cytochrome P450 1A2 (CYP1A2)] and
ventional medications (including antidepressants, racial differ-ences (see the discussion in the section
mood stabilizers and antipsychotics), a substantial ‘Limitations of the study’ below) (64, 87, 89). The
proportion do not achieve a satisfactory response agranulocyto-sis risk is still a concern for clinicians,
(67–69). This systematic review showed that cloza- but manda-tory blood monitoring has been shown to
pine may be an efficacious therapy for TRBD. First, considerably reduce the incidence of fully devel-oped
RCTs showed that: (i) clozapine add-on treatment cases of agranulocytosis (80). Thus, appropri-ate
was superior to treatment as usual in mania, and (ii) management of clozapine ADRs facilitates
clozapine plus lithium was better than clozapine plus maximization of the benefits of clozapine treat-ment,
valproate in rapid cycling BD. Secondly, and physicians and patients alike should be aware
retrospective studies of clozapine for TRBD indicated that there are a range of benefits to cloza-pine use
that the total number and dura-tion of hospitalizations that outweigh its risk (80, 90, 91).
and the number of psycho-tropic co-medications were
significantly reduced during clozapine treatment. Clozapine treatment was associated with signifi-
Thirdly, in open-label prospective studies, patients cant improvement in tardive dyskinesia in seven
treated with clozapine demonstrated a significant patients (92–94); clozapine may be useful for long-
decrease in the YMRS, term treatment to lower tardive dyskinesia risk (93,
 Clozapine for treatment-resistant bipolar disorder

95, 96). Furthermore, once tardive dyskinesia or also need to be mentioned: (i) all available trials were
dystonia is established, clozapine may be useful for included, without applying any language restrictions;
both control of the movement disorder and BD (93, and (ii) the efficacy case-by-case analysis and
96). computation of the percentage of each ADR provided
The main strength of this study is that we also some evidence supporting the use of clozapine (Table
searched Chinese databases in the systematic review, 6).
which included all TRBD clozapine trials conducted Thirdly, some trials were ‘contaminated’ by some
in China, where clozapine is widely used. Thus, it is patients with a diagnosis of schizoaffective disorder
the first review to include all trials available without or schizophrenia, which does not corre-spond to the
applying any language restric-tions. We found RCTs population described as the target population of
of clozapine monotherapy or clozapine combined interest (those with TRBD). In some trials, we could
with other medications ver-sus other treatments in not separate patients with TRBD from the patients
patients with TRBD; the comparison treatments with schizoaffective disorder or schizophrenia, but
included a mood stabilizer (97, 98), other the tables provide details of these ‘contaminated’
antipsychotics (99–102), clozapine plus a mood studies.
stabilizer versus a mood stabilizer (97, 103, 104), and A fourth limitation of this review and all the
clozapine plus a mood stabilizer ver-sus other studies reviewed in it is the lack of close attention to
antipsychotics plus a mood stabilizer (103–109) in issues regarding clozapine pharmacokinetics and
the treatment of BD in China. We also found two dosing. Clozapine dosing is influenced by racial
RCTs, one open-label prospective study, and two differences, drug–drug interactions and smoking. In
case reports on the use of clozapine for TRBD in the 1997, it was already reported that Chi-nese patients
Chinese literature, including the only placebo- tended to receive approximately half of the clozapine
controlled clozapine RCT for TRMD (41). dosage used in Western counties (110, 111) but
appeared to have roughly similar clozapine levels,
which is indicative of lower cloza-pine metabolism in
Limitations of the study Chinese patients. The literature has not stressed this
difference nor provided an explanation. Sirot et al.
A few limitations of the current review need to be (112) carried out a very important study that has not
acknowledged. First, it was not possible to conduct a received enough attention in the literature. They
meta-analysis because of the study’s heterogene-ity, described cyto-chrome P450 2C19 (CYP2C19) poor
including differences in illness phase (mania, metabolizers (PMs) as having 2.3-fold higher plasma
depression or rapid cycling BD), methodology (open- clozapine concentrations than patients with other
label trials or RCTs) and outcome definition CYP2C19 genotypes. Approximately 25% of the
(response or remission). This great heterogeneity may Chinese population are CYP2C19 PMs.
violate basic statistical assumptions and make these
analyses error-prone. Therefore, we only extracted Studies of adjunctive clozapine treatment in TRBD
data onto standard, simple forms on a case-by-case usually ignore the major differences in clozapine
basis and reported the efficacy of clozapine for TRBD metabolism associated with co-med-ication.
when available. Compared with data on efficacy, Carbamazepine is a major inducer of clozapine
there was lower heterogene-ity with ADR data, and metabolism (113) and it is possible that valproate
therefore data synthesis using ADRs was conducted may be a mild inducer (114). Fluvox-amine is a
and the percentage of each ADR was computed major inhibitor of clozapine metabolism
(Table 6). (115) and paroxetine and fluoxetine are mild inhib-
Secondly, most of the clinical trials included here itors (114, 115).
had major methodological problems. Although this In conclusion, future studies and meta-analyses of
review included 15 clinical trials, most of them were clozapine for TRBD will need to pay attention to
open-label observational trials; only two RCTs were important pharmacokinetic differences associ-ated
available. There were no obvious reporting biases in with racial differences, co-medication and smoking,
the RCTs, but reporting biases in other studies are which may have major influences on clozapine
possible. Furthermore, the GRADE approach showed dosing but at present are ignored in most published
that the quality of the evidence was ‘very low’ in CGI articles.
score and psycho-social function; other outcomes
were from ‘low’ to ‘moderate’ (Table 5). Therefore,
the current review provided limited evidence Comparison with other studies
supporting clozapine use. However, two strengths of Poon et al. (17) performed a literature review on
the current review TRBD research findings. It provided few promising
Li et al.

leads other than the use of clozapine for TRBD This comprehensive review has focused on TRBD,
mania, which is comparable to our analysis. Their but future reviews need to focus on the role of
review was limited by: (i) inclusion of only two clozapine for the treatment of BD in general. Though
clozapine trials (44, 45), (ii) lack of report on cloza- clozapine is rarely used for non-treatment-resistant
pine ADRs, and (iii) lack of inclusion of Chinese BD elsewhere in the world, emerging evidence from
studies which include larger numbers of patients. China is encour-aging (41, 97, 98). Since the early
Gitlin (116) conducted a review on this topic. That 1980s, few clozapine RCTs for BD in general have
review indicated that combining multiple agents was been published; once more clozapine RCTs have
the most commonly used clinical strat-egy for TRBD; been published, a meta-analysis of non-treatment
an approach that may be effective for treatment- resistant BD, if supportive of clozapine use, will
resistant patients included high-dose thyroid provide clinicians with more choices.
augmentation, clozapine, calcium channel blockers,
and electroconvulsive therapy, which is consistent
with our findings. However, only three studies of Acknowledgements
clozapine treatment were included (45, 60, 64), and
none of the Chinese studies were included. Similarly, The authors thank Lorraine Maw, M.A., for editorial assis-tance.
This study was supported by the Beijing Science and Technology
there was no safety analysis or data synthesis. Commission (grant D121100005012002). No com-mercial
organizations had any role in the writing of this paper for
Frye et al. (117) reviewed the use of atypical publication.
antipsychotics in the treatment of BD, focusing on
clozapine as the prototypical agent. That review Disclosures
indicated that the early clinical experience of
clozapine as a potential mood stabilizer sug-gested The authors of this paper report no financial relationship with
commercial interests within the last three years.
greater anti-manic than antidepressant properties.
However, that review only included some trials
conducted before 1998, which excluded most trials of Author contributions
clozapine for TRBD. Similarly, there was no safety X-BL and Y-LT contributed equally to the review of the arti-cles
analysis or data synthesis in that analysis. and to the writing of the first draft. C-YW and JdL con-tributed by
improving the first and later drafts. All authors contributed to and
approved the final manuscript.

Conclusions
References
TRBD is a complex, often severe and disabling 1. Kane J, Honigfeld G, Singer J, Meltzer H. Clozapine for the
psychiatric disorder and it often poses a thera-peutic treatment-resistant schizophrenic: a double-blind
challenge (17, 118). This systematic review shows comparison with chlorpromazine. Arch Gen Psychiatry
that clozapine monotherapy or its combi-nation with 1988; 45: 789–796.
other medications for TRBD may be both safe and 2. Marder SR, Van Putten T. Who should receive cloza-pine?
Arch Gen Psychiatry 1988; 45: 865–867.
effective. Long-term use of cloza-pine appeared to be 3. Meltzer HY. Dimensions of outcome with clozapine. Br J
associated with improvement in clinical symptoms, Psychiatry Suppl 1992; 17: 46–53.
measurable social and functional gains, and 4. Meltzer HY, Okayli G. Reduction of suicidality during
decreased hospitalization. Constipation, sedation, clozapine treatment of neuroleptic-resistant schizophre-
nia: impact on risk-benefit assessment. Am J Psychiatry
sialorrhea, weight gain, aches and pain were the most 1995; 152: 183–190.
commonly reported ADRs, though none were severe 5. Asenjo Lobos C, Komossa K, Rummel-Kluge C et al.
enough to result in drug discontinuation. The percent- Clozapine versus other atypical antipsychotics for schizo-
ages of patients with leukocytopenia, agranulocy- phrenia. Cochrane Library 2010; CD006633.
tosis and seizure were lower than in studies of 6. Leucht S, Cipriani A, Spineli L et al. Comparative effi-cacy
and tolerability of 15 antipsychotic drugs in schizo-
clozapine for schizophrenia, but the possibility phrenia: a multiple-treatments meta-analysis. Lancet
cannot be ruled out that they may be contami-nated 2013; 382: 951–962.
by ADR underreporting. Clozapine for TRBD may 7. Attard A, Taylor DM. Comparative effectiveness of
increase treatment compliance, which may offer atypical antipsychotics in schizophrenia. CNS Drugs 2012;
26: 491–508.
additional therapeutic benefits. On the other hand,
8. Owen R Jr, Beake B, Marby D, Dessain E, Cole J. Response
some patients may have poor adherence or may not to clozapine in chronic psychotic patients. Psy-
be willing to start cloza-pine treatment due to the chopharmacol Bull 1989; 25: 253.
blood collections required for the prevention of 9. Naber D, Hippius H. The European experience with use
agranulocytosis. of clozapine. Hosp Community Psychiatry 1990; 41: 886–
890.
 Clozapine for treatment-resistant bipolar disorder

10. Leppig M, Bosch B, Naber D, Hippius H. Clozapine in 30. Sachs GS. Treatment-resistant bipolar depression. Psy-
the treatment of 121 out-patients. Psychopharmacology chiatr Clin North Am 1996; 19: 215–236.
1989; 99: S77–S79. 31. Sharm V, Khan M, Corpse C. Role of lamotrigine in the
11. Banov MD, Zarate CA Jr, Tohen M et al. Clozapine therapy management of treatment-resistant bipolar II depression: a
in refractory affective disorders: polarity predicts chart review. J Affect Disord 2008; 111: 100–105.
response in long-term follow-up. J Clin Psychiatry 1994; 32. Phillips MR, Zhang J, Shi Q et al. Prevalence, treatment,
55: 295–300. and associated disability of mental disorders in four
12. McElroy SL, Dessain EC, Pope HG Jr et al. Clozapine in provinces in China during 2001–05: an epidemiological
the treatment of psychotic mood disorders, schizoaf- survey. Lancet 2009; 373: 2041–2053.
fective disorder, and schizophrenia. J Clin Psychiatry 1991; 33. Tsai S-Y, Chen C-C, Yeh E-K. Alcohol problems and
52: 411–414. long-term psychosocial outcome in Chinese patients with
13. Lindstrom€ LH. The effect of long-term treatment with bipolar disorder. J Affect Disord 1997; 46: 143–150.
clozapine in schizophrenia: a retrospective study in 96 34. Si TM, Shu L, Yu X et al. A cross-sectional study on
patients treated with clozapine for up to 13 years. Acta treatment patterns of bipolar disorders in China in 2006.
Psychiatr Scand 1988; 77: 524–529. Chin J Psychiatry 2012; 45: 29–34.
14. Frakenburg FR. Clozapine and bipolar disorder. J Clin 35. Chen XL, Xing BP, Zhang YQ, Jin WD. Atypical anti-
Psychopharmacol 1993; 13: 289–290. psychotics in the treatment of elder patients with mania.
15. Suppes T, Phillips KA, Judd CR. Clozapine treatment of Herald Med 2008; 27: 672–673.
nonpsychotic rapid cycling bipolar disorder: a report of 36. Fan ZG, Chen JM, Lu GH. Comparison analysis of
three cases. Biol Psychiatry 1994; 36: 338–340. clozapine treatment for 54 cases with mania Chin. J Neu-rol
16. Suppes T, McElroy SL, Gilbert J, Dessain EC, Cole JO. 1994; 39: 202.
Clozapine in the treatment of dysphoric mania. Biol Psy- 37. Liu YQ, Zhun FY, Yang LZ, Wang RJ, Chen CM. Clin-
chiatry 1992; 32: 270–280. ical treatment analysis of 528 cases of affective disorders
17. Poon SH, Sim K, Sum MY, Kuswanto CN, Baldessarini RJ. patients. Chin J Pharmacoepidemiol 1997; 5: 158–161.
Evidence-based options for treatment-resistant 38. Wang GY, Dong YM. The investigation and analysis of
adult bipolar disorder patients. Bipolar Disord 2012; 14: outpatient drug use in psychiatry. Chin J Modern Drug
573–584. Application 2007; 1: 44.
18. American Psychiatric Association. Practice guidelines 39. Liu TB, Gao H, Shen QJ. Clozapine combined with lith-
for the treatment of patients with bipolar disorder (revi- ium carbonate in the treatment of acute mania. J Clin
sion). Am J Psychiatry 2002; 159 (Suppl. 4): 1–50. Psychiatry 2000; 10: 350–351.
19. Judd LL, Akiskal HS, Schettler PJ et al. The long-term 40. Zhang L, Chen JD. A double-blind study of clozapine
natural history of the weekly symptomatic status of bipo-lar combined with lithium carbonate. J Psychiatry 2002; 15:
I disorder. Arch Gen Psychiatry 2002; 59: 530–537. 40–41.
20. Merikangas KR, Akiskal HS, Angst J et al. Lifetime and 12- 41. Yan JX. A double-blind clinical trial of add-on clozapine
month prevalence of bipolar spectrum disorder in the in the treatment of treatment-resistant depression. Clin J
National Comorbidity Survey replication. Arch Gen Psychol Med 2001; 11: 170–171.
Psychiatry 2007; 64: 543–552. 42. Xu XH. One case of treatment-resistant twin mania. J Clin
21. Hwu H-G, Joyce PR, Karam EG et al. Cross-national Psychosom Dis 2003; 9: 26.
epidemiology of major depression and bipolar disorder. 43. Tan J. The efficacy and safety of lithium and low-dose
JAMA 1996; 276: 293–299. clozapine treatment for refractory bipolar. Youjiang Med
22. McElroy SL, Frye M, Denicoff K et al. Olanzapine in 2010; 38: 150–151.
treatment-resistant bipolar disorder. J Affect Disord 1998; 44. Calabrese JR, Kimmel SE, Woyshville MJ et al. Cloza-
49: 119–122. pine for treatment-refractory mania. Am J Psychiatry 1996;
23. Goldberg JF, Burdick KE, Endick CJ. Preliminary ran- 153: 759–764.
domized, double-blind, placebo-controlled trial of pram- 45. Suppes T, Webb A, Paul B, Carmody T, Kraemer H, Rush
ipexole added to mood stabilizers for treatment-resistant AJ. Clinical outcome in a randomized 1-year trial of
bipolar depression. Am J Psychiatry 2004; 161: 564–566. clozapine versus treatment as usual for patients with
24. Ghaemi SN, Katzow JJ. The use of quetiapine for treat- treatment-resistant illness and a history of mania. Am J
ment-resistant bipolar disorder: a case series. Ann Clin Psychiatry 1999; 156: 1164–1169.
Psychiatry 1999; 11: 137–140. 46. Si TM, Shu L, Yu X et al. Antipsychotic drug patterns
25. Dell’Osso B, Mundo E, D’Urso N et al. Augmentative of schizophrenia in China: a cross-sectioned study. Chin
repetitive navigated transcranial magnetic stimulation J Psychiatry 2004; 37: 152–155.
(rTMS) in drug-resistant bipolar depression. Bipolar Disord 47. Zhang GP, Wang CY. The one-day survey on psychotro-
2009; 11: 76–81. pic drug use in 788 psychotic inpatients in Beijing An-ding
26. Erfurth A, Michael N, Stadtland C, Arolt V. Bupropion Hospital. Chin J Clin Pharmacol 2001; 17: 287–289.
as add-on strategy in difficult-to-treat bipolar depressive 48. Luo RL, Chen Q. Investigation and analysis of psycho-
patients. Neuropsychobiology 2002; 45: 33–36. tropic in 296 inpatients in one day. J Chin People Health
27. Pacchiarotti I, Mazzarini L, Colom F et al. Treatment- 2009; 21: 1755–1756.
resistant bipolar depression: towards a new definition. 49. Tang YL, Mao PX, Jiang F et al. Clozapine in China.
Acta Psychiatr Scand 2009; 120: 429–440. Pharmacopsychiatry 2008; 41: 1–9.
28. Gitlin MJ. Treatment-resistant bipolar disorder. Bull 50. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred
Menninger Clin 2001; 65: 26–40. reporting items for systematic reviews and meta-
29. Benedetti A, Di Paolo A, Lastella M et al. Augmenta-tion of analyses: the PRISMA statement. Ann Intern Med 2009;
clozapine with aripiprazole in severe psychotic 151: 264–269.
bipolar and schizoaffective disorders: a pilot study. Clin 51. Nielsen J, Kane JM, Correll CU. Real-world effective-ness
Pract Epidemiol Ment Health 2010; 6: 30–35. of clozapine in patients with bipolar disorder: results
Li et al.

from a 2-year mirror-image study. Bipolar Disord 2012; 69. Vieta E, Reinares M, Corbella B et al. Olanzapine as
14: 863–869. long-term adjunctive therapy in treatment-resistant bipo-lar
52. Varadhan KK, Neal KR, Lobo DN. Safety and efficacy of disorder. J Clin Psychopharmacol 2001; 21: 469–473.
antibiotics compared with appendicectomy for treat-ment of 70. Antonacci DJ, Swartz CM. Clozapine treatment of euphoric
uncomplicated acute appendicitis: meta-analysis of mania. Ann Clin Psychiatry 1995; 7: 203–206.
randomised controlled trials. BMJ 2012; 344: e2156. 71. Poyurovsky M, Weizman A. Safety and effectiveness of
53. Higgins JP, Green S. Cochrane Handbook for System-atic combined ECT and clozapine in treatment-resistant mania.
Reviews of Interventions. Hoboken, NJ: Wiley Online Eur Psychiatry 1996; 11: 319–321.
Library, 2008. 72. Mahmood T, Devlin M, Silverstone T. Clozapine in the
54. Higgins JPT, Altman DG, Sterne JAC. Assessing risk of management of bipolar and schizoaffective manic epi-
bias in included studies. In: Higgins JP, Green S eds. sodes resistant to standard treatment. Aust N Z J Psychi-
Cochrane Handbook for Systematic Reviews of Inter- atry 1997; 31: 424–426.
ventions. Hoboken, NJ: Cochrane Book Series, 2008: 187– 73. Chanpattana W. Combined ECT and clozapine in treat-
241. ment-resistant mania. J ECT 2000; 16: 204–207.
55. Leandro G. Meta-Analysis in Medical Research: The 74. Calabrese JR, Meltzer HY, Markovitz PJ. Clozapine
Handbook for the Understanding and Practice of Meta- prophylaxis in rapid cycling bipolar disorder. J Clin Psy-
Analysis. Wiley.com, 2008. Available from: http://www. chopharmacol 1991; 11: 396–397.
books.google.com [accessed October 2014]. 75. Lancon C, Llorca PM. Clozapine in the treatment of
56. Ried K. Interpreting and understanding meta-analysis refractory rapid cycling bipolar disorder. Encephale 1996;
graphs: a practical guide. Aust Fam Physician 2006; 35: 22: 468–469.
635–638. 76. Chen CK, Shiah IS, Yeh CB, Mao WC, Chang CC.
57. Hoffer ZS, Roth RL, Mathews M. Evidence for the par-tial Combination treatment of clozapine and topiramate in
dopamine-receptor agonist aripiprazole as a first-line resistant rapid-cycling bipolar disorder. Clin Neurophar-
treatment of psychosis in patients with iatrogenic or macol 2005; 28: 136–138.
tumorogenic hyperprolactinemia. Psychosomatics 2009; 77. Bastiampillai TJ, Reid CE, Dhillon R. The long-term
50: 317–324. effectiveness of clozapine and lamotrigine in a patient
58. Chang JS, Ha KS, Young Lee K, Sik Kim Y, Min Ahn Y. with treatment-resistant rapid-cycling bipolar disorder. J
The effects of long-term clozapine add-on therapy on Psychopharmacol 2010; 24: 1834–1836.
the rehospitalization rate and the mood polarity patterns in 78. Gupta M. Clozapine monotherapy for 66 months in
bipolar disorders. J Clin Psychiatry 2006; 67: 461–467. treatment resistant bipolar disorder: a case report. J Clin
59. Ciapparelli A, Dell’Osso L, Pini S, Chiavacci MC, Fenzi Psychopharmacol 2009; 29: 501–503.
M, Cassano GB. Clozapine for treatment-refractory 79. Vijay Sagar KJ. Treatment-refractory, juvenile-onset
schizophrenia, schizoaffective disorder, and psychotic bipolar affective disorder. Ind J Psychiatry 2005; 47: 124–
bipolar disorder: a 24-month naturalistic study. J Clin 125.
Psychiatry 2000; 61: 329–334. 80. Lieberman JA. Maximizing clozapine therapy: managing
60. Ciapparelli A, Dell’Osso L, Bandettini di Poggio A et al. side effects. J Clin Psychiatry 1998; 59 (Suppl. 3): 38–43.
Clozapine in treatment-resistant patients with schizo- 81. Safferman A, Lieberman JA, Kane JM, Szymanski S, Ki-
phrenia, schizoaffective disorder, or psychotic bipolar non B. Update on the clinical efficacy and side effects of
disorder: a naturalistic 48-month follow-up study. J Clin clozapine. Schizophrenia Bull 1991; 17: 247–261.
Psychiatry 2003; 64: 451–458. 82. Daniel DG, Goldberg TE, Weinberger DR, Kleinman JE.
61. Zarate CA Jr, Tohen M, Banov MD, Weiss MK, Cole Different side effect profiles of risperidone and cloza-pine
JO. Is clozapine a mood stabilizer? J Clin Psychiatry 1995; in 20 outpatients with schizophrenia or schizoaffec-
56: 108–112. tive disorder: a pilot study. Am J Psychiatry 1996; 153:
62. Kimmel SE, Calabrese JR, Woyshville MJ, Meltzer HY. 417–419.
Clozapine in treatment-refractory mood disorders. J Clin 83. Alvir JMJ, Lieberman JA, Safferman AZ, Schwimmer JL,
Psychiatry 1994; 55 (Suppl. B): 91–93. Schaaf JA. Clozapine-induced agranulocytosis–inci-
63. Green AI, Tohen M, Patel JK et al. Clozapine in the dence and risk factors in the United States. N J Engl Med
treatment of refractory psychotic mania. Am J Psychia-try 1993; 329: 162–167.
2000; 157: 982–986. 84. Id€anp€a€an-Heikkil€a J, Alhava E, Olkinuora M, Palva I.
64. Fehr BS, Ozcan ME, Suppes T. Low doses of clozapine Agranulocytosis during treatment with clozapine. Eur J Clin
may stabilize treatment-resistant bipolar patients. Eur Arch Pharmacol 1977; 11: 193–198.
Psychiatry Clin Neurosci 2005; 255: 10–14. 85. Atkin K, Kendall F, Gould D, Freeman H, Liberman J,
65. Kowatch RA, Suppes T, Gilfillan SK, Fuentes RM, O’Sullivan D. Neutropenia and agranulocytosis in
Granneman BD, Emlsie GJ. Clozapine treatment of chil- patients receiving clozapine in the UK and Ireland. Br J
dren and adolescents with bipolar disorder and schizo- Psychiatry 1996; 169: 483–488.
phrenia: a clinical case series. J Child Adolesc 86. Fan ZY, Cui HS, Shen AZ, Ju FY, Tang M, Bai SZ.
Psychopharmacol 1995; 5: 241–253. 3113 cases of WBC count of psychiatric patients after
66. Liu NH, Ying D. The clinical observation of clozapine add clozapine treatment. New Drug Clin 1992; 10: 330–331.
valproate and propranolol for treatment-resistant 87. Devinsky O, Honigfeld G, Patin J. Clozapine-related sei-
depression. Med J Chin People Health 2006; 18: 771. zures. Neurology 1991; 41: 369–371.
67. Tohen M, Sanger TM, McElroy SL et al. Olanzapine 88. Pacia SV, Devinsky O. Clozapine-related seizures: expe-
versus placebo in the treatment of acute mania. Am J rience with 5,629 patients. Neurology 1994; 44: 2247–
Psychiatry 1999; 156: 702–709. 2249.
68. Ketter TA, Kimbrell TA, George MS et al. Effects of mood 89. Haller E, Binder RL. Clozapine and seizures. Am J Psy-
and subtype on cerebral glucose metabolism in chiatry 1990; 147: 1069–1071.
treatment-resistant bipolar disorder. Biol Psychiatry 2001; 90. Miller DD. Review and management of clozapine side
49: 97–109. effects. J Clin Psychiatry 2000; 61 (Suppl. 8): 14–17.
 Clozapine for treatment-resistant bipolar disorder

91. Naber D, Holzbach R, Perro C, Hippius H. Clinical treatment of mania. Shandong Arch Psychiatry 2005; 18:
management of clozapine patients in relation to efficacy and 256–257.
side-effects. Br J Psychiatry Suppl 1992; 54–59. 107. Tao H, Jin WD, Liu ZC, Qiu DS, Ma YC, Shen Y et al.
92. Lieberman JA, Saltz BL, Johns CA, Pollack S, Boren- Controlled studies on olanzapine or clozapine and stabi-
stein M, Kane J. The effects of clozapine on tardive dys- lizer in patients with mania. J Clin Psychosom Dis 2004;
kinesia. Br J Psychiatry 1991; 158: 503–510. 10: 164–166.
93. Casey DE. Clozapine: neuroleptic-induced EPS and tar- 108. Dong YY, Gan JG. Randomized controlled study of
dive dyskinesia. Psychopharmacology 1989; 99: S47–S53. clozapine combined lithium versus ziprasidone combined
94. Tamminga C, Thaker G, Moran M, Kakigi T. Clozapine in lithium in the treatment of mania. J Pract Med 2010; 26:
tardive dyskinesia: observations from human and ani- 1806–1807.
mal model studies. J Clin Psychiatry 1994; 55 (Suppl. B): 109. Tang XW, Wu H, Chen Q. Randomized controlled trial
102–106. comparing quetiapine with lithium and clozapine with
95. Small JG, Milstein V, Marhenke JD, Hall DD. Treat-ment lithium in the treatment of female patients with mania.
outcome with clozapine in tardive dyskinesia, Shanghai Arch Psychiatry 2011; 23: 291–298.
neuroleptic sensitivity, and treatment-resistant psychosis. J 110. Chang W, Lin S, Lane H, Hu W, Jann M, Lin H. Cloza-
Clin Psychiatry 1987; 48: 263–267. pine dosages and plasma drug concentrations. J Form Med
96. Simpson GM, Lee JH, Shrivastava RK. Clozapine in tar- Assoc 1997; 96: 599–605.
dive dyskinesia. Psychopharmacology 1978; 56: 75–80. 111. Chong S-A, Tan C-H, Khoo Y-M et al. Clinical evalua-tion
97. Liu TZ, Song SY. A comparative study of clozapine ver- and plasma clozapine concentrations in Chinese
sus lithium in the treatment of mania. Shandong Arch patients with schizophrenia. Ther Drug Monit 1997; 19:
Psychiatry 1990; 3: 21–23. 219–223.
98. Liu HY, Gong CF, Ma ZL. A controlled study of cloza- 112. Sirot EJ, Knezevic B, Morena GP et al. ABCB1 and
pine, sodium valproate and carbamazepine in the treat-ment cytochrome P450 polymorphisms: clinical pharmacoge-
of 84 patients with mania. Health Psychol J 2002; netics of clozapine. J Clin Psychopharmacol 2009; 29: 319–
10: 228–230. 326.
99. Wu YH. Comparative study of clozapine versus olanza- 113. de Leon J, Santoro V, D’Arrigo C, Spina E. Interactions
pine in the treatment of mania. Pract Chin Clin Med 2004; between antiepileptics and second-generation antipsy-
21: 34–35. chotics. Expert Opin Drug Metab Toxicol 2012; 8: 311–
100. Wei XY, Wang QM, Yang YJ, Liu MY. The compari- 334.
son of olanzapine and clozapine in the treatment of mania. J 114. Diaz F, Santoro V, Spina E et al. Estimating the size of the
Clin Psychol Med 2002; 12: 347–348. effects of co-medications on plasma clozapine con-
101. Xie WJ, Zhao YM, Li J, Zhou Y, Chen M. A study of centrations using a model that controls for clozapine
aripiprazole versus clozapine in the treatment of mania. doses and confounding variables. Pharmacopsychiatry
Shandong Arch Psychiatry 2006; 19: 2. 2008; 41: 81–91.
102. Liao GS. A comparative study of quetiapine and cloza- 115. Spina E, De Leon J. Metabolic drug interactions with
pine in the treatment of mania episode. Med J Chin Peo-ple newer antipsychotics: a comparative review. Basic Clin
Health 2009; 21: 2993–2994. Pharmacol Toxicol 2007; 100: 4–22.
103. Yan JX, Chu PH, Liu QH, Chu JF, Lv XS, Zhang YQ. 116. Gitlin M. Treatment-resistant bipolar disorder. Mol Psy-
Clozapine combined with lithium in the treatment of mania. chiatry 2006; 11: 227–240.
Chin J Nerv Ment Dis 1997; 23: 104–106. 117. Frye MA, Ketter TA, Altshuler LL et al. Clozapine in
104. Tao Q, Yang XJ, Wang RQ. A comparative study of bipolar disorder: treatment implications for other atypi-
lithium and chlorpromazine versus lithium and clozapine cal antipsychotics. J Affect Disord 1998; 48: 91–104.
in the treatment of mania. Sichuan Arch Psychiatry 1999; 118. Ranjan R, Meltzer HY. Acute and long-term effective-
12: 56–57. ness of clozapine in treatment-resistant psychotic depres-
105. Zhang GY, Xie QM, Wang YX, Liu CM. Antipsychot-ics sion. Biol Psychiatry 1996; 40: 253–258.
combined with lithium in the treatment of mania. J Clin 119. Quante A, Zeugmann S, Bajbouj M, Anghelescu I.
Psychol Med 2006; 16: 2. Clozapine in medication- and electroconvulsive therapy-
106. Jing YL, Wang LT, Li J. A comparative study of olanza- resistant, depressed inpatients: a case series. J Clin Psy-
pine and clozapine combined with lithium in the chopharmacol 2007; 27: 715–717.