Psychosis in advanced

Parkinson’s disease:
Treatment with ondansetron, a 5-HT3 receptor antagonist
Joseph Zoldan, MD; Gali Friedberg, MD; Myron Livneh, MD; and Eldad Melamed, MD

Article abstracMPsychosis, linked to chronic levodopa and other antiparkinsonian drug treatments, is a common
and incapacitating complication of advanced Parkinson’s disease (PD). The psychosis may be due, in part, to
overstim- ulation of central serotonergic (5-HT) receptors. We treated 16 PD patients who had psychosis of 6 to 60
months’ dura- tion with the 5-HT3 receptor antagonist ondansetron (12 to 24 mg daily) in an open-label, short-term (4
to 8 weeks) trial. There was marked to moderate improvement (p < 0.01) in measures of visual hallucinations,
paranoid delusions, confusion, and the associated global functional impairment in all but one of the patients, and there
was moderate im- provement in the Brief Psychiatric Rating Scale and the Nurse’s Observation Scale for Inpatient
Evaluation; the Mini- Mental State Examination scores remained unaltered. Ondansetron did not cause any worsening
in basic PD symp- toms or levodopa efficacy and was well tolerated with no major side effects. Our study suggests
that pharmacologic blockade of central 5-HT receptors may become a strategy to attenuate PD psychosis without
inducing motor deteriora- tion or suppression of antiparkinsonian action of levodopa, and it lends support to the
hypothesis that serotonergic mechanisms are pathogenetically important in the emergence of psychosis in PD.
NEUROLOGY 1995;45:1305-1308

Psychosis, manifested mainly by visual halluci-
nosis, paranoid delusions, and confusion, is a com-
mon complication in patients with advanced Par-
kinson’s disease (PD); it can be causatively linked
to chronic levodopa, but also to other antiparkin-
sonian drug therapies-. 1 2 The psychosis is a marker
for illness deterioration,1-° an important motive for
placement of patients in nursing homes," and the
most significant limiting factor that curtails phar-
macologic attempts to improve motor function. 4
Treatment is currently unsatisfactory; reduction or
discontinuation of levodopa or its coadministration
with dopamine-blocking neuroleptics may abolish
or attenuate the psychotic phenomena, but these
strategies are usually associated with rapid and
sometimes life-risking worsening of motor disabil-
ity.' Clozapine, an atypical neuroleptic agent, may
be beneficial but frequently causes intolerable ad-
verse reactions such as lethargy and leukopenia.4 s
The cause for the emergence of parkinsonian
psychosis is not clear. The dopaminergic
hypothesis suggests that the psychosis is generated
by exces- sive stimulation of supersensitive
postsynaptic dopaminergic receptors in cortical and
limbic re- gions by dopamine formed from
exogenous levo- dopa.2 *• Another possibility is
that the psychotic phenomena are due additionally
or exclusively to
impaired serotonergic (5-HT) mechanisms."" There
are multiple direct and indirect interactions among
dopaminergic and serotonergic neurons, and it is
possible that excess dopamine formed from levo-
dopa may amplify 5-HT release in brain. In addi-
tion, exogenous levodopa can be inadvertently
taken up into 5-HT nerve terminals in the parkin-
sonian striatum, limbic system, and cortex and con-
verted therein to dopamine by the enzyme aromatic
amino acid decarboxylase. This enzyme catalyses
the conversion both of levodopa to dopamine in
dopaminergic nerve endings and of 5-hydroxy-
tryptophan to 5-HT in serotonergic nerve endings.•
In rodents, acute and chronic intraperitoneal injec-
tions of levodopa caused decreases of 5-HT and
ele- vations of its maj or metabolite, 5-hydroxyin-
dole acetic acid (5-HIAA), and the ratio of 5 -
HIAA/5-HT in striatum and cortex, indicating en-
hanced 5-HT release from storage vesicles (Mela-
med et al, in preparation). Therefore, parkinsonian
psychosis may be a result, in part, of nonphysio-
logic flushing out of serotonin from 5-HT nerve
ter- minals by the formed dopamine, leading to
over- stimulation of limbic and cortical
postsynaptic 5- HT receptors. The antipsychotic
effect of dopamine- blocking neuroleptics may be
partly due to nonse- lective inhibition of 5-HT
receptors. If true, phar-

From the Department of Neurology, Beilinson Medical Center, the Felsenatein Research Institute, and Geha Psychiatric Hospital, Petah Tiqva, and the
Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel.
Supported in part by the National Parkinson Foundation, Miami, FL.
Presented in part at the 46th annual meeting of the American Academy of Neurology, Washington, DC, May 1994.
Received September 19, 1994. Accepted in final form December 6, 1994.
Address correspondence and reprint requests to Dr. Eldad Melamed, Department of Neurology, Beilineon Medical Center, Petah Tiqva, larael 49100.

July ices NEUROLOGY 45 isos

 Vt8UAL MALLUC04ATtON8
Figure. Effect o[ondan- setron on presente and severit y (grades 2 to 4) of visual hallucinations, paranoid delusions, con- fusion, and associated
macologic blockade of 5-HT receptors ahould atten-
uate the psychosis without causing worsening of
parkinsonian signs.
Ondansetron is a novel, selective 5-HT3 receptor
antagonist used as an antiemetic in cancer patients
receiving chemotherapy.' Systemically adminis-
tered ondansetron can penetrate from the circula-
tion into brain parenchyma ' l and may be
beneficial as an antipsychotic drug in
schizophrenic pa-
tients.l2 We therefore examined the possible effi-
cacy, safety, and tolerability of treatment with on-
dansetron in a short-term, open-label trial in pa-
tients with PD who developed psychosis.
Methods. Sixteen PD patients (11 men and five women)
were included. Mean age was 72 (range, 64 to 88), and
mean duration of disease was 7.2 years (range, 1 to 14)
and of levodopa treatment 5.6 years (range, 1 to 13). The
1808 NEUROLOGY 45 July 1996
Hoehn and Yahr stage ranged from 2 to 4; all had levo-
dopa-responsive PD and 12 of 16 had response fluctua-
tions. They received levodopa at a mean daily dose of
493 mg (range, 250 to 750 mg, with carbidopa). Ten
patients were also treated with deprenyl, three with
bromocrip- tine, and three with amantadine; none were
taking anti- cholinergic agents. Patients had developed
symptoms of psychosis 6 to 60 months (mean, 15) before
entering the trial and 0.5 to 8 years (mean, 4.5) after
starting levo- dopa therapy. All suffered from frequent
visual halluci- nations and various degrees of confusion,
and nine had paranoid delusions. They previously
underwent re- peated, but unsuccessful, attempts to
attenuate the psy- chotic phenomena by changes in their
drug treatments. All patients or spouses gave informed
consent to partici- pate in the study.
The patients continued to receive their previously de-
termined optimal antiparkinsonian therapy throughout
the trial. Each patient was evaluated at baseline, before
the initiation of ondansetron therapy. They were scored
on the Unified Parkinson’s Disease Rati ng Scale
(UPDRS), the Folstein Mini-Mental State Examination
(MMSE), the Brief Psychiatric Rating Scale (BPRS),' 3
and the Nurse’s Observation Scale for Inpatient Evalua-
tion (NOSIE).'4 In addition, presence and severity of vi-
sual hallucinations, paranoid delusions, and confusion
were rated separately in each patient on a scale of 1 to 4
(1—absent; 2—mild; 3—moderate; 4—severe) based on
combinations of frequency, content, presence or absence
of insight, attitude, agitation, aggression, and ability to
function. Global functional impairment was estimated
mainly from family reports (1—absent; 2—mild; 3—
mod- erate, with preoccupation, withdrawn, lack of
initiation and motivation; 4—severe, grossly disabled,
considered for transfer away from home to a nursing
institution). Patients then received oral ondansetron at
an initial daily dose of 4 to 8 mg. Dosage was increased
by incre- ments of 4 to 8 mg per week up to a total daily
mainte- nance dose of 12 to 24 mg (mean, 18) taken in
two doses (one in the morning and one in the evening).
Patients were reevaluated after a period of 4 to 8 weeks
(5.8 + 1.9) of stable dose administration.
Results. Treatment with ondansetron was benefi-
cial in all but one of the patients (figure). Visual
hal- lucinations showed the most marked
improvement. Before treatment, these phenomena
were severe in 11 and moderate in five patients.
Following treat- ment, the hallucinations completely
or almost totally disappeared in 14, partially
improved in one, and re- mained unchanged in one
patient (figure). Paranoid delusions were present in
nine patients (five severe and four moderate).
Ondansetron induced marked improvement with total
or almost complete disap- pearance of these
phenomena in four, was partially beneficial in four,
and had no effect in one patient (figure). Various
degrees of pretreatment confusion were present in all
patients (severe in 11, moderate in four, and mild in
one). Following ondansetron ad- ministration, there
was complete remission of confu- sion in four and
partial improvement in the remain- ing 12 patients
(figure). Functional impairment was severe in 12 and
moderate in one patient. After treatment with
ondansetron, function improved markedly in eight and
moderately in four patients (figure). In one patient (the
same patient who showed no change in severity of
visual hallucina- tions and paranoid delusions),
impaired function re- mained unaltered. There were
moderate, significant improvements in the BPRS
and NOSIE scales fol- lowing ondansetron treatment
(table). The beneficial effect of ondansetron was
similar in patients with and without response
fluctuations. The patient whose symptoms did not
respond to ondansetron was not clinically different
from the other subjects, and we have no explanation
for drug failure in this case.
All patients had evidence of cognitive impairment
as shown by the baseline MMSE scale. There was no
change, either improvement or worsening, in the
cognitive function induced by ondansetron adminis-
tration (table). More important, treatment with the
5-HT antagonist did not cause any worsening of the
parkinsonian motor signs and symptoms and the
administration. This suggests that intellectual de-
Table. Effect of treatment with ondansetron on
motor, mental, and psychotic measures in patients
with Parkinson’s disease
RangeMeaO
Unified Parkinson’s Disease Rating Scale Before treatment24-4735.0
After treatment24-4835.0 8.9
8.7
Mini-Mental State Examination
Before treatment7-2417.35.4
After treatment9-2718.45.8
Brief Psychiatric Rating Scale
Before treatment31-5039.55.2
After treatment19-4330.4*8.6
Nurse’s Observation Scale for Inpatient Evaluation
Before treatment41-7150.810.2
After treatment10-5741.0†10.1
N = 16 patients; * p < 0.01 and } p < 0.05 (paired t test).
“offi’-period UPDRS measure remained unaltered
(table). Also, ondansetron did not counteract the effi-
cacy of levodopa, since in patients with response
fluctuations, the duration of drug-induced daily ‘on"
hours did not change (data not shown). Although
mood was not quantitatively evaluated in this study,
we were not impressed that there was either new
development or exacerbation of depression. Patients
and families often reported an anxiolytic effect of on-
dansetron. The drug was generally well tolerated,
and there were no major side effects except for in-
creased constipation (eight of 16 patients) and occa-
sional mild headaches (five of 16). None of the pa-
tients withdrew from the study because of adverse
reactions. After discontinuing ondansetron, symp-
toms of psychosis dzd not recur for a period of 8
weeks of follow-up in four of the 15 patients who had
a response. We have no explanation for this interest-
ing phenomenon. In the remaining 11 patients who
had a beneficial antipsychotic effect, phenomena re-
turned to the initial pretreatment severity, not im-
mediately, but within 10 to 14 days after cessation of
ondansetron.
Discussion. Our preliminary study shows that on-
dansetron is beneficial in treating psychosis in pa-
tients with advanced PD. The most marked im-
provement occurred in the visual hallucinations,
but there was also reduction of the other psychotic
phenomena, including paranoid delusions and
con- fusion, and improvement in general function.
There was less impressive, though significant,
improve- ment in the BPRS and NOSIE
measures. The drug was generally very well
tolerated and caused no major side effects.
Most of our patients showed evidence of pre-ex-
isting cognitive dysfunction of varying severity.
There is a strong correlation between the presence
of dementia and emergence of psychosis in PD pa-
tients,° 3 but there was no change (improvement or
deterioration) in cognition following ondansetron
July 1996 NEUROLOGY 45 1307
terioration in PD is not due to impaired 5-HT
mechanisms and that the psychosis and dementia
are independent markers of illness progression.
Unlike dopamine-blocking antipsychotic drugs,
ondansetron did not worsen the basic parkinsonian
motor signs and symptoms and did not counteract
the efficacy of levodopa. This indicates that it does
not block dopaminergic receptors or interfere with
absorption of levodopa or its entry into the brain,
and that its effect in PD psychosis is exclusively
mediated via central inhibition of 5-HT receptors.
Current strategies utilized to reduce PD psychosis,
eg, levodopa dose reduction or addition of
neurolep- tics, mainly suppress dopaminergic
transmission and are associated with worsening,
sometimes even catastrophic, of motor 4
Our
findings raise the hope that this type of psychosis
may be amenable to pharmacologic manipulations
of non- dopaminergic mechanisms and this
treatment may thus avoid deterioration in PD or
inhibition of an- tiparkinsonian drug efficacy.
This study lends support to the theory that
parkinsonian psychosis, and particularly the visual
hallucinations, may be due, in part, to overstimula-
tion of serotonergic receptors in the brain.* 7 On-
dansetron is a selective 5-HT3 receptor antago-
nist' o that is used to block cisplatinum-induced
vomiting by inhibiting 5-HT3 receptors in the eme-
sis center within the area postrema.' 5 5-HT3 recep-
tors are present in the central nervous system, par-
ticularly in the entorhinal and limbic areas,' 6 and
their blockade by ondansetron may be responsible
for its antipsychotic effect. Recent clinical trials
with 5-HT3 antagonists did not show therapeutic
benefit in schizophrenic patients. 17 This indicates
that different mechanisms may be responsible for
psychosis in PD and schizophrenia.
This drug is presently quite expensive, which
may curtail widespread utilization. (Note from
Edi- tor-in-Chief: “The wholesale price for 12 to
24 mg per day, in the United States, is
approximately
$800 to $1,600 per month.”) However, if the pres-
ent concept is verified, it may encourage develop-
ment of less expensive and more potent 5-HT3 an-
tagonists and antagonists for other 5-HT receptor
subtypes (eg, 5-HT2) with even better penetration
profiles from the circulation through the blood-
brain barrier.'
A controlled study should be done to determine
long-term efficacy of ondansetron in PD psychosis.
Success of this approach may greatly improve the
quality of life of afflicted patients, prevent or delay
their transfer from home to nursing institutions,
1308 NEUROLOGY 45 July lfl96
and perhaps enable increases of levodopa dosages
and the addition of other antiparkinsonian drugs to
improve motor impairment without risking exacer-
bation of psychosis.
References
1. Tanner CN1, Vogel C, Goetz CG, Klawans HL. Hallucina-
tions in Parkinson’s disease: a population study. Ann Neurol
1983;14:136-140.
2. Fahn S. Adverse effects of levodopa. In: Olanow CW, Lieber-
man AN, eds. The scientific basis of the treatment of Parkin-
son’s disease. New Jersey: Parthenon Publishing Group,
1992:89-112.
3. Goetz CG, Stebbins GT. Risk factors for nursing home
place- ment in advanced Parkinson’s disease. Neurology
1993; 43:2227-2229.
4. Friedman JH, The management of the 'levodopa psychosis.
Clin Neuropharmacol 1991;14:283-295.
5. Factor SA, Brown D, Molho ES, Podskalny GD. Clozapine: a
2-year open trial in Parkinson’s disease patients with psy-
chosis. Neurology 1994;44:544-546.
6. Goetz CG, Tanner CM, Klawans HL. Pharmacology of hallu-
cinations induced by long-term drug therapy. Am I Psychia-
try 1982;139:494-497.
7. Nausieda PA, Tanner CM, Klawans HL. Serotonergically
ac- tive agents in levodopa-induced psychiatric toxicity
reac- tions. Adv Neurol 1983;37:28-82.
8. Melamed E, Hefti F, Wurtman RJ. Non-aminergic striatal
neurons convert exogenous L-dopa to dopamine in parkin-
sonism. Ann Neurol 1980;8:558-563.
9. Melamed E, Hefti F, Wurtman RJ. L-dopa and L-5-hydroxy-
tryptophan decarboxylase activities in rat striatum: effect of
selective destructions of dopaminergic and serotonergic in-
puts. J Neurochem 1980;34:1753-1756.
10. Marty M, Pouillart P, Scholl S, et al. Comparison of the 5-
hydroxytryptamine, (serotoninJ antagonist ondansetron (GR
38032F) with high-dose metoclopramide in the control of cis-
platin-induced emesis. N Engl J Med 1990;322:816-821.
11 Rogers H, Abrams CJ, Kilpatrick GJ. Ex vivo binding of 5-
HT3 antagonists to 5-HT3 receptors in the rat entorhinal
cortex and colon [abstract]. Br I Pharmacol 1994;112:306P.
12. Hagan RM, Kilpatrick GJ, Tyers MB. Interactions between 5-
HT3 receptors and cerebral dopamine function: implica- tions
for the treatment of schizophrenia and psychoactive substance
abuse. Psychopharmacology (Berl) 1993;112:568- 575.
13. Overall JE, Gorham DR. The Brief Psychiatric Rating Scale.
Recent developments in ascertainment and scaling. Psy-
chopharmacol Bull 1988;24:97-103.
14. Honigfield G, Gillis R, Kiett CJ. NOSIE-30: a treatment-
sen- sitive ward behavior scale . Psychiatry Reports 1966;
19:180-182.
15. Aapro MS. 5-HT3 receptor antagonists: an overview of
their present status and future potential in cancer therapy-
induced emesis. Drugs 1990;42:551-568.
16. Kilpatrick GJ, Jones BJ, Tyers MB. Identification and dis-
tribution of 5-HT3- receptors in rat brain using radioligand
binding. Nature 1987;330:746-748.
17. Newcomer JW, Faustman WO, Ziparsky RB, Csernamsky
JG. Zacopride in schizophrenia: a single-blind serotonin type
3 antagonist trial. Arch Gen Psychiatry 1992;49:751-752.
18. Gerlach J. New antipsychotics: classification, efficacy, and
adverse effects. Schizophr Bull 1991;17:289-309.
Psychosis in advanced Parkinson's disease: Treatment with ondansetron, a 5HT3 receptor
antagonist
Joseph Zoldan, Gali Friedberg, Myron Livneh, et al.
Neurology 1995;45;1305-1308
DOI 10.1212/WNL.45.7.1305

This information is current as of July 1, 1995

Updated Information &

Services including high resolution figures, can be found at:
http://www.neurology.org/content/45/7/1305.full.html

Citations This article has been cited by 12 HighWire-hosted articles:

http://www.neurology.org/content/45/7/1305.full.html##otherarticles
Permissions & Licensing Information about reproducing this article in parts (figures,tables) or in its
entirety can be found online at:
http://www.neurology.org/misc/about.xhtml#permissions
Reprints Information about ordering reprints can be found online:
http://www.neurology.org/misc/addir.xhtml#reprintsus

Neurology ® is the official journal of the American Academy of Neurology. Published continuously since 1951, it
is now a weekly with 48 issues per year. Copyright Copyright 1995 by the American Academy of Neurology. All
rights reserved. Print ISSN: 0028-3878. Online ISSN: 1526-632X.