Schizophrenia Research 57 (2002) 227 – 238

www.elsevier.com/locate/schres

Extrapyramidal symptom profiles in Japanese patients with

schizophrenia treated with olanzapine or haloperidol
Toshiya Inada a,*, Gohei Yagi c, Sadanori Miura b
a
Department of Geriatric Mental Health, National Institute of Mental Health, National Center of Neurology and Psychiatry, Ichikawa, Japan
b
Department of Psychiatry, Kitasato University, Kanagawa, Japan
c
Department of Neuropsychiatry, School of Medicine, Keio University, Tokyo, Japan
Received 20 January 2001; accepted 23 July 2001

Abstract

Previous clinical trials have clearly shown the superiority of olanzapine to haloperidol in the improvement of extrapyramidal
symptoms (EPS) in schizophrenic patients. The primary purpose of this study was to compare EPS profiles in Japanese
schizophrenic patients treated with an atypical antipsychotic, olanzapine, or a typical antipsychotic, haloperidol, as measured by
the Drug-Induced Extrapyramidal Symptoms Scale (DIEPSS). The DIEPSS, which consists of eight individual parameters and
one global assessment (overall severity), was used to evaluate 182 patients enrolled in this 8-week study. The primary safety
analysis was maximum change (that could be either a decrease or increase) from baseline in DIEPSS total score. Secondary
analyses included change from baseline to maximum in DIEPSS total score, change from baseline to endpoint (LOCF) in
DIEPSS total score, and the rank sum of the maximum change (that could be either a decrease or increase) from baseline in the
DIEPSS individual items. Incidence of treatment-emergent EPS adverse events using the DIEPSS scale was also analyzed. The
olanzapine group showed statistically significant superiority to the haloperidol group on the primary analysis ( p < 0.001).
Secondary analyses also demonstrated olanzapine’s superiority in DIEPSS total, parkinsonism, akathisia and overall severity
scores (all p V 0.014). Categorical analysis of treatment-emergent akathisia and parkinsonism syndromes at endpoint showed
improvement in the olanzapine group but worsening in the haloperidol group. The results from this study suggest that
olanzapine, as in Caucasian populations, is a safe treatment in Japanese patients chronically ill with schizophrenia. D 2002
Elsevier Science B.V. All rights reserved.

Keywords: Akathisia; DIEPSS; Haloperidol; Olanzapine; Parkinsonism

1. Introduction extent, disorganized speech and behavior. However,

Typical antipsychotics (e.g., haloperidol) are effec-
tive in treating positive symptoms and to a lesser
*
Corresponding author. Department of Geriatric Mental
Health, National Institute of Mental Health, National Center of
Neurology and Psychiatry, Chiba, Japan. Tel.: +81-47-372-
0141x1272; fax: +81-47-371-2900.
E-mail address: inada@ncnp-k.go.jp (T. Inada).
these drugs lack efficacy in treating the negative
symptoms of schizophrenia (Kane and Mayerhoff,
1989; Möller, 1993) and often result in unpleasant
side effects such as extrapyramidal symptoms (EPS)
including tardive dyskinesia (TD) (Keepers et al.,
1983).
Atypical antipsychotics have recently been intro-
duced as drug therapy for schizophrenia. Patients
treated with atypical antipsychotics generally experi-

0920-9964/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 9 2 0 - 9 9 6 4 ( 0 1 ) 0 0 3 1 4 - 0
228 T. Inada et al. / Schizophrenia Research 57 (2002) 227–238
ence fewer EPS (especially acute dystonias) than
patients treated with typical antipsychotics (Casey,
1992; Meltzer, 1992; Gerlach and Peacock, 1995).
Nevertheless, haloperidol is the most widely pre-
scribed antipsychotic in much of Asia. Plasma con-
centrations of haloperidol vary considerably between
ethnic groups with Asians tending to have higher
plasma levels than Caucasians, Blacks and Hispanics
(Lam et al., 1995; Lane et al., 1997; Wood and Zhou,
1991). Therefore, Asians generally require lower
doses of haloperidol (Jibiki et al., 1993). Importantly,
Asian patients are more susceptible to EPS even after
taking lower doses of haloperidol (Lam et al., 1995;
Binder and Levy, 1981). Thus, it is imperative to
provide alternative antipsychotics to this patient pop-
ulation.
However, few studies have been conducted to
compare EPS profiles with atypical and typical anti-
psychotics in Asian patients. Olanzapine, an atypical
antipsychotic, has shown a favorable profile for EPS
and tardive dyskinesia for primarily Caucasian
patients (Tollefson et al., 1997a; Tran et al., 1997).
In addition, EPS were improved in treatment-resistant
patients taking olanzapine versus haloperidol (Breier
and Hamilton, 1999).
The Japanese versions of two standardized EPS
scales, the Abnormal Involuntary Movement Scale
(for dyskinesia) (Itoh et al., 1977) and the Barnes
Akathisia Scale (Inada et al., 1996) have been avail-
able in Japan. However, standardized rating scales that
can evaluate the entire range of EPS symptoms,
including parkinsonism and dystonia, as well as
akathisia and dyskinesia, did not exist in Japan.
Therefore, evaluation of treatment-emergent EPS in
Japanese clinical studies of antipsychotics have been
performed with simple methods such as: (1) recording
adverse events (i.e., emergent symptoms) that appear
during the clinical study and (2) comparing the
incidences of anticholinergic use between study drugs.
In response to the need for a standardized EPS
rating scale in Japan, the Drug-Induced Extrapyrami-
dal Symptoms Scale (DIEPSS) was created in 1994.
The design and reliability of this scale was established
the following year (Inada and Yagi, 1995) and a
manual on how to properly conduct an examination
for scoring the DIEPSS was published in 1996 (Inada,
1996). The scale consists of eight individual symp-
toms and one global assessment (see Appendix A).
Description and rating procedure for each symptom is
described in both English and Japanese. The DIEPSS
has now become a standard rating scale for evaluating
EPS in antipsychotic clinical trials in Japan (Inada and
Yagi, 1996).
We report the EPS profile results from a random-
ized, double-blind study conducted for 8 weeks in
Japan comparing olanzapine to haloperidol in treat-
ing chronically ill schizophrenic patients. This is the
first report focusing on EPS profiles in Asian schiz-
ophrenic patients taking either olanzapine or a typ-
ical antipsychotic. The analysis included 174 patients
at 67 investigative institutions in Japan. A full
description of overall results of safety and efficacy
for this study will be published in Ishigooka et al.
(2001).
2. Method
2.1. Subjects
All patients in this study met the F.20 category in
the ICD-10 DCR classification for schizophrenia
(WHO, 1993). Patients included men and women
between the ages of 18 and 65 (mean age 42.9).
The patients (or their guardians) gave written
informed consent prior to entering the study. Ninety-
three and eighty-nine patients were randomized to
olanzapine and haloperidol, respectively. If a patient
used no pretreatment drug, they started the study drug
immediately. If a depot form was used or if oral
haloperidol was pretreatment drug, there was a wash-
out period of 4 or 2 weeks, respectively. If pretreated
with multiple antipsychotic drugs, an observation
period of 2 weeks before study start was employed.
Prescription of pretreatment drugs was simplified and
dose reduced to 12 mg haloperidol equivalence during
the observation period.
2.2. Dosage
Initial starting dose of olanzapine was 5.0 mg/day
with dose increases (or decreases), when needed, in
increments of 2.5 mg/day to a maximum of 15.0 mg/
day. The initial starting dose for haloperidol was 4.0
mg/day with a dose increase (or decrease) of 2.0 mg/
day to a total of 12.0 mg/day.
 T. Inada et al. / Schizophrenia Research 57 (2002) 227–238
2.3. Outcome measures
The primary safety hypothesis was that there
would be a difference between the treatment groups
in EPS severity. The DIEPSS, which consists of eight
individual items and one global item, was used to
assess treatment-emergent extrapyramidal symptoms
(Inada, 1996). The eight individual items include (1)
gait, (2) bradykinesia, (3) sialorrhea (increased sali-
vation), (4) muscle rigidity, (5) tremor, (6) akathisia,
(7) dystonia, and (8) dyskinesia. The global item is
overall severity. Extrapyramidal syndromes, as meas-
ured by the DIEPSS were grouped into four catego-
ries: (1) parkinsonism, (2) akathisia, (3) dystonia, and
(4) dyskinesia. The parkinsonism syndrome consisted
of DIEPSS items 1 –5. The akathisia, dystonia and
dyskinesia syndromes consisted of DIEPSS items 6, 7
and 8, respectively. The severity of each item is rated
from 0 (normal) to 4 (severe). Extrapyramidal symp-
toms were evaluated at baseline and 1, 2, 3, 4, 6 and 8
weeks after starting administration until discontinua-
tion. To be included in this and all other analyses, the
patient required a baseline and at least one postbase-
line score.
The primary safety analysis was maximum change
from baseline in DIEPSS total score (items 1 –8). The
maximum change from baseline was determined for
each item using the score with the largest absolute
change from baseline regardless of direction. The
maximum change from baseline in DIEPSS total score
was determined as the primary analysis for the follow-
ing reason. Change from baseline to maximum, which
uses the highest postbaseline score, accounts only for
worsening of EPS, and improvement is neglected.
Change from baseline to endpoint may be confounded
with the use of anticholinergic medication since it was
allowed when EPS occurs during the study period.
Therefore, maximum change from baseline was con-
sidered most appropriate for evaluating EPS changes
during olanzapine treatment. However, since maxi-
mum change from baseline may not account for
recurrent worsening after the improvement, analysis
of change from baseline to maximum and change
from baseline to last observation carried forward
(LOCF) endpoint was considered secondary to com-
plement the analysis of EPS changes during olanza-
pine treatment. Another secondary safety analysis was
the rank sum of the maximum change from baseline
229
of the DIEPSS individual items. DIEPSS total score
and subscale scores were used for these analyses.
Analyses of maximum change from baseline in the
DIEPSS total score were performed to examine the
consistency of treatment effects over the strata of
various demographic and clinical profile character-
istics. These subgroup variables included age, gender,
duration of illness, PANSS symptomatic classifica-
tion, past haloperidol exposure, previous antipsy-
chotic dose, previous antiparkinson medication use,
and antiparkinson medication use without EPS.
The DIEPSS was used to determine the categorical
incidence of treatment-emergent syndromes of dysto-
nia, parkinsonism, akathisia and dyskinesia. Abnor-
mal for the parkinsonism syndrome at baseline was
defined as z 3 on one item or z 2 on two items;
abnormal at endpoint was defined as for baseline, plus
an increase of z 3 from baseline on the parkinsonism
total. Abnormal for akathisia, dystonia and dyskinesia
syndromes at baseline was defined as z 2; abnormal
at endpoint was defined as for baseline, plus an
increase of z 2 from baseline. Patients meeting the
abnormal criteria at baseline were not included in the
analysis.
2.4. Statistical analysis
SAS (version 6.08) was used to perform statistical
analyses (SAS, 1989). Continuous and ordinal data
were evaluated with the Wilcoxon rank-sum test due
to non-normal distribution of the data. For all analy-
ses, the effects were tested at the two-sided a level of
0.05 except when noted.
Comparability of baseline characteristics between
olanzapine and haloperidol was tested with the Wil-
coxon rank-sum test for age (median), whereas gen-
der was tested with the Fisher’s exact test (two-sided).
All illness characteristics were tested with the Wil-
coxon rank-sum test except for schizophrenic sub-
type, which was tested using the Freeman Halton test
(two-sided).
Inter-group comparisons for all analyses were
tested with the Wilcoxon rank-sum test except for
incidence of treatment-emergent EPS syndromes at
any time or at endpoint which was tested with the
Fisher’s exact test (one-sided with an a level of
0.025). Within treatment group changes were tested
with the Wilcoxon signed-rank test for maximum
230 T. Inada et al. / Schizophrenia Research 57 (2002) 227–238

change from baseline, change from baseline to max-
imum and change from baseline to endpoint (LOCF).
Treatment group differences for the rank-sum test
of maximum change from baseline was tested with the
O’Brien nonparametric rank-sum test (O’Brien,
1984). Nonparametric methods were used to test the
subgroup by treatment interactions for the profile
characteristics with the maximum change from base-
line DIEPSS total scores (Akritas et al., 1997).
3. Results
3.1. Baseline characteristics
Patient mean age was 42.9 years (median, 43.5
years) with 62.6% being male. Most (94.3%) were
taking previous medication immediately before begin-
ning the study and 79.9% were inpatients when enter-
ing the study. Hebephrenic (36.8%) and paranoid
(28.2%) were the most common schizophrenic sub-
types (Table 1). Patients had a mean and median age
of 26.3 and 23.8 years at onset of psychosis, respec-
tively. Illness duration was z 5 years in 80% of the
patients with a mean and median of 17.2 and 16.0
years, respectively. These patients demonstrated
chronic presence of symptoms. The current episode
averaged 6.3 years (median, 1.9). There was no
significant difference in any patient or illness charac-
teristic or in the baseline EPS scores between the two
treatment groups.
3.2. Drug dose
The mean ( F S.D.) for all olanzapine-treated
patients regarding modal dose (dose on a daily basis

Table 1
Illness characteristics
Variable Olanzapine (n = 90) Haloperidol (n = 84) Total (n = 174) Overall p-value
Subtype (%)
Paranoid 32.2 23.8 28.2 0.320a
Hebephrenic 31.1 42.9 36.8
Catatonic 3.3 2.4 2.9
Undifferentiated 10.0 6.0 8.0
Post-schizophrenic depression 0.0 3.6 1.7
Residual 20.0 17.9 19.0
Simple 3.3 2.4 2.9
Other schizophrenia 0.0 1.2 0.6

Age of psychosis onset (year)

Mean F S.D. (range) 27.1 F 10.0 (14.4 – 58.2) 25.4 F 8.2 (14.2 – 47.5) 26.3 F 9.2 (14.2 – 58.2) 0.328b,c

Duration of illness (year)

Mean F S.D. (range) 17.1 F 12.8 (0.6 – 48.4) 17.3 F 12.1 (0.4 – 43.4) 17.2 F 12.4 (0.4 – 48.4) 0.761b

Length of current episode (year)

Mean F S.D. (range) 6.6 F 9.2 (0.0 – 37.0) 6.0 F 8.7 (0.0 – 36.3) 6.3 F 8.9 (0.0 – 37.0) 0.845b,d

Number of previous episodes (%)

Initial 30.0 26.2 28.2 0.543b
2 20.0 21.4 20.7
3 11.1 7.1 9.2
4 5.6 3.6 4.6
z5 14.4 20.2 17.2
Unknown 18.9 21.4 20.1
a
Treatment group comparison of this baseline characteristic was tested with Freeman Halton test (two-sided).
b
Treatment group comparison of this baseline characteristic was tested with Wilcoxon rank-sum test.
c
The median for age of psychosis onset is 23.8 years for both olanzapine and haloperidol.
d
The median for length of current episode is 1.9 years for both olanzapine and haloperidol.
 T. Inada et al. / Schizophrenia Research 57 (2002) 227–238
taken most frequently for each individual patient), last
dose and maximum dose were 10.31 F 3.91,
10.53 F 3.78 and 10.86 F 3.81 mg/day, respectively.
The means for modal dose, last dose and maximum
dose were 7.36 F 3.03, 8.00 F 3.03 and 8.33 F 2.98
mg/day, respectively, for the haloperidol-treated
patients.
3.3. Concomitant medication
There were significantly fewer ( p = 0.033) olanza-
pine-treated patients (40.0%) taking antiparkinson
medication than haloperidol-treated patients (57.1%)
during study participation. For days when patients
were taking drugs, there was a significant difference
( p = 0.003) in antiparkinson medication dosage
(biperiden equivalents) during weeks 1 – 8, with olan-
zapine-treated patients (2.42 mg/day) having a lower
daily dose compared with haloperidol-treated patients
(3.51 mg/day) (Inagaki et al., 1998).
3.4. Patient disposition
A significantly greater ( p = 0.021) proportion of
olanzapine-treated patients (80.6%) completed this
study than haloperidol-treated patients (66.3%). Sig-
nificantly less ( p = 0.003) olanzapine-treated patients
(8.9%) discontinued the study due to adverse events
or an abnormal laboratory value compared to haloper-
idol-treated patients (26.2%). One haloperidol-treated
patient committed suicide during the study. From the
original 182 patients randomized to either olanzapine
or haloperidol, seven patients were excluded because
of protocol violations and a single patient was
excluded because the investigator previously commit-
ted a violation in another sponsor’s trial, which dis-
Table 2
Maximum change from baseline in DIEPSS total score
Therapy N Baseline Maximuma
Mean S.D. Mean
Olanzapine 81 2.90 3.10 2.40
Haloperidol 80 2.80 3.00 4.30
N = total number of patients with a baseline and at least one postbaseline score.
a
Score with the largest absolute change from baseline regardless of direction.
b
Within treatment group changes were tested with Wilcoxon signed-rank test.
c
Treatment group comparisons of change score distributions were tested with the Wilcoxon rank-sum test.
231
qualified him from this trial. Therefore, a total of 174
patients were included in the analyses.
3.5. Safety-extrapyramidal symptoms rating scale
Extrapyramidal symptoms in olanzapine-treated
patients versus haloperidol-treated patients were sig-
nificantly improved ( p < 0.001) in maximum change
from baseline in the total DIEPSS score (Table 2). The
DIEPSS total score for within group treatment was
significantly improved ( p = 0.03) for the olanzapine-
treated patients but significantly worsened ( p < 0.001)
for the haloperidol-treated patients. In the correspond-
ing secondary safety analysis, a statistically signifi-
cant difference ( p < 0.001) between treatment groups
in rank sum of the maximum change from baseline of
the individual DIEPSS items was observed. The
median in the olanzapine group (631.5) was less than
that observed in the haloperidol group (694.0), indi-
cating greater worsening of EPS in the haloperidol
group at any time during the study.
For change from baseline to maximum, the total
DIEPSS score, the overall severity score and the
akathisia and parkinsonism subscale scores were all
significantly improved ( p < 0.001) in the olanzapine
group compared to the haloperidol group (Table 3).
No change in median total score was observed in the
olanzapine group compared with a median increase of
1.0 in the haloperidol group. The median changes for
all items were 0.0 in both treatment groups. However,
as a whole, the mean changes were reflective of
greater worsening of symptoms in the haloperidol
group compared with the olanzapine group.
The total DIEPSS score, overall severity and the
akathisia and parkinsonism subscale scores were also
statistically significantly improved ( p V 0.014) for
Change Within group Therapy
p-valueb p-valuec
S.D. Median Mean S.D.
3.70 0.0 0.50 2.90 0.030 < 0.001
4.20 1.0 1.50 3.70 < 0.001
232 T. Inada et al. / Schizophrenia Research 57 (2002) 227–238

Table 3
Change from baseline to maximum in DIEPSS total score
DIEPSS score Therapy N Baseline Maximuma Change Within group Therapy
p-valueb p-valuec
Mean S.D. Mean S.D. Median Mean S.D.
Total Olz 81 2.90 3.10 3.10 3.90 0.0 0.20 2.20 0.963 < 0.001
Hal 80 2.80 3.00 4.70 4.10 1.0 1.90 3.10 < 0.001
Parkinsonism Olz 81 2.50 2.70 2.50 3.20 0.0 0.10 1.60 0.522 < 0.001
Hal 80 2.30 2.50 3.60 3.60 0.0 1.30 2.60 < 0.001
Akathisia Olz 81 0.30 0.70 0.40 0.80 0.0 0.10 0.80 0.373 < 0.001
Hal 80 0.20 0.60 0.80 1.10 0.0 0.50 1.00 < 0.001
Dystonia Olz 81 0.10 0.30 0.10 0.50 0.0 0.10 0.40 0.500 0.296
Hal 80 0.10 0.60 0.30 0.80 0.0 0.20 0.70 0.094
Dyskinesia Olz 81 0.10 0.30 0.10 0.50 0.0 0.10 0.40 0.375 0.994
Hal 80 0.20 0.60 0.20 0.50 0.0 0.00 0.50 1.000
Overall severity Olz 81 0.90 0.70 1.00 0.90 0.0 0.10 0.80 0.176 < 0.001
Hal 80 1.00 0.90 1.50 1.10 0.0 0.60 1.00 < 0.001
N = total number of patients with a baseline and at least one postbaseline score.
a
Score with the largest change from baseline.
b
Within treatment group changes were tested with Wilcoxon signed-rank test.
c
Treatment group comparisons of change score distributions were tested with the Wilcoxon rank-sum test.

endpoint change (LOCF) in the olanzapine-treated
patients compared to the haloperidol-treated patients
(Table 4). No significant differences in dystonia or
dyskinesia occurred between treatment groups based
on endpoint scores.
No significant interactions between subgroups (age,
gender, duration of illness) and treatments (past hal-
operidol exposure, previous antipsychotic dose, pre-
vious antiparkinson medication use) were observed,
indicating consistent treatment effects across the strata
(all p > 0.370).
3.6. Safety-treatment-emergent adverse events
The occurrence of z 1 treatment-emergent EPS
individual symptoms (DIEPSS items) was signifi-

Table 4
Change from baseline to endpoint (LOCF) in DIEPSS total score
DIEPSS score Therapy N Baseline Endpoint Change Within group Therapy
p-valuea p-valueb
Mean S.D. Mean S.D. Median Mean S.D.
Total Olz 81 2.90 3.10 2.00 3.00 0.0 0.90 2.00 < 0.001 < 0.001
Hal 80 2.80 3.00 3.30 3.90 0.0 0.50 3.00 0.114
Parkinsonism Olz 81 2.50 2.70 1.70 2.50 0.0 0.80 1.70 < 0.001 0.001
Hal 80 2.30 2.50 2.60 3.30 0.0 0.30 2.50 0.274
Akathisia Olz 81 0.30 0.70 0.20 0.50 0.0 0.10 0.50 0.047 0.001
Hal 80 0.20 0.60 0.50 0.90 0.0 0.20 0.80 0.011
Dystonia Olz 81 0.10 0.30 0.10 0.30 0.0 0.00 0.20 1.000 0.489
Hal 80 0.10 0.60 0.10 0.40 0.0 0.00 0.50 1.000
Dyskinesia Olz 81 0.10 0.30 0.10 0.40 0.0 0.00 0.30 1.000 0.539
Hal 80 0.20 0.60 0.20 0.50 0.0 0.00 0.50 1.000
Overall severity Olz 81 0.90 0.70 0.70 0.70 0.0 0.20 0.60 < 0.001 0.014
Hal 80 1.00 0.90 1.00 0.90 0.0 0.10 0.90 0.708
N = total number of patients with a baseline and at least one postbaseline score, S.D. = standard deviation, Olz = olanzapine, Hal = haloperidol.
a
Within treatment group changes were tested with Wilcoxon signed-rank test.
b
Treatment group comparisons of change score distributions were tested with the Wilcoxon rank-sum test.
 T. Inada et al. / Schizophrenia Research 57 (2002) 227–238 233

Fig. 1. Treatment-emergent individual EPS symptoms at any time using DIEPSS (percent; z 10% in any olanzapine treatment group or
statistically significant).

Fig. 2. Incidence of treatment-emergent EPS syndromes at any time as determined by DIEPSS.

234 T. Inada et al. / Schizophrenia Research 57 (2002) 227–238
cantly greater ( p < 0.001) for haloperidol-treated
patients compared to olanzapine-treated patients
(Fig. 1). Akathisia, abnormal gait, bradykinesia, sia-
lorrhea, and tremor all occurred significantly more in
the haloperidol group ( p V 0.008).
For incidence of treatment-emergent EPS syn-
dromes determined by the DIEPSS at any time,
parkinsonism was significantly greater ( p = 0.005) in
the haloperidol group (18.8%) than in the olanzapine
group (3.2%) (Fig. 2). Importantly, by endpoint, all
treatment-emergent cases of parkinsonism with olan-
zapine had resolved, whereas 7.8% of haloperidol
patients had sustained treatment-emergent parkinson-
ism (Fig. 3). Although the incidence of akathisia at
any time was only 6.8% in olanzapine-treated patients
compared to 18.2% in haloperidol-treated patients, the
difference was not significant ( p = 0.029, one-sided).
For incidence of treatment-emergent EPS determined
by the DIEPSS at endpoint, no olanzapine patients
were observed to have either parkinsonism or akathi-
sia. In contrast, there were five (7.8%) haloperidol-
treated patients with parkinsonism and five (6.5%)
diagnosed with akathisia. However, there was not a
Fig. 3. Incidence of treatment-emergent EPS syndromes at endpoint as determined by DIEPSS.
statistical difference between the two treatment groups
for these EPS ( p = 0.030, one-sided). There was one
patient in each treatment group at endpoint with
dyskinesia and one patient having dystonia in the
olanzapine group.
4. Discussion
As expected for an atypical antipsychotic, olanza-
pine demonstrated significant improvement in the
primary EPS analysis, maximum change from baseline
total score, compared to the typical antipsychotic,
haloperidol. Also, improvement was observed in the
DIEPSS total score for olanzapine-treated patients on
all secondary analyses; however, the DIEPSS total
score might be weighted toward parkinsonism since
five items are parkinsonism symptoms. The significant
improvement in the overall severity item from baseline
to endpoint is consistent with olanzapine’s favorable
EPS profile. Similar to previous studies (Tollefson et
al., 1997a; Beasley et al., 1996; Sanger et al., 1999),
both parkinsonism and akathisia were significantly
 T. Inada et al. / Schizophrenia Research 57 (2002) 227–238
improved from baseline. Haloperidol-treated patients
continued to worsen by endpoint for these two symp-
toms even though they were taking significantly
greater amounts of anticholinergic medication. This
result is not surprising because it has been known that
typical antipsychotics trigger the appearance of, or
worsen, existing EPS (Gerlach, 1999; Glazer, 2000).
Patients treated with olanzapine experienced signifi-
cant reduction in the severity of two syndromes
(parkinsonism and akathisia) even though the patient
population had an average illness duration of 17.2
years and a current length of treatment (mostly with
typical drugs) of nearly 6 years. No substantial differ-
ences were observed in dyskinetic symptoms probably
because of this study’s short length. No significant
difference for dystonia was observed because the
baseline scores were very low in both treatment groups
and little change occurred during the study. This would
be expected in a patient population chronically treated
with antipsychotic drugs.
The estimated occurrence of EPS in patients tak-
ing typical antipsychotics is 50% to 90% (Casey and
Keepers, 1988). Over the last decade, patients taking
atypical antipsychotics have shown a much lower
risk for EPS (Casey, 1999). An analysis of three trials
that included 1796 patients treated with olanzapine
(5– 20 mg/day) and 810 patients treated with halo-
peridol (5 – 20 mg/day) showed that significantly
fewer olanzapine-treated patients experienced treat-
ment-emergent EPS adverse events of dystonia, par-
kinsonism, akathisia and dyskinesia as compared to
haloperidol-treated patients (Tran et al., 1997). Olan-
zapine-treated patients also were significantly im-
proved compared to haloperidol-treated patients on
parkinsonism and akathisia as measured by the
Simpson – Angus and Barnes Akathisia scales, res-
pectively. Significantly, fewer olanzapine-treated pa-
tients discontinued treatment because of EPS and
anticholinergic use was significantly greater in the
haloperidol group.
A recent prospective, open study found that
patients treated with olanzapine had fewer EPS com-
pared to not only haloperidol, but also to the atypical
antipsychotic, risperidone (Gómez et al., 2000). The
incidence of akathisia in olanzapine patients (2.8%)
was significantly lower ( p < 0.001) versus both hal-
operidol and risperidone patients (17.2% and 7.2%,
respectively). The olanzapine group also demonstra-
235
ted significantly lower ( p V 0.002) incidences of
hypertonia, hypokinesia and tremor versus the halo-
peridol and risperidone groups.
The likelihood of a patient on long-term typical
antipsychotic therapy to acquire tardive dyskinesia
(TD) at any time ranges from 20% to 50%, with the
elderly being at higher risk (Casey, 1999; Kane et al.,
1992). According to the evidence that suggests a
relationship between EPS and TD development (Inada
and Yagi, 1996; Kane et al., 1992, 1982; Jeste et al.,
1995; Saltz et al., 1991), treatment with olanzapine
would have, theoretically, a lower risk of TD than
haloperidol, as well as risperidone, based on the results
of this and previous studies (Tollefson et al., 1997a;
Tran et al., 1997; Gómez et al., 2000; Beasley et al.,
1999). Tollefson et al. (1997b) found in a long-term
study that the incidence of newly emergent TD was
significantly less ( p < 0.003) at endpoint for olanza-
pine-treated patients as compared to haloperidol-trea-
ted patients. Moreover, there have been few reports of
possible tardive dyskinesia onset during olanzapine
use (Ananth and Kenan, 1999; Herrán and Vázquez-
Barquero, 1999; Snoddgrass and Labbate, 1999).
A study by Matsuda et al. (1996) showed that
Asians with schizophrenia, as compared to Cauca-
sians, required lower doses of the atypical antipsy-
chotic clozapine to obtain similar levels of efficacy.
Another study that treated Asian patients with cloza-
pine found that parkinsonism, as measured by the
Simpson –Angus scale, did not improve (excluding
patients with no parkinsonian symptoms at baseline)
or worsened in the majority of those patients (Chong
et al., 1997). In contrast, this study shows that
olanzapine-treated Japanese patients improved signif-
icantly from baseline in the akathisia and parkinson-
ism item scores as well as on the total and overall
severity scores ( p < 0.001). The olanzapine group was
also significantly improved versus the haloperidol
group on the same EPS scores even though they were
on relatively lower doses of olanzapine compared to
Caucasian counterparts. Taken together, these results
suggest that olanzapine has a more favorable EPS
profile compared to the typical agents in Japanese
schizophrenic patients. These results might reasonably
be extended to other Asian populations but further
studies are needed to confirm this.
In summary, olanzapine-treated patients had a
much better EPS profile compared to haloperidol-
236 T. Inada et al. / Schizophrenia Research 57 (2002) 227–238
treated patients. The EPS profile is very similar to
what has been shown previously in trials conducted
primarily in Caucasians comparing olanzapine to
haloperidol. Olanzapine-treated patients from this
study received a lower mean modal dose than what
has been noted in studies consisting primarily of
Caucasians. This supports the possibility that Japa-
nese patients might benefit from lower doses of
antipsychotics than their Caucasian counterparts. In
conclusion, the EPS findings in this study suggest that
olanzapine is a safe treatment for chronically ill
schizophrenic patients in a Japanese population.
Acknowledgements
This study was sponsored by Eli Lilly and
Company.
Principal investigators in the Olanzapine 301E
Study Group:
. A. Aoba; St. Marianna University School of Med-
icine
. S. Endo; Nippon Medical School
. C. Hirotsu; Meisei University
. G. Ikawa; Sobu Hospital
. K. Kamijima; Showa University, School of Med-
icine
. K. Kosaka; Yokohama City University, School of
Medicine
. Y. Koshino; Kanazawa University, School of
Medicine
. T. Koyama; Hokkaido University, School of
Medicine
. Y. Kudo; Asakayama Hospital
. M. Kurihara; Clinic in Ministry of Finance Japan
. S. Kuroda; Okayama University Medical School
. Y. Machiyama; Umayabashi Hospital
. T. Mita; Iwate Medical University; School of
Medicine
. M. Murasaki; Kitasato University, School of
Medicine
. T. Nakajima; Bukkyo University
. Y. Nakane; Nagasaki University, School of Med-
icine
. S. Niwa; Fukushima Medical University
. J. Nomura; Suzuka Sakura Hospital
. C. Ogura; Ryukyu University, Faculty of Medicine
. K. Omori; Dokkyo University School of Medicine
. Late M. Saito; Kansai Medical University
. T. Sakai; Aino College
. J. Suzuki; Toho University, School of Medicine
. M. Takeda; Osaka University, School of Medicine
. A. Tamura; Hatano Kosei Hospital
. N. Tashiro; Kyushu University, Faculty of Medi-
cine
. M. Toru; Mental Clinic Ogikubo
. S. Ushijima; The Jikei University, School of
Medicine
. S. Watanabe; Kawasaki University of Medical
Welfare
. S. Yamagami; Kumeda Hospital
. T. Yamauchi; Saitama Medical School
. S. Yamawaki; Hiroshima University, Faculty of
Medicine
. F. Yoshimasu; Wakayama Medical College
Appendix A
A brief description of the DIEPSS subscales. Con-
tact Dr. Inada for detailed descriptions of each sub-
scale.
The severity of each item is rated from 0 (normal)
to 4 (severe).
(1) Gait: shuffling, slow gait. Evaluate the degree
of reduction in speed and step, decrease in pendular
arm movement, stooped posture and propulsion phe-
nomenon.
(2) Bradykinesia: slowness and poverty of move-
ments: Delay and/or difficulty in initiating and/or
terminating movements. Rate degree of poverty of
facial expression (mask-like face) and monotonous,
slurred speech, as well.
(3) Sialorrhea: excess salivation.
(4) Muscle rigidity: resistance to flexion and ex-
tension of upper arms. Rate cogwheeling, waxy flex-
ibility, lead-pipe rigidity and the degree of flexibility
of wrists, as well.
(5) Tremor: repetitive, regular (4 –8 Hz), and rhyth-
mic movements observed in the oral region, fingers,
extremities and trunk.
(6) Akathisia: subjective inner restlessness and re-
lated distress; awareness of the inability to remain
seated, restless legs, fidgety feelings, desire to move
constantly, etc. Rate increased motor phenomena
 T. Inada et al. / Schizophrenia Research 57 (2002) 227–238
(body rocking, shifting from foot to foot, stamping in
place, crossing and uncrossing legs, pacing around,
etc.), as well.
(7) Dystonia: symptoms induced by the hypertonic
state of muscles. Stiffness, twisting, and persistent
abnormal position of muscles observed in tongue,
neck, extremities, trunk, etc. Rate tongue protrusion,
torticollis, retrocollis, trismus, oculogyric crisis, Pisa
syndrome, etc.
(8) Dyskinesia: hyperkinetic abnormal movements.
Apparently purposeless, irregular, and involuntary
movements observed in face, mouth, tongue, extrem-
ities and/or trunk. Include choreic and athetoid move-
ments, but do not rate tremor.
(9) Overall severity: overall severity of extrapyr-
amidal symptoms.
References
Akritas, M.G., Arnold, S.F., Brunner, E., 1997. Nonparametric hy-
potheses and rank statistics for unbalanced factorial designs. J.
Am. Stat. Assoc. 92, 258 – 265.
Ananth, J., Kenan, J., 1999. Tardive dyskinesia associated with
olanzapine monotherapy. J. Clin. Psychiatry 60, 870.
Beasley Jr., C.M., Tollefson, G.D., Tran, P., Satterlee, W., Sanger, T.,
Hamilton, S., The Olanzapine HGAD Study Group, 1996. Olan-
zapine versus placebo and haloperidol: acute phase results of the
North American double-blind olanzapine trial. Neuropsycho-
pharmacology 14, 111 – 123.
Beasley, C.M., Dellva, M.A., Tamura, R.N., Morgenstern, H.,
Glazer, W.M., Ferguson, K., Tollefson, G.D., 1999. Randomised
double-blind comparison of the incidence of tardive dyski-
nesia in patients with schizophrenia during long-term treat-
ment with olanzapine or haloperidol. J. Psychiatry 174, 23 – 30.
Binder, R.L., Levy, R., 1981. Extrapyramidal reactions in Asians.
Am. J. Psychiatry 138, 1243 – 1244.
Breier, A., Hamilton, S.H., 1999. Comparative efficacy of olanza-
pine and haloperidol for patients with treatment-resistant schiz-
ophrenia. Biol. Psychiatry 45, 403 – 411.
Casey, D.E., 1992. What makes a neuroleptic atypical? In: Meltzer,
H.Y. (Ed.), Novel Antipsychotic Drugs. Raven Press, New York,
NY, pp. 241 – 251.
Casey, D.E., 1999. Tardive dyskinesia and atypical antipsychotic
drugs. Schizophr. Res. 35 (Suppl.), S61 – S66.
Casey, D.E., Keepers, G.A., 1988. Neuroleptic side effects: acute
extrapyramidal syndromes and tardive dyskinesia. In: Casey,
D.E., Christensen, A.V. (Eds.), Psychopharmacology: Current
Trends. Springer Verlag, Berlin, pp. 74 – 93.
Chong, S.-A., Tan, C.-H., Khoo, Y.-M., Lee, H.-W., Wong, K.-E.,
Ngui, F., Winslow, M., 1997. Clinical evaluation and plasma
clozapine concentrations in Chinese patients with schizophrenia.
Ther. Drug Monit. 19, 219 – 223.
237
Gerlach, J., 1999. The continuing problem of extrapyramidal symp-
toms: strategies for avoidance and effective treatment. J. Clin.
Psychiatry 60 (Suppl. 23), 20 – 24.
Gerlach, J., Peacock, L., 1995. New antipsychotics: the present
status. Int. Clin. Psychopharmacol. 10 (Suppl. 3), 39 – 48.
Glazer, W.M., 2000. Extrapyramidal side effects, tardive dyskinesia,
and the concept of atypicality. J. Clin. Psychiatry 61 (Suppl. 3),
16 – 21.
Gómez, J.C., Sacristan, J.A., Hernandez, J., Breier, A., Ruiz Car-
rasco, P., Anton Saiz, C., Fontova Carbonell, E., 2000. The safety
of olanzapine compared with other antipsychotic drugs: results
of an observational prospective study in patients with schizo-
phrenia (EFESO Study). Pharmacoepidemiologic study of olan-
zapine in schizophrenia. J. Clin. Psychiatry 61, 335 – 343.
Herrán, A., Vázquez-Barquero, J.L., 1999. Tardive dyskinesia asso-
ciated with olanzapine. Ann. Intern. Med. 131, 72.
Inada, T., 1996. Evaluation and Diagnosis of Drug-Induced Extrap-
yramidal Symptoms: Commentary on the DIEPSS and Guide to
its Usage. Seiwa Shoten, Tokyo.
Inada, T., Yagi, G., 1995. Current topics in tardive dyskinesia in
Japan. Psychiatry Clin. Neurosci. 49, 239 – 244.
Inada, T., Yagi, G., 1996. Current topics in neuroleptic-induced
extrapyramidal symptoms in Japan. Keio J. Med. 45, 95 – 99.
Inada, T., Matsuda, G., Kitao, Y., Nakamura, A., Miyata, R., Ina-
gaki, A., Koshiishi, M., Kanba, S., Yagi, G., 1996. Barnes aka-
thisia scale: usefulness of standardized videotape method in
evaluation of the reliability and in training raters. Int. J. Methods
Psychiatr. Res. 6, 49 – 52.
Inagaki, A., Inada, T., Fujii, Y., Yagi, G., 1998. Dose equivalence of
psychotropic drugs: Part 8. Dose equivalence of antiparkinso-
nian drugs. Jpn. J. Clin. Psychopharmacol. 1, 969 – 972.
Ishigooka, J., Inada, T., Miura, S., 2001. Olanzapine versus halo-
peridol in the treatment of patients with chronic schizophrenia:
results of the Japan multicenter, double-blind olanzapine trial.
Psychiatry Clin. Neurosci. (Jpn.) 55, 403 – 414.
Itoh, H., Yagi, G., Ogita, K., Otsuka, N., Sakurai, S., Tashiro, I.,
Suzuki, T., Kaizawa, S., Kamijima, K., Koga, Y., Miura, S.,
1977. International collaborative comparative study on the effi-
cacy and the safety of the antipsychotic drug treatment: 2nd
report: a study on the reliability of AIMS (NIMH). Annu.
Rep. Pharmacopsychiatr. Res. Found. 9, 218 – 225.
Jeste, D.V., Caligiuri, M.P., Paulsen, J.S., Heaton, R.K., Lacro, J.P.,
Harris, M.J., Bailey, A., Fell, R.L., McAdams, L.A., 1995. Risk
of tardive dyskinesia in older patients. Arch. Gen. Psychiatry 52,
756 – 765.
Jibiki, I., Kubota, T., Fujimoto, K., Sakamoto, H., Hasegawa, M.,
Furuta, H., Yamaguchi, N., 1993. Effective clinical response at
low plasma levels of haloperidol in Japanese schizophrenics with
acute psychotic state. Jpn. J. Psychiatry Neurol. 47, 627 – 629.
Kane, K.M., Mayerhoff, D., 1989. Do negative symptoms respond
to pharmacological treatment? Br. J. Psychiatry 155 (Suppl. 7),
115 – 118.
Kane, J.M., Woerner, M., Weinhold, P., Wegner, J., Kinon, B., 1982.
A prospective study of tardive dyskinesia development: prelimi-
nary results. J. Clin. Psychopharmacol. 2, 345 – 349.
Kane, J.M., Jeste, D.V., Barnes, T.R.E., Casey, D.E., Cole, J.O.,
Davis, J.M., Gualtieri, C.T., Schooler, N.R., Sprague, R.L.,
238 T. Inada et al. / Schizophrenia Research 57 (2002) 227–238
Wettstein, R.M., 1992. Tardive Dyskinesia: A Task Force Report
of the American Psychiatric Association. American Psychiatric
Association, Washington, DC.
Keepers, G.A., Clappison, V.J., Casey, D.E., 1983. Initial anticho-
linergic prophylaxis for neuroleptic-induced extrapyramidal
syndromes. Arch. Gen. Psychiatry 40, 1113 – 1117.
Lam, Y.W., Jann, M.W., Chang, W.-H., Yu, H.-S., Lin, S.-K., Chen,
H., Davis, C.M., 1995. Intra- and interethnic variability in re-
duced haloperidol to haloperidol ratios. Pharmacoepidemiology
35, 128 – 136.
Lane, H.-Y., Lin, H.-N., Hu, O.Y.-H., Chen, C.-C., Jann, M.W.,
Chang, W.-H., 1997. Blood levels of reduced haloperidol versus
clinical efficacy and extrapyramidal side effects of haloperidol.
Prog. Neuropsychopharmacol. Biol. Psychiatry 21, 299 – 311.
Matsuda, K.T., Cho, M.C., Lin, K.-M., Smith, M.W., Young, A.S.,
Adams, J.A., 1996. Clozapine dosage, serum levels, efficacy,
and side-effect profiles: a comparison of Korean – American
and Caucasian patients. Psychopharmacol. Bull. 32, 253 – 257.
Meltzer, H.Y., 1992. The mechanism of action of clozapine in rela-
tion to its clinical advantages. In: Meltzer, H.Y. (Ed.), Novel
Antipsychotic Drugs. Raven Press, New York, pp. 1 – 13.
Möller, H.-J., 1993. Neuroleptic treatment of negative symptoms in
schizophrenic patients: efficacy problems and methodological
difficulties. Eur. Neuropsychopharmacol. 3, 1 – 11.
O’Brien, P., 1984. Procedures for comparing samples with multiple
endpoints. Biometrics 40, 1079 – 1087.
Saltz, B.L., Woerner, M.G., Kane, J.M., Lieberman, J.A., Ma, J.,
Alvir, J., Bergmann, K.J., Blank, K., Koblenzer, J., Kahaner, K.,
1991. Prospective study of tardive dyskinesia incidence in the
elderly. JAMA 266, 2402 – 2406.
Sanger, T.M., Lieberman, J.A., Tohen, M., Grundy, S., Beasley
Jr., C., Tollefson, G.D. 1999. Olanzapine versus haloperidol
treatment in first-episode psychosis. Am. J. Psychiatry 156,
79 – 87.
SAS Institute, 1989. SAS/STAT User’s Guide, version 6, 4th edn.
vols. 1 and 2, SAS Institute, Cary, NC.
Snoddgrass, P.L., Labbate, L.A., 1999. Tardive dyskinesia from
risperidone and olanzapine in an alcoholic man. Can. J. Psychia-
try 44, 921.
Tollefson, G.D., Beasley, C.M., Tamura, R.N., Tran, P.V., Potvin,
J.H., 1997a. Blind, controlled, long-term study of the compara-
tive incidence of treatment emergent tardive dyskinesia with
olanzapine or haloperidol. Am. J. Psychiatry 154, 1248 – 1254.
Tollefson, G.D., Beasley Jr., C.M., Tran, P.V., Street, J.S., Krueger,
J.A., Tamura, R.N., Graffeo, K.A., Thieme, M.E. 1997b. Olan-
zapine versus haloperidol in the treatment of schizophrenia and
schizoaffective and schizophreniform disorders: results of an
international collaborative trial. Am. J. Psychiatry 154, 457 –
465.
Tran, P.V., Dellva, M.A., Tollefson, G.D., Beasley Jr., C.M., Potvin,
J.H., Kiesler, G.M. 1997. Extrapyramidal symptoms and toler-
ability of olanzapine versus haloperidol in the acute treatment of
schizophrenia. J. Clin. Psychiatry 58, 205 – 211.
Wood, A.J.J., Zhou, H.H., 1991. Ethnic differences in drug dispo-
sition and responsiveness. Clin. Pharmacokinet. 20, 350 – 373.
World Health Organization, 1993. The International Classification
of Mental and Behavioural Disorders: Diagnostic Criteria for
Research. WHO, Geneva, Switzerland.