. 61
( 68 .)


Akathisia scales did not differ between groups. And no treatment-emergent extra-
pyramidal adverse events differed between treatment groups.
A larger, olanzapine“valproate double-blind comparison has been published
(Tohen et al., 2002a). In this trial, inpatients were randomized to olanzapine
5“20 mg/day (starting dose 15 mg, n ¼ 125) or valproate 500“2500 mg/day (starting
dose 750 mg/day, n ¼ 123). YMRS improvement at 3 weeks was superior on
olanzapine (49%) compared to valproate (37%) (P<0.03). Response rates, defined
as improvement of at least 50% on the YMRS, were 54% on olanzapine versus 42%
for valproate (P ¼ 0.06). This study found an interaction (P ¼ 0.06) based on psychosis
status. Among patients with psychotic mania, YMRS improvement was similar on
olanzapine (44% reduction) and on valproate (41% reduction). However, among
362 R. W. Baker et al.

Table 16.1 Improvement from baseline Young Mania Rating Scale in psychotic versus
non-psychotic patients treated with olanzapine or a comparator

Duration Group Psychotic Non-psychotic Interaction

34% (n ¼ 38) 39% (n ¼ 32) P ¼ 0.880
3 weeks Olanzapine
19% (n ¼ 33) 17% (n ¼ 33)
44% (n ¼ 62) 54% (n ¼ 63) P ¼ 0.061
3 weeks Olanzapine
41% (n ¼ 51) 34% (n ¼ 74)
52% (n ¼ 33) 51% (n ¼ 21) P ¼ 0.110
4 weeks Olanzapine
13% (n ¼ 28) 44% (n ¼ 28)
Olanzapine þ lithium or valproate 54% (n ¼ 70) 62% (n ¼ 150) P ¼ 0.440
6 weeks
Placebo þ lithium or valproate 45% (n ¼ 38) 39% (n ¼ 76)

non-psychotic individuals, mania rating improvement was superior on olanzapine
(54%) than on valproate (34%, P<0.001). No interaction was found (P ¼ 0.95)
based on manic versus mixed episode, suggesting the relative performance of
olanzapine and divalproex is similar across this diagnostic dimension. Hamilton
Depression rating scale reduction did not differ significantly between the groups
(33% improvement on olanzapine and 26% on valproate). Discontinuation rates
were 31% on olanzapine and 36% on divalproex. Though treatment-emergent
tremor was more common on olanzapine (10% versus 3%), mean changes in
extrapyramidal symptom ratings (Simpson-Angus and Barnes Akathisia scales)
did not differ between groups. Tables 16.1 and 16.2 summarize improvement in
manic symptoms seen in olanzapine- or comparator-treated patients with psychotic
versus non-psychotic and manic versus mixed manic symptoms, respectively.
Given the question of worsening of depression with some antipsychotic agents,
the findings of a recent trial of olanzapine for bipolar I depression are of interest
(Tohen et al., 2003b). To our knowledge no other study has been conducted of
antipsychotic monotherapy for bipolar depression. In this 8-week, double-blind
study, mean Montgomery-Asberg Depression symptom rating improvements were
significantly greater among 370 patients treated with olanzapine 5“20 mg/day (39%
improvement) than among 377 placebo-treated patients (30% improvement from
baseline). Of note, this study also included a group treated with olanzapine 6 or 12
mg/day plus fluoxetine 25 or 50 mg/day; depressive symptoms improved signifi-
cantly more in this group (55%) than in either of the other groups, while the rate of
switch to mania was not increased. Eight-week completion rate was significantly
higher on olanzapine (48%) than placebo (39%), and completion rate on the
olanzapine“fluoxetine combination (64%) was significantly greater than in both
other groups. Extrapyramidal side-effects, measured by Simpson-Angus and
Abnormal Involuntary Movement scales, did not differ among the groups.
363 Use of atypical antipsychotic agents

Table 16.2 Improvement from baseline Young Mania Rating Scale in manic versus mixed
patients treated with olanzapine or a comparator

Duration Group Manic Mixed Interaction

37% (n ¼ 58) 28% (n ¼ 12) P ¼ 0.998
3 weeks Olanzapine
16% (n ¼ 54) 25% (n ¼ 12)
49% (n ¼ 69) 48% (n ¼ 56) P ¼ 0.950
3 weeks Olanzapine
36% (n ¼ 74) 38% (n ¼ 52)
53% (n ¼ 31) 50% (n ¼ 23) P ¼ 0.706
4 weeks Olanzapine
26% (n ¼ 33) 30% (n ¼ 23)
Olanzapine þ lithium or valproate 60% (n ¼ 99) 58% (n ¼ 121) P ¼ 0.708
6 weeks
Placebo þ lithium or valproate 45% (n ¼ 60) 34% (n ¼ 54)

Finally, initial results are available of a 1-year comparison of olanzapine 5“20
mg/day and lithium (mean level 0.77 mmol/l) for relapse prevention among
bipolar patients stabilized on an open combination of olanzapine and lithium
and a 1-year comparison of olanzapine to placebo for relapse prevention among
manic patients openly stabilized on olanzapine (Tohen et al., 2004). The lithium
comparison study however enrolled very few (<10%) rapid-cycling patients. On
the other hand, rapid-cycling patients constituted roughly half of the placebo
comparison study. Overall relapse rates were modestly higher among rapid-cycling
patients, but olanzapine was significantly better than placebo in delaying relapse
for both rapid and non-rapid cycling patients. To our knowledge, no controlled
maintenance data or information on slowing the course of rapid cycling are
available for the other agents reviewed in this chapter.

Two double-blind studies have been reported demonstrating efficacy of quetiapine
as adjunctive treatment for acute mania. The first (DelBello et al., 2002) was a 6-week
trial in adolescent (aged 12“18) inpatients in acute manic or mixed episodes,
receiving open treatment with valproate 20 mg/kg per day. They were randomized
to adjunctive treatment with valproate 450 mg/day (n ¼ 15) or placebo (n ¼ 15).
Repeated measures of analysis of variance of completers only found greater improve-
ment of YMRS in the quetiapine group at two of the seven ratings (days 21 and 42).
Information available to date does not state whether the groups differed in intent-
to-treat analysis of mania ratings. Forty percent of patients receiving quetiapine and
13% on placebo discontinued treatment during the 6-week study. Changes in
extrapyramidal symptom ratings did not differ between treatment groups.
The second, larger trial (Sachs et al., 2004) was conducted in adult inpatients with
acute mania receiving open treatment with lithium or divalproex. It appears that the
364 R. W. Baker et al.

initiation of mood-stabilizer treatment did not necessarily antedate starting quetia-
pine or placebo. Patients were randomized to 3 weeks of adjunctive treatment with
quetiapine 200“800 mg/day (n ¼ 91) or placebo (n ¼ 100); significantly greater
improvement in YMRS from baseline occurred in the adjunctive quetiapine group
(44% reduction) than in the placebo group (32%) (P ¼ 0.021). Response, defined as
50% or greater YMRS improvement, was more common in patients taking adjunc-
tive quetiapine (54.9%) than placebo (32.6%, P ¼ 0.005). Montgomery-Asberg
depression ratings improved from baseline in both groups (25% reduction on
quetiapine versus 20% on placebo, difference not significant) and the number of
patients with treatment-emergent depression also did not differ significantly
between groups (quetiapine 17%, placebo 14%). Unfortunately, the report does
not shed light on subgroup efficacy; the authors do not mention whether improve-
ment is consistent irrespective of presence of psychosis, and both rapid-cycling
patients and those in current mixed episodes were excluded from the trial. Drop-
out rate did not differ significantly between placebo-treated (51%) and quetiapine-
treated patients (39%) for this 3-week adjunctive-treatment study. Changes in
extrapyramidal symptom ratings (Simpson-Angus and Barnes Akathisia scales)
were modest and apparently did not differ between the groups.

At this writing, one double-blind trial of ziprasidone in bipolar disorder has been
reported (Keck et al., 2003a). This 3-week, placebo-controlled study in patients with
acute manic or mixed episodes found superior YMRS improvement on ziprasidone
80“160 mg/day (46% reduction, n ¼ 131) than on placebo (29%, n ¼ 64)
(P<0.005). Response rates, defined as 50% or greater improvement in YMRS scores,
were 50% and 35% for ziprasidone and placebo groups, respectively (P<0.05). Raw
improvement in mania ratings during ziprasidone treatment was similar among
patients in classic manic (n ¼ 85) or mixed episodes (n ¼ 46). However, reports to
date do not address whether effect size, or improvement relative to placebo, differs
between manic and mixed subgroups. Similarly, relative efficacy for psychotic
versus non-psychotic subjects and rapid versus non-rapid-cycling patients was not
addressed in the first manuscript. Discontinuation rate in the ziprasidone group was
46% versus 56% for placebo, P-value not provided. Changes in depressive symp-
toms are not addressed and apparently were not assessed in this trial.
Extrapyramidal side-effect ratings (Simpson-Angus and Barnes Akathisia Rating
Scales, and Abnormal Involuntary Movement Scale) did not differ significantly
between ziprasidone and placebo groups. However, treatment-emergent events
reported more frequently on ziprasidone than on placebo included hypertonia
(11% versus 3%) and akathisia (11% versus 6%). Tests of statistical significance
were not reported for these adverse events.
365 Use of atypical antipsychotic agents

Aripiprazole is the most recently introduced antipsychotic agent, and at least two
controlled monotherapy trials have been conducted in acute mania. One of these
had positive findings (Keck et al., 2003b). Among hospitalized patients with acute
mania, mean YMRS improved from baseline to the 3-week endpoint significantly
more frequently among patients treated with aripiprazole (29% reduction,
n ¼ 130, mean dose 27.9 mg/day) compared to placebo (11%, n ¼ 132).
Depression ratings significantly improved in aripiprazole-treated patients versus
depression (clinical global Impression-Bipolar severity of Depression). The
authors do not address any of the potentially important response by subgroup
interactions, that is, based on rapid cycling, psychosis, or mixed state status. Drop-
out rates were high in this 3-week study (58% on aripiprazole and 79% on
placebo). Measurements of extrapyramidal symptoms (Simpson“Angus and
Barnes Akathisia scales) worsened significantly on aripiprazole compared to
placebo. Extrapyramidal adverse events reported more commonly in the aripipra-
zole than placebo groups included akathisia (11% versus 2%), tremor (6% versus
3%), and increased salivation (6% versus 1%), but it is not reported whether any of
these findings were statistically significant.

Use of atypical antipsychotic medications in mania: psychotic and
non-psychotic patients
A diverse array of medications has evidence of usefulness in mania, including
lithium, anticonvulsants, antipsychotics, atypical antipsychotics, benzodiazepines,
and calcium channel blockers. There is little information on the relative effects of
such treatments, including whether a specific medication differentially targets a
subset of manic symptoms. Nevertheless, in the case of antipsychotic agents, clinical
guidelines, unsurprisingly, are particularly supportive of their use for patients with
psychotic mania (and in some cases for more agitated or severely ill patients)a (Sachs
et al., 2000, American Psychiatric Association, 2002; Baldessarini, 2002).
A secondary analysis was performed for the double-blind studies of olanzapine in
acute mania to evaluate effectiveness for psychotic and non-psychotic subgroups, as

There is some evidence that, at least in the case of olanzapine, usefulness extends beyond more severely ill
patients. In a previously unpublished secondary analysis of a 3-week olanzapine“divalproex monotherapy
comparison for acute mania (Tohen et al., 2002a), treatment groups were divided into more and less
severely ill groups by a median split of baseline YMRS. Among olanzapine-treated patients, baseline to
endpoint improvement was similar in more and less severely ill patients. As compared to those receiving
divalproex, improvement was similar in the more severely ill cohort, but superior on olanzapine among
less ill patients. In that all patients in this study were hospitalized and had bipolar I disorder, the less
severely ill patients probably still had relatively severe symptoms. Nevertheless, these data suggest that the
usefulness of olanzapine, at least, extends beyond the sickest patients.
366 R. W. Baker et al.

defined by baseline diagnosis confirmed by Structured Clinical Interview for
Diagnostic and Statistical Manual IV using Diagnostic and Statistical Manual of
Mental Disorders, 4th edn (DSM-IV): American Psychiatric Association (1994)
criteria. These data are not available for two of the trials (Berk et al., 1999; Zajecka
et al., 2002). The principal intent of these analyses was to seek an interaction between
diagnostic subgroup and clinical outcome, that is, is efficacy better among psychotic
patients, thereby suggesting that antimanic effects derive from antipsychotic proper-
ties? The answer appears to be no.


American Psychiatric Association (1994). Diagnostic and Statistical Manual of Mental
Disorders, 4th edn (DSM-IV). Washington, DC: American Psychiatric Association.
American Psychiatric Association (2002). Practice guideline for the treatment of patients with
bipolar disorder (revision). Am. J. Psychiatry, 159 (suppl. 4), 1“50.
Baldessarini, R. J. (2002). Treatment research in bipolar disorder: issues and recommendations.
CNS Drugs., 16, 721“9.
Berk, M., Ichim, L., and Brook, S. (1999). Olanzapine compared to lithium in mania:
a double“blind randomized controlled trial. Int. Clin. Psychopharmacol., 14, 339“43.
Brotman, M. A., Post, R. M., Fergus, E. L., and Leverich, G. S. (2000). High exposure to neuro-
leptics in bipolar patients: a retrospective review. J. Clin. Psychiatry, 61, 68“72.
Calabrese, J. R., Kimmel, S. E., Woyshville, M. J., et al. (1996). Clozapine for treatment-
refractory mania. Am. J. Psychiatry, 153, 759“64.
DelBello, M. P., Schwiers, M. L., Rosenberg, H. L., et al. (2002). A double-blind, randomized,
placebo-controlled study of quetiapine as adjunctive treatment for adolescent mania. J. Am.
Acad. Child Adolesc. Psychiatry, 41, 1216“23.
Garfinkel, P. E., Stancer, H. C., and Persad, E. (1980). A comparison of haloperidol, lithium
carbonate and their combination in the treatment of mania. J. Affect. Disord., 2, 279“88.
Glazer, W. (2000a). Review of incidence studies of tardive dyskinesia associated with typical
antipsychotics. J. Clin. Psychiatry, 61, 15“20.
Glazer, W. (2000b). Extrapyramidal side effects, tardive dyskinesia, and the concept of atypi-
cality. J. Clin. Psychiatry, 61, 16“21.
Glazer, W. M. (2000c). Expected incidence of tardive dyskinesia associated with atypical anti-
psychotics. J. Clin. Psychiatry, 61, 21“6.
Goodwin, F. K. and Jamison, K. R. (1990). Manic“Depressive Illness. New York, NY: Oxford
University Press.
Green, A. I., Tohen, M., Patel, J. K., et al. (2000). Clozapine in the treatment of refractory
psychotic mania. Am. J. Psychiatry, 157, 982“6.
Hirschfeld, R., Keck, P. E., Karcher, K., Kramer, M., and Grossman, F. (2004). Rapid antimanic
effect of risperidone monotherapy: a 3-week multicenter, double-blind, placebo-controlled
trial. Am. J. Psychiatry, 161, 1057“65.
367 Use of atypical antipsychotic agents

Kapur, S., and Seeman, P. (2001). Does fast dissociation from dopamine d(2) receptors explain
the action of atypical antipsychotics? A new hypothesis. Am. J. Psychiatry, 158, 360“9.
Keck, P. E., Jr., Versiani, M., Ptkin, S. et al. (2003a). Ziprasidone in the treatment of acute bipolar
mania. A 3-week, placebo-controlled, double-blind, randomized trial. Am. J. Psychiatry,
160, 741“8.
Keck, P. E., Jr., Marcus, R., Tourkodimitris, S., et al. (2003b). Aripiprazole versus placebo in
acute mania. Am. J. Psychiatry, 160, 1651“8.
Khanna, R., Das, A., and Damodaran, S. S. (1992). Prospective study of neuroleptic-induced
dystonia in mania and schizophrenia. Am. J. Psychiatry, 149, 511“13.
Koukopoulos, A., Reginaldi, D., Laddomada, P., et al. (1980). Course of the manic depressive
cycle and changes caused by treatment. Pharmacopsychiatria, 13, 156“67.
Krakowski, M., Czobor, P., and Volavka, J. (1997). Effect of neuroleptic treatment on depressive
symptoms in acute schizophrenic episodes. Psychiatry Res., 71, 19“26.
Morgan, H. G. (1972). The incidence of depressive symptoms during recovery from hypomania.
Br. J. Psychiatry, 120, 537“9.


. 61
( 68 .)