Schizophrenia Bulletin Advance Access published January 1, 2017

Schizophrenia Bulletin
doi:10.1093/schbul/sbw171

Immediate vs Gradual Discontinuation in Antipsychotic Switching:

ASystematic Review and Meta-analysis

HiroyoshiTakeuchi*,13, NavotKantor4, HiroyukiUchida3,5, TakefumiSuzuki3,6, and GaryRemington1,2,7,8

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at University of California, San Diego on January 3, 2017

1
Schizophrenia Division, Complex Care & Recovery Program, Centre for Addiction and Mental Health, Toronto, ON, Canada;
2
Department of Psychiatry, University of Toronto, Toronto, ON, Canada; 3Department of Neuropsychiatry, Keio University School of
Medicine, Tokyo, Japan; 4Faculty of Medicine, University of Ottawa, Ottawa, ON, Canada; 5Geriatric Mental Health Program, Centre
for Addiction and Mental Health, Toronto, ON, Canada; 6Department of Psychiatry, Inokashira Hospital, Tokyo, Japan; 7Institute
of Medical Science, University of Toronto, Toronto, ON, Canada; 8Campbell Family Mental Health Research Institute, Centre for
Addiction and Mental Health, Toronto, ON, Canada
*To whom correspondence should be addressed; Schizophrenia Division, Complex Care & Recovery Program, Centre for Addiction and
Mental Health, 250 College Street, Toronto, ON M5T 1R8, Canada; tel:+1-416-535-8501 (ext. 30547), fax: +1-416-979-4292, e-mail:
hirotak@dk9.so-net.ne.jp

Background: Antipsychotic switching is routine in clinical
practice, although it remains unclear which is the preferable
switching method: immediate discontinuation of the cur-
rent antipsychotic or a gradual tapering approach. The first
strategy has been implicated in rebound/withdrawal symp-
toms and emergence/exacerbation of symptoms, whereas
the gradual approach is thought to pose a risk of additive
or synergistic side effects if employed in the context of a
crossover approach. Methods: MEDLINE, Embase, and
Cochrane Central Register of Controlled Trials were sys-
tematically searched. Randomized controlled trials exam-
ining immediate vs gradual antipsychotic discontinuation in
antipsychotic switching in patients with schizophrenia and/
or schizoaffective disorder were selected. Data on clinical
outcomes, including study discontinuation, psychopathol-
ogy, extrapyramidal symptoms, and treatment-emergent
adverse events, were extracted. Results: A total of 9 stud-
ies involving 1416 patients that met eligibility criteria were
included in the meta-analysis. No significant differences in
any clinical outcomes were found between the 2 approaches
(all Ps > .05). Sensitivity analyses revealed that the find-
ings remained unchanged in the studies where switching to
aripiprazole was performed or where immediate initiation
of the next antipsychotic was adopted, while some signifi-
cant differences were observed in switching to olanzapine
or ziprasidone. Conclusions: These findings indicate that
either immediate or gradual discontinuation of the cur-
rent antipsychotic medication represents a viable treatment
option. Clinicians are advised to choose an antipsychotic
switching strategy according to individual patient needs.
This said, immediate discontinuation may be advantageous
both for simplicity and because a stalled cross-titration
process in antipsychotic switching could end up in antipsy-
chotic polypharmacy.
Key words: antipsychotics/discontinuation/schizophrenia/
switching strategy
Introduction
Switching to a different antipsychotic agent frequently
occurs in clinical practice during the treatment of schizo-
phrenia, most often related to suboptimal efficacy and/
or poor tolerability of the current agent.13 The process
of antipsychotic switching provides the opportunity for
multiple strategies48 that arise from the fundamental
options available for each drug (ie, immediate vs grad-
ual discontinuation of the current antipsychotic, as well
as full dose vs incremental titration of the new antipsy-
chotic). Variations can also include a gap between dis-
continuing the first and initiating the next antipsychotic.9
As summarized in supplementary table 1, immediate
antipsychotic discontinuation has been associated with
the following risks: (1) dopamine supersensitivity syn-
dromes (eg, supersensitivity psychosis and withdrawal
dyskinesia); (2) rebound syndromes related to cholin-
ergic, histaminergic, and serotonergic activity; and (3)
emergence/exacerbation of symptoms.4,5,7,8,10,11 On the
other hand, the gradual discontinuation has been linked
with an increased risk of side effects that may be addi-
tive or synergistic when used in the context of a crossover
approach.4,5,7,8,11 It has been reported that clinicians are
more likely to perform abrupt rather than cross-titration
switching,1215 while previous reviews of antipsychotic

The Author 2017. Published by Oxford University Press on behalf of the Maryland Psychiatric Research Center.
All rights reserved. For permissions, please email: journals.permissions@oup.com

Page 1 of 10
H. Takeuchi etal
switching strategies recommend gradual antipsychotic
discontinuation as a safer method in general7,8,10,11,16; such
endorsement is based on empirical evidence but not on
actual data from clinical trials.
There have been several randomized controlled tri-
als (RCTs) examining the 2 approaches.9,1724 In fact, a
meta-analysis in 2004 by our group reported no differ-
ences in either efficacy or safety25 although only 4 RCTs
met criteria for inclusion at that time.9,1719 However, a
further 5 RCTs comparing these approaches have since
been reported.2024 In light of how common antipsy-
chotic switching occurs, this question remains no less rel-
evant. In order to update information on this clinically
important topic, we conducted a systematic review and
meta-analysis of RCTs examining immediate vs gradual
antipsychotic discontinuation in patients with schizo-
phrenia undergoing a switch.
Methods
Literature Search and Study Selection
Two authors (H.T. and N.K.) independently conducted
a systematic literature search in accordance with the
PRISMA (Preferred Reporting Items for Systematic
Reviews and Meta-analyses) Statement.26 MEDLINE,
Embase, and Cochrane Central Register of Controlled
Trials (CENTRAL) were searched using the following
keywords: (schizophr* OR schizoaff*) AND antipsy-
chotic* AND switch* AND (strateg* OR tactic* OR
techni* OR procedure*), with a limitation of English
language (last search: August 30, 2016). Studies that met
the following eligibility criteria were selected: (1) RCTs;
(2) studies including patients with schizophrenia and/or
schizoaffective disorder; and (3) studies including both
immediate and gradual antipsychotic discontinuation
arms in antipsychotic switching. Risk of bias for each
included study was assessed according to the Cochrane
Handbook for Systematic Reviews of Interventions
(available at http://handbook.cochrane.org).
Data Extraction
Two authors (H.T.and N.K.) independently extracted the
following clinical outcome data in both immediate and
gradual antipsychotic discontinuation groups from the
selected studies: (1) the number of patients who discon-
tinued the study due to all causes, lack of efficacy, and
adverse events; (2) mean SD changes from baseline to
endpoint in the Positive and Negative Syndrome Scale
(PANSS)27 or the Brief Psychiatric Rating Scale (BPRS)28
total and positive symptom, the PANSS negative sub-
scale, and the Clinical Global Impression-Severity scale
(CGI-S)29 scores as psychopathology measures, and the
Simpson-Angus Scale (SAS)30 total, the Barnes Akathisia
Rating Scale (BARS)31 total or global, and the Abnormal
Involuntary Movement Scale (AIMS)29 total or Item 8
Page 2 of 10
scores as extrapyramidal symptom (EPS) measures; and
(3) the number of patients who experienced treatment-
emergent adverse events (TEAEs) that were reported in
3 out of the identified 9 studies (these included akathi-
sia, anxiety, diarrhea, headache, insomnia, nausea, and
somnolence).
If reports on the studies did not provide sufficient data,
we contacted the corresponding authors and/or funding
pharmaceutical company, accessed the ClinicalTrails.gov
website, and/or applied to data-sharing organizations to
which pharmaceutical companies sometimes belong in an
Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at University of California, San Diego on January 3, 2017
attempt to obtain additional information.
There are 2 points that should be noted in the data
extraction. First, since Lee et al17 did not clearly report
the number of patients who discontinued the study due
to inefficacy, we used 0 for both groups, estimated by sub-
tracting the total number of patients who discontinued
the study due to specific causes from those discontinuing
due to all causes in each group. Second, Kinon etal9 did
not report somnolence but rather drowsiness, which we
treated as synonomous. In addition, various descriptors
of insomnia were reported including shortened sleep,
difficulty falling asleep, interrupted sleep, and early
waking9; we chose to use shortened sleep for insomnia
but also conducted sensitivity analyses using each of the
descriptors.
Data Analysis
Meta-analyses were performed using Review Manager
(RevMan) version 5.3. Outcome data were combined and
compared between immediate and gradual antipsychotic
discontinuation groups. For dichotomous and continu-
ous outcomes, pooled estimates of risk ratios (RRs) and
standardized mean differences (SMDs) were calculated
with 2-sided 95% confidence intervals (CIs) using a ran-
dom-effects model, respectively.
Because we were able to obtain from Pfizer the individ-
ual data on all 3 studies included in a pooled analysis,19
we treated this pooled analytic study as 3 separate studies
(ie, Weiden 2003a, Weiden 2003b, and Weiden 2003c); in
addition, to evaluate the impact of treating one pooled
study as 3 separate studies, we conducted sensitivity anal-
yses using pooled results. If any study included another
switching strategy, we disregarded these group(s) as we
were focused on comparing the immediate and gradual
antipsychotic discontinuation strategies (supplementary
table 2). Since Kinon et al9 compared 4 antipsychotic
switching strategies: (1) immediate antipsychotic discon-
tinuation + immediate antipsychotic initiation; (2) imme-
diate antipsychotic discontinuation + wait-and-gradual
antipsychotic initiation; (3) gradual antipsychotic discon-
tinuation + immediate antipsychotic initiation; and (4)
gradual antipsychotic discontinuation + wait-and-grad-
ual antipsychotic initiation, we treated this investigation
as 2 separate studies (ie, Kinon 2000a and Kinon 2000b)

and compared arm (1) with arm (3) in one study, and arm
(2) with arm (4) in the other study to match the antipsy-
chotic initiation strategies (supplementary table 2); in
addition, to evaluate the impact of treating one study as
2 separate studies, we conducted sensitivity analyses only
using either one (ie, Kinon 2000a or Kinon 2000b). As
a consequence, a total of 12 comparisons of immediate
and gradual antipsychotic discontinuation were yielded.
As sensitivity analyses, we separately analyzed the follow-
ing 5 sets of studies: (1) blinding raters (N=4); (2) adopting
an immediate antipsychotic initiation strategy (8 compari-
sons); (3) switching to olanzapine (N=2); (4) switching to
ziprasidone (N=2); and (5) switching to aripiprazole (N=2).
All effect sizes with a P < .05 were considered signifi-
cant. Study heterogeneities were quantified using I2 sta-
tistic with I2s 50% indicating significant heterogeneity.
Results
Included Studies
A total of 9 studies9,1724 involving 1416 patients (n=714
and n=702 for the immediate and gradual antipsychotic
discontinuation, respectively) that met eligibility criteria
were identified and included in the meta-analysis (supple-
mentary figure1). The characteristics of these studies are
summarized in table 1. All studies represented parallel-
group RCTs with a relatively short duration, ranging
from 3 to 12 weeks, and were published in 2000 or later.
Only 1 study was conducted in a double-blind fashion, 3
were performed with raters blinded, and the remaining
5 were open-label studies. In the majority of studies, ris-
peridone, olanzapine, or haloperidol were prescribed as
the current antipsychotic prior to randomization; risperi-
done (N=1), olanzapine (N=2), ziprasidone (N=2),
aripiprazole (N = 2), iloperidone (N = 1), or clozapine
(N = 1) were introduced as the new antipsychotic. In
terms of gradual antipsychotic discontinuation proce-
dures, the current antipsychotic were discontinued in 1,
2, 3, and 4 weeks in 3, 4, 1, and 1 report(s), respectively.
In terms of initiation strategies of the new antipsychotic,
8 out of 12 comparisons used immediate initiation, while
4 used gradual or wait-and-gradual initiation (table1 and
supplementary table2).
The results of risk of bias assessment are displayed in
supplementary figure2.
Study Discontinuation
There was no significant difference in the number of
patients who discontinued the study due to all causes
(RR=1.10, 95% CI=0.931.30), inefficacy (RR=1.14,
95% CI=0.622.09), or intolerability (RR=1.08, 95%
CI=0.701.66) between the immediate and gradual anti-
psychotic discontinuation groups (figure1); no significant
study heterogeneities were observed for all study discon-
tinuation outcomes among the studies.
Immediate vs Gradual Antipsychotic Discontinuation
Psychopathology
No significant differences were found in any psychopa-
thology outcomes including the PANSS/BPRS total
(SMD = 0.03, 95% CI = 0.17 to 0.11) and positive
symptom (SMD=0.00, 95% CI=0.14 to 0.14), PANSS
negative subscale (SMD = 0.05, 95% CI = 0.19 to
0.10), and CGI-S scores (SMD=0.02, 95% CI=0.14
to 0.10) between the 2 groups (figure2). All I2s were 0%,
indicating no study heterogeneities.
Extrapyramidal Symptoms
Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at University of California, San Diego on January 3, 2017
There was no significant difference in scores for the
SAS (SMD = 0.12, 95% CI = 0.11 to 0.35), BARS
(SMD = 0.16, 95% CI = 0.33 to 0.65), or AIMS
(SMD = 0.27, 95% CI = 0.03 to 0.56) between the 2
groups (figure 3). Although not statistically significant,
the AIMS score tended in favor of gradual antipsychotic
discontinuation (P = .07); however, both baseline and
endpoint AIMS Item 8 (ie, global) scores were <1 (mini-
mal) for all studies included in the meta-analysis. No sig-
nificant heterogeneities were found in all EPS outcomes;
however, there was a marginally significant study hetero-
geneity for the BARS (P=.06).
Treatment-Emergent AdverseEvents
There were no significant differences in the number of
patients who experienced any TEAEs (ie, akathisia, anxi-
ety, diarrhea, headache, insomnia, nausea, and somno-
lence) between the immediate and gradual antipsychotic
discontinuation groups (figure 4); however, the former
tended to have a less favorable outcome in terms of
insomnia (P = .09). None of the study heterogeneities
for any TEAEs was significant. Although all types of
insomnia reported in Kinon et al9 were separately used
for insomnia in sensitivity analyses, none was significant
(data not shown).
Sensitivity Analyses
No significant differences in any clinical outcomes were
observed between the 2 groups in the following study data
sets: blinding raters; adopting an immediate antipsychotic
initiation strategy; and switching to aripiprazole (all Ps > .05)
(data not shown). In addition, no significant differ-
ences were found in either data set using Kinon 2000a
or Kinon 2000b, and the data set using pooled results of
Weiden 2003a, Weiden 2003b, and Weiden 2003c. On the
other hand, in switching to olanzapine there was a sig-
nificant difference regarding insomnia (RR=2.62, 95%
CI=1.305.29, P=.007) in favor of gradual discontinu-
ation, while in switching to ziprasidone there were sig-
nificant differences in the SAS scores (SMD=0.33, 95%
CI = 0.000.65, P = .05) and somnolence (RR = 0.25,
95% CI = 0.060.97, P = .05) in favor of gradual and
immediate discontinuation, respectively.
Page 3 of 10
 Table1. Randomized Controlled Trials Examining Immediate vs Gradual Antipsychotic Discontinuation in Antipsychotic Switching in Patients With Schizophrenia

Switching Strategy

Current Antipsychotics New Antipsychotics

Page 4 of 10
H. Takeuchi etal

Type (Frequency) Type and

Study Blinding Duration N and Mean Dosea Discontinuation Strategy Mean Dose Introduction Strategy

Casey 2003 Open-label 8 wk 104 RIS (37%), OLZ (55%), ID (reduce by 100% at day 0) APZ 30mg/d II (introduce 30mg/d at day 0)
TAPs (8%)
104 RIS (36%), OLZ (56%), GD (reduce by 50% at day 0, 75% at
TAPs (8%) week 1, and 100% at week 2)
Ganguli 2008 Rater-blind 6 wk 41 OLZ (100%) 14.4mg/d ID (reduce by 100% at day 0) RIS 4.3mg/d GI (introduce 2mg/d at day 0,
40 OLZ (100%) 15.5mg/d GD (reduce by 50% at day RIS 4.9mg/d increase to 4mg/d at day 3, and
0 and 100% at week 1) titrate the dose after week 1)
Kinon 2000a Open-label 3 wk 52 RIS (33%), HPD (31%), ID (reduce by 100% at day 0) OLZ 10mg/d II (introduce 10mg/d at day 0)
(discontinuing strategy) other TAPs (37%)
Double-blind 54 RIS (20%), HPD (28%), GD (reduce by 50% by week
(introducing strategy) other TAPs (52%) 1 and 100% by week 2)
Kinon 2000b 53 RIS (25%), HPD (28%), ID (reduce by 100% at day 0) WGI (introduce 5mg/d at week 1
other TAPs (47%) and increase to 10mg/d at week 2)
50 RIS (32%), HPD (22%), GD (reduce by 50% by week
other TAPs (46%) 1 and 100% by week 2)
Lee 2002 Open-label 6 wk 54 HPD (39%), ID (reduce by 100% at day 0) OLZ 12.2mg/d II (introduce 10mg/d at day 0 and
other APs (61%) then titrate 520mg/d)
54 HPD (41%), GD (reduce by 100% by week 2) OLZ 11.5mg/d
other APs (59%)
Pae 2009 Rater-blind 12 wk 31 RIS (35%), OLZ (42%), ID (reduce by 100% at day 0) APZ (dose not II (introduce 10mg/d at day 0 and
other AAPs (15%) reported) then titrate 1030mg/d)
29 RIS (52%), OLZ (36%), GD (reduce by 50% by week
other AAPs (12%) 2 and 100% by week 4)
Stip 2010 Open-label 6 wk 18 HPD (72%), other ID (reduce by 100% at day 0) ZIP 86.1mg/d II (introduce 80mg/d at day 0 and
TAPs (28%) titrate 80160mg/d after day 2)
18 HPD (67%), other GD (reduce by 50% at day ZIP 87.6mg/d
TAPs (33%) 0 and 100% at week 1)
Takeuchi 2016 Double-blind 8 wk 15 OLZ (67%), other ID (reduce by 100% at day 0) CLZ 342mg/d GI (introduce 12.5mg/d at day
(discontinuing strategy) APs (33%) 0, increase by 25mg/d everyday
Open-label 18 OLZ (44%), other GD (reduce by 25% at day 0, CLZ 292mg/d to 300mg/d at day 12, and titrate
(introducing strategy) APs (56%) 50% at week 1, 75% at week 2, after week 5)
and 100% at week 3)
Weiden 2003a Rater-blind 6 wk 35 TAPs (100%) ID (reduce by 100% at day 0) ZIP 90.3mg/d II (introduce 80mg/d at day 0 and
titrate 40160mg/d after day 2)
39 TAPs (100%) GD (reduce by 50% at day 0 ZIP 91.9mg/d
and 100% at week 1)
Weiden 2003b 34 OLZ (100%) ID (reduce by 100% at day 0) ZIP 88.6mg/d
36 OLZ (100%) GD (reduce by 50% at day 0 ZIP 91.1mg/d
and 100% at week 1)

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at University of California, San Diego on January 3, 2017

 Immediate vs Gradual Antipsychotic Discontinuation

Discussion

increase to 12mg/d by day 4, and

GI (introduce 2mg/d at day 0,

titrate 1224mg/d after day 4)

Our meta-analysis revealed that none of the clinical out-

Note: AAPs, atypical antipsychotics; APs, antipsychotics; APZ, aripiprazole; CLZ, clozapine; GD, gradual discontinuation; GI, gradual introduction; HPD, haloperidol;
comes, including study discontinuation, psychopathology,

ID, immediate discontinuation; II, immediate introduction; ILO, iloperidone; OLZ, olanzapine; RIS, risperidone; TAPs, typical antipsychotics; WGI, wait-and-gradual
Introduction Strategy EPS, and TEAEs, significantly differed between immedi-
ate vs gradual discontinuation of the current agent when
switching antipsychotics in patients with schizophrenia.
This was the case for the vast majority of the individual
studies included in this meta-analysis, which resulted in
no study heterogeneities across all clinical outcomes over-
all. These findings are consistent with a previous meta-
New Antipsychotics

analysis of 4 RCTs published in 2005,25 adding 5 more

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at University of California, San Diego on January 3, 2017

RCTs and a more robust methodology.
ILO 16.7mg/d
ZIP 97.3mg/d
ZIP 92.7mg/d

These findings are notably at odds with numerous

Mean Dose

reviews of antipsychotic switching strategies that recom-

Type and

mend gradual discontinuation of the current antipsy-

chotic to avoid rebound/withdrawal symptoms.7,8,10,11,16
One possible explanation relates to the fact that the
GD (reduce by 50% at day 0, 75% at

half-life for antipsychotic-related dopamine D2 receptor

occupancy centrally, as assessed with positron emission
ID (reduce by 100% at day 0)

ID (reduce by 100% at day 0)

tomography (PET), is considerably longer than plasma

week 1, and 100% at week 2)
GD (reduce by 50% at day 0

half-life.32,33 This, in turn, may act as a buffer in the face

Discontinuation Strategy

of abrupt antipsychotic discontinuation. It should be

and 100% at week 1)

noted, though, that marginally significant differences

in tardive dyskinesia and insomnia were observed, at
face value favoring gradual antipsychotic discontinua-
tion although the severity of tardive dyskinesia at both
baseline and endpoint was minimal (ie, <1 in the AIMS
global score) for all studies included in the meta-analysis.
However, this finding could as well reflect different phar-
macological properties beyond dopamine, as well as dose
260 RIS (36%), OLZ (29%),

240 RIS (34%), OLZ (33%),

Current Antipsychotics

equivalents, between the current and new antipsychotic

Switching Strategy

being initiated.7,8,10,11,16 For example, insomnia may be

Type (Frequency)
and Mean Dosea

caused by a rapid switch to an antipsychotic agent with

17 RIS (100%)
20 RIS (100%)

less potential to antagonize histamine H1 receptors.8,10,11,16

APZ (35%)

APZ (33%)

Notably, no differences in psychopathology were found

between immediate and gradual antipsychotic discon-
tinuation in switching, which is of interest given that lit-
erature suggests dopamine supersensitivity psychosis can
N

occur if patients stop antipsychotics abruptly.34 Finally, a

switch that entails moving to a more effective or tolerable
Duration

antipsychotic may mitigate risk of clinical worsening or

12 wk

adverse effects regardless of switching strategy.

In this meta-analysis, we focused on immediate and
gradual discontinuation of the current antipsychotic.
Nonetheless, 1 RCT35 and 4 arms of the included RCTs1922
comparing gradual vs wait-and-gradual discontinuation
introduction; ZIP, ziprasidone.

(supplementary table2) failed to demonstrate significant

Open-label

differences in the majority of clinical outcomes between

Blinding

the 2 strategies, which further supports the notion that

discontinuation strategies of the current antipsychotic
Table1. Continued

do not have any substantial clinical impact. As dis-

Weiden 2003c

cussed, how the new antipsychotic is introduced is also

Weiden 2014

If reported.

important. Strategies can vary as a function of dose

and titration strategies, and on occasion strategies may
Study

also employ a gap between discontinuation of the first

a

Page 5 of 10
H. Takeuchi etal
Fig.1. Study discontinuation.
antipsychotic and initiation of the next agent. Sensitivity
analysis including 8 comparisons that used an immedi-
ate introduction approach of the next antipsychotic
found no significant differences in any clinical outcomes
between immediate and gradual discontinuation of the
current antipsychotic. This left only 4 comparisons that
used other types of introduction strategies, hampering
any conclusions that can be drawn regarding how to
introduce the new antipsychotic.
In addition, differences in dose equivalents as well
as pharmacological properties must also be taken into
consideration when performing an antipsychotic switch.
For example, a switch to aripiprazole or, alternatively, a
switch from quetiapine/clozapine calls into play issues
not relevant to switches involving other antipsychotics,
the former related to aripiprazoles partial dopamine
agonist properties and low affinity vis-a-vis other recep-
tors, while the latter must take into account their low D2
receptor affinity.8 Sensitivity analyses revealed that none
of the clinical outcomes significantly differed between
Page 6 of 10
Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at University of California, San Diego on January 3, 2017
immediate and gradual antipsychotic discontinuation
strategies in those studies switching to aripiprazole, while
in switching to olanzapine or ziprasidone some significant
differences were found, suggesting that clinicians need to
take into account the pharmacological profiles of the next
antipsychotic, in particular for antipsychotics that have a
sedative effect such as olanzapine. However, each involves
only 2 reports and further studies are clearly needed.
There are limitations to the present study that warrant
comment. First, we were not able to obtain certain addi-
tional data, including the number of patients who dis-
continued in one of the largest studies of Kinon etal.9 In
addition, 1 RCT examining immediate vs wait-and-grad-
ual discontinuation of the current antipsychotic with a
switch to sertindole36 was not included in the meta-analysis
because no usable data were presented; this study reported
no significant differences in the vast majority of clinical
outcomes between the 2 strategies. Second, there were
some variations in duration of gradual antipsychotic dis-
continuation strategies among the studies, as noted in the
 Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at University of California, San Diego on January 3, 2017
Immediate vs Gradual Antipsychotic Discontinuation

Page 7 of 10
Fig.3. Extrapyramidal symptoms.
Fig.2.Psychopathology.
 Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at University of California, San Diego on January 3, 2017

Fig.4. Treatment-emergent adverse events.

H. Takeuchi etal

Page 8 of 10

Results section. Third, only 1 study was conducted in a
double-blind fashion, which may have influenced findings.
Fourth, study durations were relatively short, although
rebound/withdrawal symptoms are reported to usually
occur within weeks following an antipsychotic switch.4
Fifth, despite the frequency of antipsychotic switching in
actual clinical practice, the number of studies investigating
this issue remains limited, raising the possibility of type II
error. Lastly, most of the studies were presumed to largely
recruit patients with chronic schizophrenia, as reflected in
the established mean age of approximately 3545years (ill-
ness duration was provided in only 3 studies), which may
limit the generalizability of the present findings. Further
long-term, double-blind RCTs in patients with schizophre-
nia at various stages of the illness are needed.
In conclusion, the current meta-analysis of 9 RCTs
examining antipsychotic switching strategies demon-
strated no significant differences in clinical outcomes
between immediate and gradual antipsychotic discontin-
uation in schizophrenia, indicating that either immediate
or gradual discontinuation of the current antipsychotic
medication represents a viable treatment option. In
clinical practice, gradual discontinuation is frequently
observed because of the widely held notion that this
method diminishes risk of symptom exacerbation and/
or side effectsthis strategy is recommended as a safer
method in previous reviews. However, evidence thus far
does not necessarily support this notion. Clinicians are
advised to choose an antipsychotic switching strategy
according to individual patient needs. This said, immedi-
ate antipsychotic discontinuation may be advantageous
both for simplicity and because a stalled cross-titration
process in antipsychotic switching could end up in anti-
psychotic polypharmacy.3740
Supplementary Material
Supplementary data are available at Schizophrenia
Bulletin online.
Funding
Dr Takeuchi is supported through the Canadian Institutes
of Health Research (CIHR) Fellowship program. This
funding source had no role in study design, statistical
analysis or interpretation of findings, or in manuscript
preparation or submission for publication.
Acknowledgments
The authors thank Ms Sadhana Thiyanavadivel for
her kind help. Dr Takeuchi has received fellowship
grants from Centre for Addiction and Mental Health
(CAMH) Foundation, the Japanese Society of Clinical
Neuropsychopharmacology, and Astellas Foundation
for Research on Metabolic Disorders, and manuscript
Immediate vs Gradual Antipsychotic Discontinuation
fees from Sumitomo Dainippon Pharma. Mr Kantor
has no competing interests to disclose. Dr Uchida has
received grants from Astellas, Eisai, Eli Lilly, Meiji-
Seika Pharma, Mochida, Otsuka, Shionogi, Sumitomo
Dainippon Pharma, and Yoshitomi Yakuhin, and
speakers honoraria from Eli Lilly, Janssen, Meiji-Seika
Pharma, MSD, Otsuka, Pfizer, Shionogi, Sumitomo
Dainippon Pharma, and Yoshitomi Yakuhin. Dr Suzuki
has received manuscript or speakers honoraria from
Astellas, Eli Lilly, Elsevier Japan, Janssen, Meiji Seika
Pharma, Novartis, Otsuka, Sumitomo Dainippon
Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at University of California, San Diego on January 3, 2017
Pharma, and Wiley Japan. Dr Remington has received
research support from Novartis and consultant fees from
Synchroneuron.
References
1. Hugenholtz GW, Heerdink ER, Meijer WE, Stolker JJ,
Egberts AC, Nolen WA. Reasons for switching between
antipsychotics in daily clinical practice. Pharmacopsychiatry.
2005;38:122124.
2. Linden M, Pyrkosch L, Dittmann RW, Czekalla J. Why do
physicians switch from one antipsychotic agent to another?
The physician drug stereotype. J Clin Psychopharmacol.
2006;26:225231.
3. Roussidis A, Kalkavoura C, Dimelis D, et al. Reasons and
clinical outcomes of antipsychotic treatment switch in out-
patients with schizophrenia in real-life clinical settings: the
ETOS observational study. Ann Gen Psychiatry. 2013;12:42.
4. Weiden PJ, Aquila R, Dalheim L, Standard JM. Switching
antipsychotic medications. J Clin Psychiatry. 1997;58(suppl
10):6372.
5. Edlinger M, Baumgartner S, Eltanaihi-Furtmller N,
Hummer M, Fleischhacker WW. Switching between sec-
ond-generation antipsychotics: why and how? CNS Drugs.
2005;19:2742.
6. Correll CU. Real-life switching strategies with second-gener-
ation antipsychotics. J Clin Psychiatry. 2006;67:160161.
7. Lambert TJ. Switching antipsychotic therapy: what to expect
and clinical strategies for improving therapeutic outcomes. J
Clin Psychiatry. 2007;68(suppl 6):1013.
8. Cerovecki A, Musil R, Klimke A, et al. Withdrawal symp-
toms and rebound syndromes associated with switch-
ing and discontinuing atypical antipsychotics: theoretical
background and practical recommendations. CNS Drugs.
2013;27:545572.
9. Kinon BJ, Basson BR, Gilmore JA, Malcolm S, Stauffer
VL. Strategies for switching from conventional antipsy-
chotic drugs or risperidone to olanzapine. J Clin Psychiatry.
2000;61:833840.
10. Buckley PF. Receptor-binding profiles of antipsychotics:
clinical strategies when switching between agents. J Clin
Psychiatry. 2007;68(suppl 6):59.
11. Newcomer JW, Weiden PJ, Buchanan RW. Switching antip-
sychotic medications to reduce adverse event burden in
schizophrenia: establishing evidence-based practice. J Clin
Psychiatry. 2013;74:11081120.
12. Detke H, McDonnell D, Andersen S, Watson S. Olanzapine
long-acting injection in patients with schizophrenia at risk of
relapse: 12-week switching data. Eur Neuropsychopharmacol.
2008;18:S435.
Page 9 of 10
H. Takeuchi etal
13. Alptekin K, Hafez J, Brook S, et al. Efficacy and tolerabil-
ity of switching to ziprasidone from olanzapine, risperidone
or haloperidol: an international, multicenter study. Int Clin
Psychopharmacol. 2009;24:229238.
14. Cazorla P, Mackle M, Zhao J, Ha X, Szegedi A. Safety
and tolerability of switching to asenapine from other
antipsychotic agents: pooled results from two randomized
multicenter trials in stable patients with persistent nega-
tive symptoms in schizophrenia. Neuropsychiatr Dis Treat.
2012;8:247257.
15. de Smidt C, Haffmans J, Hoencamp E. Antipsychotics
switching strategies in real life: a longitudinal study in clinical
practice. Eur J Psychiatry. 2012;26:4149.
16. Correll CU. Switching and combining antipsychotics. CNS
Spectr. 2010;15:811.
17. Lee CT, Conde BJ, Mazlan M, etal. Switching to olanzapine
from previous antipsychotics: a regional collaborative multi-
center trial assessing 2 switching techniques in Asia Pacific.
J Clin Psychiatry. 2002;63:569576.
18. Casey DE, Carson WH, Saha AR, et al.; Aripiprazole
Study Group. Switching patients to aripiprazole from other
antipsychotic agents: a multicenter randomized study.
Psychopharmacology (Berl). 2003;166:391399.
19. Weiden PJ, Simpson GM, Potkin SG, OSullivan RL.
Effectiveness of switching to ziprasidone for stable but symp-
tomatic outpatients with schizophrenia. J Clin Psychiatry.
2003;64:580588.
20. Ganguli R, Brar JS, Mahmoud R, Berry SA, Pandina GJ.
Assessment of strategies for switching patients from olanzap-
ine to risperidone: a randomized, open-label, rater-blinded
study. BMC Med. 2008;6:17.
21. Pae C-U, Serretti A, Chiesa A, et al. Immediate versus
gradual suspension of previous treatments during switch to
aripiprazole: results of a randomized, open label study. Eur
Neuropsychopharmacol. 2009;19:562570.
22. Stip E, Zhornitsky S, Potvin S, Tourjman V. Switching
from conventional antipsychotics to ziprasidone: a ran-
domized, open-label comparison of regimen strategies.
Prog Neuropsychopharmacol Biol Psychiatry. 2010;34:
9971000.
23. Weiden PJ, Citrome L, Alva G, et al. A trial evaluating
gradual- or immediate-switch strategies from risperidone,
olanzapine, or aripiprazole to iloperidone in patients with
schizophrenia. Schizophr Res. 2014;153:160168.
24. Takeuchi H, Lee J, Fervaha G, Foussias G, Agid O,
Remington G. Switching to clozapine using immediate versus
gradual antipsychotic discontinuation: a pilot, double-blind,
randomized controlled trial. J Clin Psychiatry. In press.
Page 10 of 10
25. Remington G, Chue P, Stip E, Kopala L, Girard T, Christensen
B. The crossover approach to switching antipsychotics: what
is the evidence? Schizophr Res. 2005;76:267272.
26. Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA Group.
Preferred reporting items for systematic reviews and meta-
analyses: the PRISMA statement. BMJ. 2009;339:b2535.
27. Kay SR, Fiszbein A, Opler LA. The positive and negative
syndrome scale (PANSS) for schizophrenia. Schizophr Bull.
1987;13:261276.
28. Overall JE, Gorham DR. The Brief Psychiatric Rating Scale.
Psychol Rep. 1962;10:799812.
29. Guy W. ECDEU Assessment Manual for Psychopharmacology.
Rockville, MD: National Institute of Mental Health; 1976.
Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at University of California, San Diego on January 3, 2017
30. Simpson GM, Angus JW. A rating scale for extrapyramidal
side effects. Acta Psychiatr Scand Suppl. 1970;212:1119.
31. Barnes TR. A rating scale for drug-induced akathisia. Br J
Psychiatry. 1989;154:672676.
32. Tauscher J, Jones C, Remington G, Zipursky RB, Kapur S.
Significant dissociation of brain and plasma kinetics with
antipsychotics. Mol Psychiatry. 2002;7:317321.
33. Takano A, Suhara T, Ikoma Y, etal. Estimation of the time-
course of dopamine D2 receptor occupancy in living human
brain from plasma pharmacokinetics of antipsychotics. Int J
Neuropsychopharmacol. 2004;7:1926.
34. Yin J, Barr AM, Ramos-Miguel A, Procyshyn RM.
Antipsychotic induced dopamine supersensitivity psychosis:
a comprehensive review [published online ahead of print June
5,2016]. Curr Neuropharmacol.
35. Takeuchi H, Suzuki T, Uchida H, etal. A randomized, open-
label comparison of 2 switching strategies to aripiprazole
treatment in patients with schizophrenia: add-on, wait, and
tapering of previous antipsychotics versus add-on and simul-
taneous tapering. J Clin Psychopharmacol. 2008;28:540543.
36. Alptekin K, Akdede B, Soygur H, Group TS. Tries: an open,
randomized, prospective, multicenter study, searching the
best switch policy of sertindole in patients with schizophre-
nia. Schizophr Res. 2012;136:S164S165.
37. Tapp A, Wood AE, Secrest L, Erdmann J, Cubberley L,
Kilzieh N. Combination antipsychotic therapy in clinical
practice. Psychiatr Serv. 2003;54:5559.
38. Sernyak MJ, Rosenheck R. Clinicians reasons for antipsy-
chotic coprescribing. J Clin Psychiatry. 2004;65:15971600.
39. Barnes TR, Paton C. Antipsychotic polypharmacy in schizo-
phrenia: benefits and risks. CNS Drugs. 2011;25:383399.
40. Grech P, Taylor D. Long-term antipsychotic polyphar-
macy: how does it start, why does it continue? Ther Adv
Psychopharmacol. 2012;2:511.