respiratory dyskinesia plasma risperidone levels and...

Respiratory Dyskinesiaas Discontinuation

Effect of Risperidone

To the Editors:

Atypical antipsychotics, among themrisperidone, cause less extrapyramidalside effects, including tardive dyskinesia(TD), than conventional antipsychotics.1

One of the many presentations of TDreported is respiratory dyskinesia (RD)which is characterized by involuntarymovements of respiratory muscles, ir-regular respiration, grunting, dyspnea,and reactive hyperventilation.2 To thebest of our knowledge, there are no re-ports of risperidone causing RD. Wepresent a case of RD after the discon-tinuation of risperidone in a dementedelderly patient.

CASE REPORTAn 84-year-old woman with demen-

tia was referred to the geriatric psychiatricoutpatient clinic. The Mini-Mental State Ex-amination score was 16/30, and the Cam-bridge Cognitive Examination score was55/105. She had visual hallucinations, wasprone to falls, and presented with fluctua-tions in cognitive functioning. We diagnoseddementia with Lewy bodies (DLB).3 Hercurrent medication was 10 mg donepezil, and2 mg risperidone in the previous 3 months.During this period, the only nonpsychotro-pic substance she received was ranitidine150 mg. There was no record or indicationof prior use of psychotropic medication.

During examination, rigidity in the up-per limbs with cogwheel phenomenon wasnoted.

Risperidone was gradually withdrawnover a period of 1 week, and quetiapinewas introduced at a dose of 25 and, later,50 mg/d. Three days after risperidone with-drawal, the caregivers reported that thepatient’s condition deteriorated. Reportedsymptoms included anxiety with bouts ofhyperventilation.

During examination, she displayed in-voluntary movements of the respiratory mus-culature. She grunted and displayed signs ofdistress resulting from irregular respiration.No additional dyskinetic symptoms werenoted. We diagnosed RD and restarted 2 mgrisperidone. This resulted in immediate symp-

tom alleviation. Risperidone dosage wasreduced in 0.5-mg intervals over a periodof 3 months, during which respiratory symp-toms were subjectively reported as beingmuch less severe with no accompanyingdiscomfort. The treatment with quetiapineremained unchanged. Six months after theinitial observation of RD, no symptomsremained.

To our knowledge, this is the firstreport on RD after risperidone treatment.This condition is most often observed in pa-tients displaying symptoms of TD after treat-ment with antipsychotics. In our patient withDLB, it occurred as discontinuation effect.The prevalence of RD is 2.3% in chronicinpatients and 7.4% among those with TD.4

It is often underrecognized or misdiagnosedas anxiety or restlessness.5 Patients withParkinson disease can also develop sponta-neous RD.6 This is worth considering becauseParkinson disease and DLB seem to shareimportant neuropsychiatric and pathobio-logic features, including degeneration of thedopaminergic nigrostriatal system.

Some authors have reported low in-cidence of TD with the administration ofrisperidone in dementia,1 but our case under-lines the high vulnerability for side effectson antipsychotics in patients with DLB.

Uwe Ehrt, MD

Friederike Fritze, MD

Dag Aarsland, MD, PhDResearch Center for Clinical Nueroscience

Stavanger University Hospital

Psychiatric Clinic,

Stavanger, Norway

[email protected]

REFERENCES

1. Jeste DV, Okamoto A, Napolitano J, et al. Lowincidence of persistent tardive dyskinesia in el-derly patients with dementia treated with risperi-done. Am J Psychiatry. 2000;157(7):1150–1155.

2. Kruk J, Sachdev P, Singh S. Neuroleptic-induced respiratory dyskinesia. J Neuropsychi-atry Clin Neurosci. 1995;7(2):223–229.

3. McKeith I, O’Brien J. Dementia with Lewybodies. Aust N Z J Psychiatry. 1999;33(6):800–808.

4. Yassa R, Lal S. Respiratory irregularity andtardive dyskinesia. A prevalence study. ActaPsychiatr Scand. 1986;73(5):506–510.

5. Chiu HF, Lee S, Chan CH. Misdiagnosis ofrespiratory dyskinesia. Acta Psychiatr Scand.1991;83(6):494–495.

6. Shill H, Stacy M. Respiratory function inParkinson’s disease. Clin Neurosci. 1998;5(2):131–135.

Plasma RisperidoneLevels and Clinical

Response in PatientsWith First-episode

Psychosis

To the Editors:

Risperidone (RSP) is an atypicalantipsychotic with antagonistic propertiesat serotonin 5-HT2 and dopamine D2

receptors. Orally administered RSP isreadily absorbed and extensively metab-olized to various metabolites includingthe main metabolite 9-hydroxyrisperidone(9-OH-RSP) by cytochrome P450 en-zymes, especially CYP2D61 and CYP3A.2

9-OH-RSP may reach the brain at sub-stantial concentrations and contributeimportantly to pharmacological actionsand subsequent clinical outcomes afterRSP doses.3 Therefore, previous studiessuggest that it is important to measuresteady-state plasma levels of total activemoiety by analyzing both RSP and 9-OH-RSP for plasma drug monitoring.4

The pharmacokinetics of RSPshows a wide interindividual variabilityunder the influence of various phys-iological, genetic, and environmentalfactors such as age, CYP2D6 genotype,and so on.5–7 In a recent study, Spinaet al3 assessed the relationship betweenplasma levels of RSP and 9-OH-RSPand clinical response in 42 schizophren-ic patients who experienced an acuteexacerbation of the disorder. In thisopen-label study, the authors concludedthat the plasma levels correlated withthe occurrence of extrapyramidal sideeffects (EPSEs), but not with the degreeof clinical improvement.

The aim of this study was toassess the relationship between plasmalevels of RSP and its active metabolite,9-OH-RSP, and the clinical response indrug-naive Asian patients with first-episode psychosis.

The study population comprised19 consecutive referrals to the inpa-tient and outpatient services of the Ear-ly Psychosis Intervention Programme,

Letters to the Editors

Journal of Clinical Psychopharmacology � Volume 25, Number 6, December 2005 609

Copyr ight © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Singapore. Drug-naive patients with first-episode psychosis who were started onRSP monotherapy and who gave a writ-ten informed consent were recruited.Patients were excluded if they had re-ceived concomitant therapy with otherantipsychotic, antidepressant, or moodstabilizer medication, as well as if theyhad a history of current substance use orserious medical illness. Diagnosis wasmade using the Structured Clinical In-terview for Diagnostic and StatisticalManual of Mental Disorders, FourthEdition Axis I disorders.8

Patients were assessed at baselinebefore being started on RSP and then atweeks 2, 6, and 12. Each assessment in-cluded the following measures: Positiveand Negative Syndrome Scale (PANSS)to assess the psychopathology9 andSimpson-Angus Scale for Extrapyra-midal Side Effects10 to rate the severityof EPSEs. The assessments were doneby experienced psychiatrists who weretrained in the use of the rating instru-ments. Interrater reliability coefficientamong the raters was established at 0.80.

RSP was started at a dose of 0.5 to1 mg/d and increased gradually to atarget dose of 1 to 3 mg/d, based on thetherapeutic response and developmentof side effects. The patients were ad-ministered the entire dose at night be-fore sleep.

Blood samples for the determina-tion of RSP and 9-OH-RSP levels wereobtained at the end of week 12. Bloodsamples were taken in the morning, 12hours after the last dose. The determi-nation of the RSP and 9-OH-RSP levelswas performed by high-performanceliquid chromatography method usingultraviolet detection.11

Spearman rank correlation coeffi-cient was used to test the correlationsbetween the dose and plasma drug lev-els, between the dose and clinical re-sponse, and between plasma levels andclinical response. Statistical significancewas set at P < 0.05.

Nineteen Chinese patients (10 menand 9 women) with a mean (SD) age of28.74 (6.7) years and mean (SD) dura-tion of untreated illness of 11.91 (22.04)months completed the study. The mean(SD) daily dose of RSP prescribed was1.47 (0.56) mg. The mean (SD) reduc-tion in PANSS score was from 64.58

(3.33) at baseline to 41.32 (2.20) atweek 12. Only 2 patients developedEPSE as defined by Simpson-AngusScale for Extrapyramidal Side Effectsscore of 0.3 or more.12

The mean (SD) plasma levelswere as follows: RSP, 2.77 (3.66) ng/mL;9-OH-RSP, 9.62 (6.10) ng/mL; andactive moiety (RSP + 9-OH-RSP),12.39 (8.21) ng/mL. There was a sig-nificant correlation between plasma lev-els of RSP and 9-OH-RSP (r = 0.66,P = 0.002). We also observed a signifi-cant correlation between the dose ofRSP per kilogram of body weight atweek 12 and plasma levels of 9-OH-RSP (r = 0.46, P = 0.046), but not withplasma levels of RSP and active moiety.After adjusting for dose, there were nosignificant correlations between age orsex and plasma levels of RSP, 9-OH-RSP, and active moiety.

We did not find a significant cor-relation between plasma levels of RSP,9-OH-RSP, and active moiety with im-provement in PANSS scores (%) (Fig. 1).

Two patients (9.50%) developedEPSE as defined by Simpson-AngusScale for Extrapyramidal Side Effectsscore of 0.3 or more.12 Patients who de-veloped EPSE did not have significantlyhigher plasma levels of RSP, 9-OH-RSP, and active moiety than patientswho did not develop EPSE.

DISCUSSIONAtypical antipsychotic drugs are

gradually becoming the first-line agents

in the treatment of first-episode psy-chosis. By examining the correlationbetween the dose, plasma levels, andclinical response, we sought to betterunderstand the role of therapeutic plas-ma drug monitoring for RSP in patientswith first-episode psychosis.

There was a significant correla-tion between weight-normalized dose ofRSP and 9-OH-RSP plasma levels, butnot RSP plasma levels. This findingsuggests that 9-OH-RSP is an importantand clinical active metabolite of RSPand needs to be measured in the ther-apeutic plasma monitoring.

We did not find a significantcorrelation between dose of RSP andclinical response. We also did not find acorrelation between plasma drug levelsand clinical response. However, a closerlook at the scatterplot figure indicatedthat this was because of 2 subjects whowere nonresponders, and when theywere not included in the correlationalanalysis, there was indeed a significantcorrelation between clinical responseand plasma levels of RSP, 9-OH-RSP,and active moiety. A closer clinical lookat these 2 patients indicated that one ofthem showed a delayed response (afterthe end of study period). In the secondpatient, the psychotic symptoms hadresolved; however, there was an emer-gence of depressive and anxiety symp-toms, and this increased the PANSSscores. Hence, there may be a trend to-ward a correlation between plasma druglevels and clinical response. However,

FIGURE 1. Correlation between plasma levels of active moiety and PANSS.

Letters to the Editors Journal of Clinical Psychopharmacology � Volume 25, Number 6, December 2005

610 n 2005 Lippincott Williams & Wilkins


our sample size was small, and this mayhave led to type II error.

Our study had few limitations: thedose of RSP was not controlled, and wehad a single plasma determination. Thiswas a naturalistic study, and hence,factors such as the treatment setting(inpatient vs. outpatient) and the diag-nosis, as well as severity of psychopa-thology, may have also influenced thetreatment response.

Swapna K. Verma, MD*Chay Hoon Tan, MMed, PhDy

Yiong Huak Chan, PhDySiow Ann Chong, MMed*

*Institute of Mental Health and Woodbridge

Hospital and the yNational University

of Singapore, Singapore

[email protected]

REFERENCES

1. Huang ML, Van Peer A, Woestenborghs R,et al. Pharmacokinetics of the novel antipsy-chotic agent risperidone and the prolactinresponse in healthy subjects. Clin PharmacolTher. 1993;54:257–268.

2. Bork JA, Rogers T, Wedlund PJ, et al. A pilotstudy on risperidone metabolism: the roleof cytochromes P450 2D6 and 3A. J ClinPsychiatry. 1999;60(7):469–476.

3. Spina E, Avenoso A, Facciola G, et al.Relationship between plasma risperidone and9-hydroxyrisperidone concentrations and clin-ical response in patients with schizophrenia.Psychopharmacology. 2001;153:238–243.

4. Aravagiri M, Marder SR, Nuechterlein KH,et al. Intra- and interindividual variations insteady-state plasma concentrations of risper-idone and 9-hydroxyrisperidone in schizo-phrenic patients treated chronically withvarious doses of risperidone. Ther Drug Monit.2003;25:657–664.

5. Snoeck E, Van Peer A, Sack M, et al.Influence of age, renal and liver impairmenton the pharmacokinetics of risperidone inman. Psychopharmacology. 1995;122:11–19.

6. Anderson C, True J, Ereshefsky L, et al.Risperidone clinical efficacy: role of metab-olite 9-hydroxy-risperidone. Psychopharma-col Bull. 1994;30:88.

7. Darby JK, Pasta DJ, Elfand L, et al.Risperidone dose and blood variability: accu-mulation effects and interindividual and intra-individual variability in the nonresponderpatient in the clinical practice setting. J ClinPsychopharmacol. 1997;17:478–484.

8. First MB, Spitzer RL, Gibbon M, et al.Structured Clinical Interview for DSM IVAxis I Disorders, Patient Edition (SCID-P),Version 2. New York: New York State Psy-chiatric Institute, Biometrics Research; 1994.

9. Kay S, Fiszbein R, Opler A. The Positive andNegative Syndrome Scale (PANSS) for schizo-phrenia. Schizophr Bull. 1987;13(2):261–276.

10. Simpson GM, Angus JWS. Drug inducedextrapyramidal disorders: a rating scale forextrapyramidal side effects. Acta PsychiatrScand Suppl. 1970;212:11–19.

11. Remmerie BM, Sips LL, De Vries R, et al.Validated method for the determinationof risperidone and 9-hydroxyrisperidone inhuman plasma by liquid chromatography–tandem mass spectrometry. J Chromatogr BAnalyt Technol Biomed Life Sci. 2003;783:461–472.

12. Halliday J, Farrington S, Macdonald S, et al.Nithsdale schizophrenia surveys 23: move-ment disorders: 20 year review. Br J Psychi-atry. 2002;181:422–427.

‘‘Extended’’Antipsychotic Dosing

Rationale andPilot Data

To the Editors:

In chronic psychotic illnesses suchas schizophrenia, oral antipsychotic med-ications are administered daily duringmaintenance treatment. Efforts to re-duce exposure through prolonged drugholidays (eg, intermittent treatment) oruse only when symptoms emerge (eg,targeted therapy) have been associatedwith increased relapse rates.1–3

Several recent lines of investiga-tion, however, caused us to rethink theissue of daily administration. DopamineD2 receptor blockade appears to be thesine qua non of antipsychotic activity,4

and more recent work involving posi-tron emission tomography indicates thatthe optimal chance of clinical responseoccurs when D2 occupancy exceeds 65%.5

Therapeutic doses of risperidone andolanzapine demonstrate D2 occupancyin the range of 70% even 48 hours af-ter the last administration.6 In addition,drugs such as quetiapine achieve anti-psychotic response with only transientlyhigh D2 daily occupancy.7 Along similarlines, haloperidol decanoate maintainsresponse despite D2 occupancy levelsdropping to less than 65% by the end ofthe injection interval.8 It has also beendemonstrated that the injection intervalfor fluphenazine decanoate can be ex-tended from every 2 weeks to every 6weeks without compromising response.9

Postulating that extended (but regular)dosing may be sufficient to maintain

antipsychotic response, we conducted apilot study as proof of principle.

Subjects were recruited from theSchizophrenia Program at the Centre forAddiction and Mental Health, a tertiarycare setting associated with the Univer-sity of Toronto. The study was approvedby the human subjects review commit-tee, and subjects provided written in-formed consent after receiving detailedinformation about the protocol.

Inclusion criteria included (i) out-patient status, (ii) Diagnostic and Statis-tical Manual of Mental Disorders, FourthEdition diagnosis of schizophrenia orschizoaffective disorder, (iii) stabiliza-tion on a single oral antipsychotic 3months or longer with no change on theClinical Global Impression Scale over2 weeks before study entry, (iv) no depotantipsychotic use in the last year, (v) nocurrent history of substance-related dis-orders, and (vi) antipsychotic compli-ance; that is, both study candidate andtreating clinician independently assessedcompliance with current antipsychotictreatment to be greater than 90%.

Subjects could be on any oral anti-psychotic, typical or atypical, except clo-zapine and quetiapine, based on currentthinking regarding their D2 binding pro-file, that is, rapid dissociation10,11 andanecdotal evidence linking abrupt dis-continuation of such agents with rapidrelapse.12

The trial was open, and at baseline,participants were instructed that theirantipsychotic would be decreased in 2phases, each 3 months in duration. Theplan was to administer the medicationat the current dose every second dayfor the first 3 months, and should thepatients remain stable, the interval wasto increase to every third day. A calen-dar detailing the schedule was provided,and follow-up visits were scheduledevery 2 weeks, with the primary out-come measure being change in clinicalsymptoms using the Brief PsychiatricRating Scale and Clinical Global Impres-sion Scale.13

Thirteen subjects were recruited,with individual data summarized inTable 1. Six individuals completed the6-month trial, and only 1 patient re-lapsed and required hospitalization afterdeciding to discontinue the antipsychot-ic medication altogether. Three subjects

Journal of Clinical Psychopharmacology � Volume 25, Number 6, December 2005 Letters to the Editors

n 2005 Lippincott Williams & Wilkins 611


TA

BLE

1.

Dat

aon

Ind

ivid

ual

Sub

ject

sin

Exte

nd

ed

Dosi

ng

Tri

al(n

=13)

Ba

seli

ne

En

dP

oin

t

Su

bje

ctS

exA

ge

DS

M-I

V

Dia

gn

osi

s

Du

rati

on

of

Illn

ess

(y)

An

tip

sych

oti

c/

Da

ily

Do

se

Oth

erP

sych

otr

op

ic

Dru

gs/

Da

ily

Do

sea

tB

ase

lin

e

Tri

al

(wk

)B

PR

SC

GI

BP

RS

CG

IC

om

men

ts

1M

ale

23

Sch

izo

ph

ren

ia2

.8L

ox

apin

e3

0m

gL

ora

zep

am1

mg�

3,

ben

ztro

pin

e2

mg�

224

18

21

91

Sid

eef

fect

sdim

inis

hed

on

day

so

ffan

tip

sych

oti

c

2M

ale

39

Sch

izo

ph

ren

ia6

Ola

nza

pin

e5

mg

43

33

36

3W

ith

dre

wco

nse

nt

3*

Fem

ale

55

Sch

izo

ph

ren

ia2

9O

lan

zap

ine

5m

g2

42

12

22

2L

ess

‘‘h

eav

ines

s’’

on

day

so

ffan

tip

sych

oti

c

4M

ale

23

Sch

izo

ph

ren

ia4

Ola

nza

pin

e1

5m

g2

42

12

19

1M

isse

dse

dat

ing

qu

alit

ies

of

dai

lym

edic

atio

n

5*

Mal

e4

0S

chiz

op

hre

nia

23

Ola

nza

pin

e2

0m

g6

26

12

01

Wit

hd

rew

con

sen

tb

ecau

seo

fin

crea

sed

app

oin

tmen

ts

6M

ale

45

Sch

izo

ph

ren

ia2

8H

alo

per

ido

l2

0m

g2

43

64

36

5S

leep

ing

less

7M

ale

56

Sch

izo

ph

ren

ia3

0O

lan

zap

ine

5m

g6

20

13

54

Sto

pp

edan

tip

sych

oti

cco

mp

lete

lyan

dre

lap

sed

8F

emal

e5

9S

chiz

oaf

fect

ive

12

Flu

pen

thix

ol

3m

gV

alp

roat

e1

00

0m

g2

41

91

18

1D

escr

ibed

‘‘cl

eare

rth

ink

ing

’’

9M

ale

22

Sch

izo

ph

ren

ia4

.5L

ox

apin

e3

0m

g1

22

93

31

4‘‘

No

tso

slu

gg

ish

,’’

wit

hd

rew

con

sen

td

ue

tosc

hed

uli

ng

vis

its

10y

Fem

ale

27

Sch

izo

ph

ren

ia4

Ola

nza

pin

e7

.5m

g2

42

12

21

2L

ess

fati

gu

ed

ays

off

anti

psy

cho

tic

11

Mal

eS

chiz

op

hre

nia

2R

isp

erid

on

e1

.5m

g2

32

3W

ith

dre

wco

nse

nt

12

*F

emal

e2

4S

chiz

op

hre

nia

4O

lan

zap

ine

15

mg

62

73

27

3D

isco

nti

nu

edb

ecau

seo

fid

enti

fied

sub

stan

ceab

use

13

Fem

ale

28

Sch

izo

ph

ren

ia3

Ola

nza

pin

e7

.5m

g6

26

32

43

Wit

hd

rew

fro

mst

ud

yas

rece

ived

job

off

er

DS

M-I

Vin

dic

ates

Dia

gn

ost

ica

nd

Sta

tist

ica

lM

an

ua

lo

fM

enta

lD

iso

rder

s,F

ou

rth

Ed

itio

n;

BP

RS

,B

rief

Psy

chia

tric

Rat

ing

Sca

le;

CG

I,C

linic

alG

lobal

Impre

ssio

n.

*B

enzo

dia

zepin

epre

scri

bed

for

inso

mnia

inin

itia

lw

eeks

of

tria

l(c

lonaz

epam

0.5

mg

or

lora

zepam

1m

gnig

htl

y,

asnee

ded

).y A

t3

mo

nth

s,su

bje

ctw

asst

able

bu

tre

qu

este

dto

con

tin

ue

alte

rnat

ed

ayd

osi

ng

for

the

seco

nd

3m

on

ths.




were prescribed benzodiazepine as need-ed for sleep disturbance on the drug-free nights.

Data analysis included 11 subjects,excluding the individual who relapsedbecause of noncompliance and a subjectwho withdrew consent and did not attendbeyond the first visit. Paired t tests usingbaseline versus end point scores for theBrief Psychiatric Rating Scale and Clin-ical Global Impression Scale failed toshow significant differences for each: t11

= 0.50, P = 0.63; and t11 = 0.000, P = 1.0,respectively. Scales to evaluate extrapy-ramidal symptoms were completed; how-ever, scores at baseline were so low thatstatistical analyses were not carried out.Because antipsychotics have also beenlinked to the induction of depressionand secondary negative symptoms, BriefPsychiatric Rating subscales addressingwithdrawal/retardation, anxiety/depres-sion, and activation were analyzed, butagain, paired t tests failed to show sig-nificant differences: t10 = 1.20, P = 0.26;t10 = 0.11, P = 0.92; and t10 = 1.15, P =0.28, respectively.

Advances in understanding regard-ing the relationship between D2 block-ade, clinical response, and side effects5,6

led us to hypothesize that efficacy ofantipsychotic drugs might be maintain-ed with intermittent but regular, thatis, extended, antipsychotic dosing, andthese pilot data suggest that this may bepossible. Only 1 of the 13 individuals en-rolled in the study, who stopped his anti-psychotic completely, relapsed.

Why would we entertain extendeddosing that is, in practice, more com-plex than regular daily medication ad-ministration? The goal in antipsychotictreatment is one of minimizing expo-sure without compromising efficacy.All currently available antipsychoticshave dopamine-blocking properties andcan adversely influence cognition, af-fect, and motivation,14 even at relativelylow doses. Reducing antipsychotic doseis therefore beneficial in this regard, butthe D2 occupancy data would suggestthat there is a threshold, in the levelof 65%, below which response may becompromised.5 In contrast, it appearsthat sustained D2 occupancy above thisthreshold is not required, offering anoth-er means of decreasing overall antipsy-chotic exposure.7–9

Although we were unable to es-tablish improvement in these other clini-cal domains with the current measures,various participants did describe subjec-tive improvement and a preference for theextended dosing approach. In addition,issues such as weight gain, metabolicdisturbances, and lipid dysregulationhave come to the fore with the atypicalagents, and while it is not known yetwhether decreasing exposure can reducesuch effects, it is a reasonable possibilityto explore. Finally, studies such as thishave important theoretical implicationsregarding the underlying mechanismsof action necessary to maintain response.

We highlight the preliminary na-ture of these data and their limitations.This was a small, open, pilot study, andthe patient sample represented a selectsubgroup: stable, has minimal polyphar-macy, reasonable to excellent symptomcontrol, compliant, and has no concom-itant substance abuse. Whereas only oneindividual relapsed, that is, the personwho stopped his medication completely,5 others discontinued for other reasons,resulting in 6-month data on only 6subjects. It can also be argued that alonger follow-up interval is required tofully evaluate the risk of relapse.

We are, however, encouraged bythe findings and have now undertaken alarger double-blind trial to better eval-uate our hypothesis.

Gary Remington, MD, PhD, FRCPC*yPhilip Seeman, MD, PhD, FRSCyz

Chekkera Shammi, MBBS, DPM,

MRCPsych, FRCPC*ySteve Mann, BA*

Shitij Kapur, MD, PhD, FRCPC*y*Schizophrenia Program, Centre for

Addiction and Mental Health; and the

yDepartments of Psychiatry and

zPharmacology, University of Toronto,

Toronto, Ontario, Canada

[email protected]

ACKNOWLEDGMENTS

The authors thank the patients fortheir participation.

REFERENCES

1. Carpenter WT Jr, Hanlon TE, Heinrichs DW,et al. Continuous versus targeted medicationin schizophrenic outpatients: outcome results.Am J Psychiatry. 1990;147:1138–1148.

2. Buchanan RW, Carpenter WT Jr. Targetedmaintenance treatment in schizophrenia: is-sues and recommendations. CNS Drugs. 1996;4:240–245.

3. Gitlin M, Nuechterlein K, Subotnik KL, et al.Clinical outcome following neuroleptic dis-continuation in patients with remitted recent-onset schizophrenia. Am J Psychiatry. 2001;158:1835–1842.

4. Kapur S, Remington G. Dopamine D2 recep-tors and their role in atypical antipsychoticaction: still necessary and may even besufficient. Biol Psychiatry. 2001;50:873–883.

5. Kapur S, Zipursky R, Jones C, et al. Rela-tionship between dopamine D2 occupancy,clinical response, and side effects: a double-blind PET study of first-episode schizophrenia.Am J Psychiatry. 2000;157:514–520.

6. Tauscher J, Jones C, Remington G, et al.Significant dissociation of brain and plasmakinetics with antipsychotics. Mol Psychiatry.2002;7:317–321.

7. Kapur S, Zipursky R, Jones C, et al. A positronemission tomography study of quetiapine inschizophrenia. Arch Gen Psychiatry. 2000;57:553–559.

8. Nyberg S, Farde L, Halldin C, et al. D2 do-pamine receptor occupancy during low-dosetreatment with haloperidol decanoate. Am JPsychiatry. 1995;152:173–178.

9. Carpenter WT Jr, Buchanan RW, KirkpatrickB, et al. Comparative effectiveness of flu-phenazine decanoate injections every 2 weeksversus every 6 weeks. Am J Psychiatry. 1999;156:412–418.

10. Seeman P, Tallerico T. Antipsychotic drugswhich elicit little or no parkinsonism bindmore loosely than dopamine to brain D2 recep-tors, yet occupy high levels of these receptors.Mol Psychiatry. 1998;3:123–134.

11. Seeman P. Atypical antipsychotics: mechanismof action. Can J Psychiatry. 2002;47:27–38.

12. Shore D. Clinical implications of clozapinediscontinuation: report of an NIMH workshop.Schizophr Bull. 1995;21:333–338.

13. Guy W, ed. Assessment Manual for Psycho-pharmacology. Rockville, Md: National Insti-tute of Mental Health; 1976:157–165, 217–222.

14. Kapur S, Lecrubier Y, eds. Dopamine in thePathophysiology and Treatment of Schizophre-nia. London, England: Martin Dunitz; 2003:49–67.

The Prevalence ofHyperprolactinemia

After Long-termHaloperidol Use in

Patients With ChronicSchizophrenia

To the Editors:

Chronic prolactin elevations shouldbe a major focus in clinical practice




because of the potential risk of seriouschronic medical problems such as men-strual abnormalities, gynecomastia, ga-lactorrhea, osteoporosis, and cardiovascu-lar disease.1 However, there have been fewstudies on the prevalence of chronichyperprolactinemia in the patients withschizophrenia. To examine the prevalenceof hyperprolactinemia in the chronicschizophrenia with long-term haloperidoluse, serum prolactin levels were examined.All patients with oral haloperidol mono-therapy in 2 large psychiatric facilities inKorea were evaluated. Inclusion criteriaincluded subjects with a Diagnostic andStatistical Manual of Mental Disordersdiagnosis of schizophrenia and betweenthe ages of 18 to 45 years. In addition, onlythose on haloperidol monotherapy (>1year) and who received the same dosagefor at least 3 months (mean dose, 15.7 ±10.9 mg/d; median dose, 15.0 ± 12.5 mg/d)were included. Benztropine and benzodi-azepine were the only concomitant med-ications permitted. Sixty subjects (28women and 32 men) were enrolled inthis study. The mean illness durationwas 15.5 ± 7.8 years. Hyperprolactinemiawas defined as a serum prolactin level ofmore than 20 ng/mL for men and morethan 24 ng/mL for women. Two-tailedStudent t tests were used to evaluate theclinical variables. Spearman rank correla-tion was use to examine the relationshipof prolactin level to haloperidol doseand weight. All tests were 2-tailed, andsignificance was defined as an a less than0.05. All patients were enrolled into thisstudy after giving written consent. Thisstudy was approved by the institutionalreview board of Busan Paik Hospital,Inje University.

Overall, 66% of the subjects hadhyperprolactinemia. Prolactin levels werenot significantly related to body massindex. There were no significant differ-ences in age (38.9 ± 5.0 vs. 40.6 ± 4.6years), duration of haloperidol treatment(86.5 ± 62.2 vs. 95.5 ± 55.5 months),and dose of haloperidol (15.0 ± 12.1vs. 16.3 ± 10.0 mg/d) between womenand men (Table 1). However, the meanprolactin level for women was signifi-cantly higher than that for men (73.8 ±46.2 vs. 24.15 ± 15.56 ng/mL,t58 = �5.52, P < 0.0001). The prevalenceof hyperprolactinemia for women (93%)was significantly higher than that formen (47%; m2

1 = 14.59, P < 0.0001;

Table 1). A significant correlation wasobserved between haloperidol dose andserum prolactin level in women (r =0.4262, P = 0.021), but not in men (r =0.15, P = 0.414). Although a statisti-cally significant correlation was foundbetween haloperidol dose and serumprolactin level in women, there was awide range of prolactin levels observedwith each fixed dose of haloperidol. Forexample, the range of prolactin levelsfor the women (n = 8) taking 5 mg/dof haloperidol varied from 27.1 to199.9 ng/mL. So one should be carefulwhen interpreting the correlation be-tween haloperidol dose and serum pro-lactin level.

Our data are consistent with pre-vious studies regarding the prevalence ofhyperprolactinemia.2,3 It is well knownthat sex-specific side effects of anti-psychotics treatment occur with suchside effects such as hyperprolactinemia,osteoporosis, and weight gain.4 Accord-ingly, our results also show that hy-perprolactinemia occurs more frequentlyamong women than among men. How-ever, the prevalence of hyperprolactine-mia for women in our group was higherthan other findings. Kinon et al3 reportedthe prevalence of hyperprolactinemia inchronic schizophrenia patients withlong-term antipsychotic drug treatment.

Among women of reproductive age andmen, the percent of hyperprolactinemiawas 65.6% and 42.4%, respectively. Bydrug group, among women, this corre-sponded to 88% with risperidone treat-ment and 47.6% with conventional drug.This prevalence is noted to be lowerthan our results; however, dose in ourstudy is higher than the dose reportedby Kinon et al3 which may explainsome of the variance. Mean antipsychot-ic dosing in our subjects was 14.9 ±12.1 mg/d of haloperidol versus 401.9 ±398.6 mg/d chlorpromazine equivalentsreported in the other study. Althoughthe dose of both groups was sufficientto block most D2 receptors in the tuber-oinfundibular tract, it is difficult tocompare their results directly with ours,because they included a host of con-ventional antipsychotics. Interethnic dif-ference from a pharmacogenetic stand-point should also be considered toexplain the higher rate of prolactin ele-vations found in our study. It is knownthat the activity of cytochrome P450enzyme such as CYP2D6 is differentbetween Asians and whites, and this cancause an interethnic difference in thera-peutic outcome, adverse effects, andtoxicity in subjects of different ethnicorigin undergoing antipsychotic drugtreatment.4,5

TABLE 1. A Comparison Between Women and Men

Patients With

Schizophrenia (n = 60)

Women

(n = 28)

Men

(n = 32) Statistics

Age (y) 38.9 ± 5.0 40.6 ± 4.6 t = 1.39, df = 58, P = 0.173

Body mass index (kg/m2) 23.9 ± 3.4 23.9 ± 2.6 t = �0.02, df = 58, P = 0.982

Duration of takinghaloperidol (mo)

86.5 ± 62.2 95.5 ± 55.5 t = �0.61, df = 58, P = 0.544

Duration of taking same doseof haloperidol (mo)

31.4 ± 39.2 31.2 ± 30.0 t = �0.02, df = 58, P = 0.750

Mean dose of haloperidol(mg/d)

15.0 ± 12.1 16.3 ± 10.0 t = 0.46, df = 58, P = 0.649

Median dose of haloperidol(mg/d)

10.0 ± 9.8 15.0 ± 10.0 NA

Mean serum prolactin level(ng/mL)

73.8 ± 46.3 24.2 ± 15.6 t = �5.42, df = 58, P < 0.0001

% With hyperprolactinemia (n) 92.9 (26) 46.9 (15) X 2 = 14.59, P < 0.0001

% With menstrualdisturbances (n)

96.4 (27) NA NA

Values given are mean ± SD, median ± interquartile range. NA indicates not assessed.




In our study, the prevalence ofhyperprolactinemia was notably higherthan that in short-term haloperidol treat-ment studies.2,6,7 It is interesting whetherlong-term treatment with haloperidolproduces cumulative effects on the prev-alence of hyperprolactinemia. Therehave been only a few studies on the timecourse of hyperprolactinemia. Kinonet al3 reported that patients with hyper-prolactinemia sustained the effect overtime. David et al7 showed early peaks inprolactin with haloperidol which thendecreased, yet still remained above thereference range after completing theearly phase. Our findings support thatprolactin elevations with haloperidol areat least sustained and can be cumulative;however, this should be confirmed withfurther prospective studies.

A few limitations should be con-sidered in interpreting the findings ofthis report. First, it is a cross-sectionalstudy with no comparator. Next, thesample size was small. Lastly, the doseof haloperidol was higher than dosesused in other clinical studies, because ofchronic nature of patients of sample.

Our study suggests that hyper-prolactinemia with long-term haloperi-dol treatment is more frequent andserious than in short-term treatment.Schizophrenia is a chronic mental disor-der, and most patients with schizophre-nia need antipsychotic drug throughouttheir lifetime. While more long-termstudies are needed to characterize therisk of hyperprolactinemia with adverseconsequences, a few recent studies havedemonstrated bone loss8 and breastcancer.9 Furthermore, poor subjectivewell-being regarding hormonal side ef-fects and sexual dysfunction related tohyperprolactinemia can complicate theclinical outcomes of treatment. Ourfindings also suggest that sex specificitymust be considered when one interpretsthe prolactin response to antipsychoticdrug in the clinical practice as well as inthe research setting. Because of the im-portance of long-term health outcomesfor patients, more attention should bepaid to the minimization of this adverseeffect in both sexes.

Do-Un Jung, MD*Young-Soo Seo, MD, PhDyJae-Hoon Park, MD, PhDy

Chi-Young Jeong, MD*Robert R. Conley, MDz

Deanna L. Kelly, PharmD, BCPPzJoo-Cheol Shim, MD, PhDx

*Department of Psychiatry, Dongseo

Hospital, Masan; yDepartment of

Psychiatry, Dongrae Hospital, Busan, Korea;

zMaryland Psychiatric Research Center,

University of Maryland, Baltimore, MD and

xDepartment of Psychiatry and Clinical

Pharmacology, Busan Paik Hospital,

Inje University, Busan, Korea

[email protected]

ACKNOWLEDGMENTS

This study was supported by agrant of the Korea Health 21 R&DProject, Ministry of Health and Wel-fare, Republic of Korea (0412-CT02-0704-0006).

REFERENCES

1. Haddad PM, Wieck A. Antipsychotic-inducedhyperprolactinaemia: mechanisms, clinical fea-tures and management. Drugs. 2004;64:2291–2314.

2. Crawford AM, Beasley CM Jr, Tollefson GD.The acute and long-term effect of olanzapinecompared with placebo and haloperidol onserum prolactin concentrations. Schizophr Res.1997;26:41–54.

3. Kinon BJ, Gilmore JA, Liu H, et al. Hyper-prolactinemia in response to antipsychoticdrugs: characterization across comparative clin-ical trials. Psychoneuroendocrinology. 2003;28(suppl 2):69–82.

4. Kim K, Johnson JA, Derendorf H. Differ-ences in drug pharmacokinetics between EastAsians and Caucasians and the role of geneticpolymorphisms. J Clin Pharmacol. 2004;44:1083–1105.

5. Nikoloff D, Shim JC, Fairchild M, et al.Association between CYP2D6 genotype andtardive dyskinesia in Korean schizophrenics.Pharmacogenomics J. 2002;2:400–407.

6. Zhang XY, Zhou DF, Cao LY, et al. Prolactinlevels in male schizophrenic patients treatedwith risperidone and haloperidol: a double-blind and randomized study. Psychopharma-cology (Berlin). 2005;178:35–40.

7. David SR, Taylor CC, Kinon BJ, et al. Theeffects of olanzapine, risperidone, and haloper-idol on plasma prolactin levels in patients withschizophrenia. Clin Ther. 2000;22:1085–1096.

8. Abraham G, Paing WW, Kaminski J, et al.Effects of elevated serum prolactin on bonemineral density and bone metabolism in femalepatients with schizophrenia: a prospectivestudy. Am J Psychiatry. 2003;160:1618–1620.

9. Halbreich U, Shen J, Panaro V. Are chronicpsychiatric patients at increased risk for de-veloping breast cancer? Am J Psychiatry. 1996;153:559–560.

Delusional DisorderWith Delusions of

Parasitosis and JealousyAfter Stroke

Treatment WithQuetiapine and Sertraline

To the Editors:

Delusional parasitosis is a rarepsychiatric disorder characterized by thefalse and persistent belief of being in-fested by parasites. Pimozide usuallyhas been considered the treatment ofchoice,1 although the increasing and suc-cessful use of atypical antipsychoticswith a more favorable side effects pro-file has raised doubts about this practice.We report the case of a 74-year-oldwoman with very mild vascular demen-tia who developed a mixed delusionaldisorder that included delusions of par-asitosis secondary to stroke. Partialclinical remission was achieved withrelatively low quetiapine doses andsertraline, with no significant side ef-fects. Few reports on the use of quetia-pine or olanzapine in delusional parasit-osis have been published.2,3 Aw et al4 didnot find quetiapine to be useful in 2cases. However, Kim et al5 and Wenninget al6 reported isolated successful cases.Thus, to our knowledge, the present re-port offers the first evidence of thesuccessful combined use of quetiapineand sertraline in patients with delusionalparasitosis.

CASE REPORTMrs A., a 74-year-old woman, had

right hemispheric stroke at age 71 years(November 2001), resulting in left hemi-paresis and confinement to a wheelchair.She was moved to an old people’s homefor appropriate medical care. One year later,she referred itching particularly aroundorifices. She thought that this was due tobeing infested by minute but macroscopicblack insects smaller than lice which sheclaimed to be able to catch and try to killoff by using ammonia. In July 2003, 2months after her husband began to live inthe same old people’s home, she beganto accuse him of cheating on her despitethe lack of objective data. Secondarily, she




became depressed, with difficulties initiatingand maintaining sleep, and had dysphoria,depressed mood most of the day, frequentthoughts of death (although without suicidalideation), decreased appetite, and markedlydiminished pleasure in most of her activities.She firmly stated that all depressive symp-toms were due to her husband cheating onher, and she based her accusation on theobservation that he spoke with other womenin the residence. Previously, no affectivesymptoms had been reported. Finally, inSeptember 2003, she began to explain thatshe kept a gold medallion which was veryprecious to her in one of her shoes, inan attempt to prevent it from being stolen.This was when a first psychiatric assessmentwas carried out. Her medical history com-prised hypertension, type II diabetes mellitus,and 2 previous right hemispheric strokes.No previous psychiatric antecedents werefound. Her mental status was consistent withmild dementia, and brain magnetic reso-nance imaging disclosed diffuse cortical re-traction, a hypodense zone in the right lobe,multiple hypodensities suggestive of cere-brovascular disease, and a lesion in the rightbrainstem and mesencephalic region. Labo-ratory testing revealed the following: bloodglucose, 398 mg/dL; cholesterol, 369 mg/dL;folate, vitamin B12, iron, and thyroid hor-mones, all within reference range; and neg-ative syphilis testing. Mild vascular demen-tia and a Diagnostic and Statistical Manualof Mental Disorders, Fourth Edition (TextRevision) diagnosis of psychotic disordersecondary to general medical conditionwere established. She was put on sertraline100 mg/d, ziprasidone 40 mg/d, and broma-zepam 3 mg/d. Two weeks later, bothziprasidone and bromazepam were dis-continued due to confusional state and ir-ritability. As she continued showing thepreceding psychiatric symptoms, the sertra-line dose was increased to 150 mg/d, andquetiapine was introduced at an initial doseof 25 mg. This was progressively increasedin 25-mg steps every 3 days to avoid potentialhypotension, to a maximum of 300 mg/d.Two months later, no evidence of depres-sive syndrome was observed. She was alsoable to fully criticize her jealousy delusionand partially her delusional parasitosis andspent much less time scratching herself(having been convinced that she had elimi-nated the insects).

DISCUSSIONAlthough few controlled studies

have been carried out, pimozide contin-ues to be considered the treatment ofchoice in delusional parasitosis, basedmainly on case-report series.7 Full re-

mission with pimozide is between28%1,2 and 50%8 to 82%.9 However,treatment with pimozide is limited by itsside effects profile, and atypical anti-psychotics have also proven effective—especially risperidone.3,10,11 Sertindolehas been mentioned as useful only in asingle case file12 and has not been provento be particularly effective. In addition,the drug has been withdrawn from themarket.

It has been proposed that atypicalantipsychotics may be considered earlyin the management of patients diag-nosed with delusional parasitosis, be-cause these patients are very reluctant toaccept psychiatric treatments, and atyp-ical antipsychotics offer a better sideeffects profile.4,8 However, caution hasbeen advised with the use of risperidone,due to the risk of extrapyramidal reac-tions.1,3 Moreover, considering that, inthe present case report, delusional para-sitosis was secondary to vascular de-mentia and that, recently, the Committeeon Safety of Medicines recommendedavoiding risperidone and olanzapine inpatients with dementia due to concernsover the excess risk of stroke,13 and de-spite the fact that another recent pub-lication found both drugs to pose nostatistically significant increase in therisk of cerebrovascular accidents versustypical antipsychotics,14 it seems man-datory to question whether the rest ofatypical antipsychotics could be an alter-native in these patients. In a preliminarywork, Schneider et al15 reported no evi-dence of an increased risk of cerebro-vascular adverse events associated withthe use of quetiapine in elderly patientswith dementia.

It should also be taken into accountthat, in our patient, vascular dementiawas probably secondary to previoushypertension, type 2 diabetes, hypercho-lesterolemia, and 2 previous right hemi-spheric strokes. It seems that type-2diabetes is an independent risk factorfor cognitive decline and dementia.16

Indeed, in a 6-year follow-up of 274elderly participants (36 with diabetesand 238 without diabetes), Hassing et al17

found type 2 diabetes to be associatedwith accelerated cognitive decline in oldage that may result in dementia.

Quetiapine is a second-generationantipsychotic of demonstrated efficacyin application to schizophrenia18 and

bipolar disorder.19 Although the labeleddose range is 400 to 800 mg/d, higherdoses have been used with good clinicaloutcome and a favorable side effectsprofile.20 Blockade of the 5-HT recep-tors—specifically of 5-HT2a—has beenhypothesized to be important in delu-sional parasitosis.1,3 However, quetiapineis a relatively low–5-HT2a affinity anti-psychotic and was found to be useful inthis case report—although the potentialrole of sertraline should also be takeninto account, because the patient wastreated with both sertraline and quetia-pine at the time of improvement.21

In addition, clomipramine monotherapyhas been found to be effective in thetreatment of somatic-type delusional dis-orders, thus suggesting some associa-tion between somatic-type delusionaldisorder and serotoninergic dysfunction.22

Clomipramine may have been a poten-tially useful treatment in this case re-port, because the patient also presentedsecondary depression due to delusionsof jealousy. The absence of prior de-pressive symptoms after stroke and thechronological coincidence of the depres-sive symptoms with her husband begin-ning to live with her in the sameresidence led us to diagnose secondarydepression due to delusions of jealousyand not poststroke depression, becausethe disturbance was better accountedfor by the former diagnosis. Further-more, both the delusions of jealousyand the depressive episode subsided atthe same time after the treatment wasstarted. On the other hand, only a weakrelationship has been described be-tween poststroke depression and righthemispheric lesions.23

Quetiapine offers several advan-tages in the treatment of delusional statesin the elderly, due to its favorable sideeffects profile, with a lack of anticholin-ergic activity,24 extrapyramidal symp-toms, or cardiovascular adverse out-comes.25 Moreover, the commonestside effects of quetiapine—drowsiness,dizziness, and postural hypotension—rarely result in withdrawal from therapyin elderly patients with psychotic disor-ders.13 However, the potential use ofquetiapine in nonschizophrenic psychot-ic syndromes and nonpsychotic disor-ders remains unexplored,12,13,26 andfurther research is needed to considerquetiapine as a first-line treatment option




for delusional parasitosis and otherpsychotic syndromes in the elderly.

Hilario Blasco-Fontecilla, MD*Maria Dolores Bragado Jimenez, MDy

Luis Marıa Garcıa Santos, MDyJose Manuel Barjau Romero, MDy

*Ciudad Lineal Mental Health Center,Madrid and yPsychiatry Department,

Provincial Hospital, Toledo, [email protected]

REFERENCES

1. Zomer AF, de Wit RFE, Van BronswijkJEHM, et al. Delusions of parasitosis. A psy-chiatric disorder to be treated by dermatolo-gists? An analysis of 33 patients. Br J Derm.1998;138:1030–1032.

2. Slaughter JR, Zanol K, Rezvani H, et al.Psychogenic parasitosis. A case series and lit-erature review. Psychosomatics. November–December 1998;39(6):491–500.

3. Le L, Gonski PN. Delusional parasitosismimicking cutaneous infestation in elderlypatients. Med J Aust. August 18, 2003;179(4):209–210.

4. Aw DC, Thong JY, Chan HL. Delusionalparasitosis: case series of 8 patients and re-view of the literature. Ann Acad Med Singa-pore. 2004;33(1):89–94.

5. Kim C, Kim J, Lee M, et al. Delusionalparasitosis as ‘folie a deux’. J Korean MedSci. June 2003;18(3):462–465.

6. Wenning MT, Davy LE, Catalano G, et al.Atypical antipsychotics in the treatment ofdelusional parasitosis. Ann Clin Psychiatry.September–December 2003;15(3):233–239.

7. Elmer MKB, George MRM, Peterson MK.Therapeutic update: use of risperidone forthe treatment of monosymptomatic hypochon-driacal psychosis. J Am Acad Dermatol. 2000;43(4):683–686.

8. Trabert W. 100 Years of delusional parasit-osis: meta-analysis of 1,223 case reports.Psychopathology. 1995;28:236–246.

9. Munro A, Chmara J. Monosymptomatic hy-pochondriacal psychosis: a diagnostic check-list based on 50 cases of the disorder. Can JPsychiatry. 1982;27:374–376.

10. De Leon OA, Furmaga KM, Canterbury AL,et al. Risperidone in the treatment of delu-sions of infestation. Int J Psychiatry Med.1997;27(4):403–409.

11. Freyne A, Keeny E, Cooney C. Delusions ofinfestation—a case report of response torisperidone. Ir Med J. 1999;92(7):435.

12. Yorston G. Treatment of delusional parasito-sis with sertindole. Int J Geriatr Psychiatry.1997;12(11):1127–1128.

13. Mowat D, Fowlie D, MacEwan T. CSMwarning on atypical psychotics and strokemay be detrimental for dementia. BMJ.2004;328(7450):1262.

14. Herrmann N, Mamdani M, Lanctot KL.Atypical antipsychotics and risk of cerebro-vascular accidents. Am J Psychiatry. 2004;161(6):1113–1115.

15. Schneider L, Zhong K, Tariot P, et al.

Cerebrovascular adverse events and quetia-pine: a pooled analysis in elderly patients withdementia. Presented at the International Con-ference on Alzheimer’s Disease and RelatedDisorders Meeting; Philadelphia, Penn: July17–22, 2004; [poster no. P2-443].

16. Messier C, Awad N, Gagnon M. The relation-ships between atherosclerosis, heart disease,type 2 diabetes and dementia. Neurol Res.July 2004;26(5):567–572.

17. Hassing LB, Grant MD, Hofer SM, et al. Type2 diabetes mellitus contributes to cognitivedecline in old age: a longitudinal population-based study. J Int Neuropsychol Soc. July2004;10(4):599–607.

18. Buckley PF. Efficacy of quetiapine for thetreatment of schizophrenia: a combined analy-sis of three placebo-controlled trials. Curr MedRes Opin. September 2004;20(9):1357–1363.

19. Cole P, Rabasseda X. Quetiapine in bipolardisorder: increasing evidence of efficacy andtolerability. Drugs Today (Barcelona). October2004;40(10):837–852.

20. Bobes J, Garcia-Portilla MP, Saiz PA, et al.High degree of tolerability for monotherapywith high doses of quetiapine: a case report.J Clin Psychiatry. 2002;63(11):1048–1049.

21. Tyson PJ, Roberts KH, Mortimer AM. Arethe cognitive effects of atypical antipsychoticsinfluenced by their affinity to 5HT-2A recep-tors? Int J Neurosci. 2004;114(6):593–611.

22. Wada T, Kawakatsu S, Nadaoka T, et al.Clomipramine treatment of delusional disor-der, somatic type. Int Clin Psychopharmacol.May 1999;14(3):181–183.

23. Yu L, Liu CK, Chen JW, et al. Relationshipbetween post-stroke depression and lesionlocation: a meta-analysis. Kaohsiung J MedSci. August 2004;20(8):372–380.

24. Tariot PN, Ismail MS. Use of quetiapine inelderly patients. J Clin Psychiatry. 2002;63(suppl 13):21–26.

25. Tariot PN, Salzman C, Yeung PP, et al. Long-term use of quetiapine in elderly patients withpsychotic disorders. Clin Ther. 2000;22(9):1068–1084.

26. Adityanjee, Schulz SC. Clinical use of quetia-pine in diseases states other than schizophre-nia. J Clin Psychiatry. 2002;63(13):32–38.

Periodic Restless LegsSyndrome AssociatedWith Quetiapine Use

A Case Report

To the Editor:

Restless legs syndrome (RLS) is acommon disorder with prevalence rang-ing from 2% to 15% of general adultpopulation1 but is often undiagnosed ormisdiagnosed.2 Some second-genera-tion antipsychotics such as risperidoneand olanzapine have been reported tocause secondary RLS,3,4 whereas others

have been reported to help it.4,5 No casesof RLS attributed to quetiapine werefound on MEDLINE and EMBASEsearches done on February 28, 2005.We report a case of RLS possibly in-duced by quetiapine.

A 68-year-old white female withbipolar I disorder, manic with psychot-ic symptoms, developed parkinsoniansymptoms of cogwheel rigidity, brady-kinesia, and tremor on a combination ofrisperidone and lithium (serum level,1.2 mEq/L). There was no akathisia atthat time. The parkinsonian symptomsresolved completely on tapering off therisperidone and reducing the dose oflithium. Two months after discontinu-ing risperidone, she was started onquetiapine 100 mg/d for insomnia, de-pressed mood, and mild agitation. Thedosage was titrated up by 50 mg/d.Within 24 hours of receiving 200 mg/d,she complained of uncomfortable andannoying sensations in both her legsthat were worse at rest and at night andrelieved on moving the legs. These sen-sations worsened the initial and middleinsomnia, and the patient would walkabout at night to relieve her symptoms.Neurological examination did not showabnormalities. There was no prior his-tory of RLS, and other causes includingincreased caffeine intake, anemia, renalfailure, and other medication use wereexcluded. Serum ferritin level was with-in reference range at 50 ng/mL. Poly-somnography could not be done becauseof short duration of symptoms. Therewas history of idiopathic RLS in asecond-degree relative. Quetiapine wasreduced to 150 mg/d, and the sensations,restlessness, and sleep disturbance re-solved completely within 48 hours.

Six weeks later, an attempt wasmade again to increase the dose ofquetiapine to 200 mg/d to control mildagitation as there was lack of efficacywith 150 mg/d. Within a day, she hadrecurrence of unpleasant sensations inher legs that were worse in the eveningand at night and relieved with movementof legs, and her sleep was disturbed.After 3 days, she reduced the dose to150 mg/d with complete resolution ofthe symptoms, and she remains symp-tom-free at 6 months.

The case described satisfies theRLS diagnostic criteria established by




International Restless Leg SyndromeStudy Group.6 Restless leg syndromehas to be differentiated from akathisiaas the 2 conditions share some clinicalfeatures, and quetiapine has been re-ported to cause akathisia.7 The symp-toms of dysesthesias in the legs, desireto move because of the dysesthesiasrather than internal restlessness, in-creased severity at rest, diurnal varia-tion with worsening at night, significantsleep disturbance from dysesthesias,and a family history of RLS suggest adiagnosis of RLS.

Restless leg syndrome is fre-quently associated with periodic limbmovement disorder that can only bediagnosed with polysomnography.8 Pol-ysomnography was not done in thisinstance, and hence, the association withperiodic limb movement disorder cannotbe ascertained. Close temporal relationbetween use of quetiapine 200 mg/d,complete resolution with a decrease indose, and reemergence on rechallengewith higher dose indicate that quetiapinewas the likely etiologic agent and thatit was a dose-dependent side effect.

Dopaminergic dysfunction is con-sidered to be a pathophysiologic mech-anism in the development of RLS, anddrugs that block dopamine receptors arelikely to induce RLS.9 Of the second-generation antipsychotics, quetiapinehas lower and more transient D2 dopa-mine receptor occupancy of 58% to64%10 and may be least likely to induceRLS. It has also been shown to improveRLS induced by other second-genera-tion antipsychotics such as risperidone.4

However, physicians should be awarethat quetiapine could cause RLS whenthere are associated risk factors such asfamily history, older age, and femalesex, and reduction in dosage or discon-tinuation of the medication may beadequate to resolve the RLS.

Narsimha Reddy Pinninti, MD*Rajnish Mago, MDy

James Townsend, DO*Karl Doghramji, MDy

*Department of Psychiatry, School of

Osteopathic Medicine–University of

Medicine and Dentistry, Cherry Hill, NJ

and yThomas Jefferson University,

Philadelphia, PA

[email protected]

REFERENCES

1. Zucconi M, Ferini-Strambi L. Epidemiologyand clinical findings of restless legs syn-drome. Sleep Med. 2004;5:293–299.

2. Allen RP, Picchietti D, Hening WA, et al.Restless legs syndrome: diagnostic criteria,special considerations, and epidemiology. Areport from the restless legs syndrome diag-nosis and epidemiology workshop at the Na-tional Institutes of Health. Sleep Med. 2003;4:101–119.

3. Kraus T, Schuld A, Pollmacher T. Periodicleg movements in sleep and restless legs syn-drome probably caused by olanzapine. J ClinPsychopharmacol. 1999;19:478–479.

4. Wetter TC, Brunner J, Bronisch T. Restless legssyndrome probably induced by risperidonetreatment. Pharmacopsychiatry. 2002;35(3):109.

5. McLean AJ. The use of the dopamine-receptor partial agonist aripiprazole in thetreatment of restless legs syndrome. Sleep.2004;27:1022.

6. The International Restless legs SyndromeStudy Group. Validation of the InternationalRestless Legs Syndrome Study Group ratingscale for restless legs syndrome. Sleep Med.2003;4:121–132.

7. Prueter C, Habermeyer B, Norra C, et al.Akathisia as a side effect of antipsychotictreatment with quetiapine in a patient withParkinson’s disease. Mov Disord. 2003;18:712–713.

8. Lesage S, Hening WA. The restless legssyndrome and periodic limb movement disor-der: a review of management. Semin Neurol.2004;24:249–259.

9. Allen R. Dopamine and iron in the patho-physiology of restless legs syndrome (RLS).Sleep Med. 2004;5:385–391.

10. Kapur S, Zipursky R, Jones C, et al. Apositron emission tomography study of que-tiapine in schizophrenia: a preliminary findingof an antipsychotic effect with only transient-ly high dopamine D2 receptor occupancy.Arch Gen Psychiatry. 2000;57:553–559.

Lisinopril MayAugment

AntidepressantResponse

To the Editors:

The ability of the central nervoussystem to maintain euthymic mood re-quires a number of structural features,including:� Anatomic integrity of the neural cir-

cuitry involved in maintenance ofnormal mood;

� Intact regulation of neurotransmittersynthesis, storage, and release;

� Biochemical regulation of second andsubsequent messenger pathways;

� Functional neuroendocrine regulation;and

� Ability of the blood flow system atthe cellular level to provide glucoseand oxygen for oxidative phosphory-lation and energy to operate.

A number of medications pre-scribed mainly for the treatment of car-diovascular disorders are known to haveeffects on mood and cognition. A histor-ically early antihypertensive, reserpine,was soon shown to have a significanteffect in causing major depressive dis-order (MDD) in some patients. On theother hand, there is no clear relation-ship between whether compounds raiseor lower blood pressure, and the na-ture of influence mood or cognition. Anumber of antihypertensives also donot clearly influence mood or cognition,such as diuretics. Two other antihyper-tensives, clonidine and guanfacine, havebeen reported to improve performancein many patients diagnosed with mood-lifting properties. In replicated trials,verapamil, a calcium-channel blocker,has been shown to have mood-stabilizingproperties.

It is our preliminary observation in10 patients that the angiotensin-convert-ing enzyme (ACE) inhibitor lisinoprilmay also demonstrate mood-stabilizingpotential. It is our purpose in this briefcommunication to describe some ofthe characteristics of these patients andthe results of their treatment. Internistsor cardiologists were treating all 10patients for hypertension primarily,without reference to their mood disor-ders. Nine of these patients had hadMDD. At least 1 of these 9 had a strongfamily history of bipolar disorder. Thetenth patient had documented bipolardisorder.

The diagnoses in all but 2 of thesepatients (both of these patients withMDD) were confirmed by one of us(M.H.), using Diagnostic and StatisticalManual of Mental Disorders, FourthEdition criteria. Thus, the diagnoses in-cluded the required criteria for MDDor bipolar disorder, as well as excludingpatients with alternative, or confound-ing reasons for mood disorders.

Our original observations in thepresent series were the following:

A 50-year-old white woman wasbeing treated for refractory MDD with




bupropion, to a dose of 300 mg everyday (qd). During the course of treat-ment, her blood pressure increased, andbupropion was discontinued. At the sametime, lisinopril, 10 mg qd, was pre-scribed for hypertension by her inter-nist. Contrary to expectation, a monthlater, her mood had improved signifi-cantly and has remained so since.

A 60-year-old white man had MDDand a strong family history of bipolardisorder. Lithium, 1200 mg qd, failedto improve his depression and was dis-continued. His mood began to improvewhen his cardiologist instituted lisino-pril for hypertension (which had pre-ceded the MDD). His mood has beeneuthymic since, on a combination of ser-traline, up to 200 mg qd, and lisinopril,10 mg qd.

A 37-year-old married white manwith hypertension was being treatedwith a diuretic for his hypertension.Months later, treatment of MDD wasinstituted, with bupropion 150 mg qd,a partial success. Some months later,his internist added lisinopril, 10 mg qd(later, to 20 mg qd) to his antihyperten-sive regimen. The patient immediatelynoticed his mood lift. This effect wasreproduced several times later when thedosage of lisinopril was progressivelyraised.

This 35-year-old single black wom-an was being treated for attention defi-cit disorder with stimulant. Althoughclearly having MDD, at first, she de-clined antidepressant treatment. Her in-ternist prescribed lisinopril, 10 mg qd,for hypertension. This had some mod-est mood-elevating effect. She hassince acceded to additional antidepres-sant treatment.

A 53-year-old white man required aregimen of mood stabilizer (tiagabine, 12mg qd), antidepressant (fluoxetine, 20 mgqd), and stimulant (amphetamine salts, 20mg qd) to maintain euthymia over anextended number of years. Eventually,lisinopril, 10 mg qd, was superimposedupon this regimen and provided someadditional improvement of mood.

A 47-year-old married white wom-an came to the psychiatrist on a com-bination of fluoxetine for MDD andanxiety, not otherwise specified, andlisinopril for hypertension. Upward ad-justment of the fluoxetine dosage has

resulted in sustained improvement inmood and anxiety. The exception tothis was a brief period in which her in-ternist lowered her dose of lisinoprilfrom 20 to 10 mg qd. This resulted ina decrement of mood, which was thenoffset by an upward adjustment of thefluoxetine dosage.

A 45-year-old black woman pre-sented with recurrent MDD and hyper-tension. The MDD originally respondedto fluoxetine. Eventually, repeated rein-stitution of antidepressant ceased toimprove her mood. At about this time,her internist instituted lisinopril 10 mgfor the first time to treat her hyperten-sion. She has maintained a reasonablystable mood for 1.5 years on lisinoprilalone.

The remaining patients also hadimproved mood with a combinationof antidepressant and lisinopril. Noneof these patients reported significantside effects from lisinopril treatment,whether alone or in combination withother medications.

DISCUSSIONTo our knowledge, this is the first

report of possible mood-stabilizing ef-fects from the ACE inhibitor, lisinopril.In none of these cases was lisinoprildeliberately instituted to achieve thiseffect. However, it appeared to be toler-ated well.

The relationship between depres-sive illness and hypertension is probablycomplex. Depressive disorders are over-represented in individuals who havehypertension,1 although no clear evi-dence supporting a causal relationshiphas been established. Coffey et al2 haddemonstrated microvascular changes (un-identified bright objects on T2-weightedmagnetic resonance imaging) in de-pressed patients before electroconvulsivetherapy. Dysregulated cortisol releasein depressed patients3 may contributeto increased blood pressure. Robinsonet al4 have elucidated poststroke depres-sions, and hypertension is a clear riskfactor for cerebrovascular disease.

There is at least some indirectevidence, which may indicate a role forACE in the central nervous system as afactor in regulating affective state. Thecentral nervous system is rich in recep-tors for angiotensin II.5 Such receptors

have been demonstrated in preclinicalmodels to be involved in the regulationof dopamine6 and in the regulation ofthe balance between dopamine and se-rotonin.7 Other studies have implicatedangiotensin II receptors in regulation ofsubstance P,8 a current target for novelantidepressant development.9 Centralnervous system ACE activity has beenreported to be increased in suicidalpatients.10

The relationship of ACE inhibi-tion to successful antidepressant aug-mentation remains speculative. Bioamineregulation might account for at leastsome of this effect. An intriguing alter-native is that angiotensin II inhibitionmay promote increases in microvascu-lar cerebral profusion. Such an in-crease in perfusion might facilitateneuronal activation in circuits wherereceptor down-regulation by antidepres-sants acts synergistically to improvemood. Studies in depressed patients sug-gest region-specific cortical activationderangements, which can be reversedby appropriate frequency transcranialmagnetic stimulation.11

At the present time, we believethat it is as yet unjustified to treat pa-tients for mood problems with lisinoprilalone. However, this small series sug-gests that expanding the sample ofpatients with mood disorders who hap-pen then to be placed upon lisinoprilcoincidentally may render a firmer state-ment as to the extent to which lisinoprilmay have mood-stabilizing properties.We are presently expanding the sam-ple under study by having internistsand cardiologists prescribe as they ordi-narily would, and then obtaining patientself-completed mood rating scaled be-fore and during treatment with lisinopril.

This study also opens the questionas to whether other ACE inhibitors or,perhaps, other antihypertensive medica-tions may also have mood-stabilizingproperties.

Marc Hertzman, MD*Lawrence W. Adler, MD

y

Bryan Arling, MDz

Martha Kern, MD**Private Practice, yClinical Insights and

zGeorge Washington University,

Washington, DC

[email protected]




REFERENCES

1. Scherrer JF, Xian H, Bucholz KK, et al. Atwin study of depression symptoms, hyper-tension, and heart disease in middle-aged men.Psychosom Med. 2003;65:548–557.

2. Coffey CE, Figiel GS, Djang WT, et al. Whitematter hyperintensity on magnetic resonanceimaging: clinical and neuroanatomic correlatesin the depressed elderly. J NeuropsychiatryClin Neurosci. 1989;1:135–144.

3. Parker KJ, Schatzberg AF, Lyons DM. Neu-roendocrine aspects of hypercortisolism inmajor depression. Horm Behav. 2003;43(1):60–66.

4. Robinson RG, Kubos KL, Starr LB, et al.Mood disorders in stroke patients. Importanceof location of lesion. Brain. 1984;107(pt 1):81–93.

5. Barnes JM, Steward LJ, Barber PC, et al.Identification and characterization of angio-tensin II receptor subtypes in human brain.Eur J Pharmacol. 1993;230(3):251–258.

6. Brown DC, Steward LJ, Ge J, et al. Abilityof angiotensin II to modulate striatal dopaminerelease via the AT1 receptor in vitro and invivo. Br J Pharmacol. 1996;118(2):414–420.

7. Mendelsohn FA, Jenkins TA, Berkovic SF.Effects of angiotensin II on dopamine andserotonin turnover in the striatum of con-scious rats. Brain Res. 1993;613(2):221–229.

8. Arinami T, Li L, Mitsushio H, et al. Aninsertion/deletion polymorphism in the angio-tensin converting enzyme gene is associatedwith both brain substance contents and affec-tive disorders. Biol Psychiatry. 1996;40(11):1122–1127.

9. Kramer MS, Winokur A, Kelsey J, et al.Demonstration of the efficacy and safety ofa novel substance P (NK1) receptor antagonistin major depression. Neuropsychopharmacol-ogy. 2004;29(2):385–392.

10. Hong CJ, Wang YC, Tsai SJ. Association studyof angiotensin I–converting enzyme polymor-phism and symptomatology and antidepressantresponse in major depressive disorders. JNeural Transm. 2002;109(9):1209–1214.

11. George MS, Wassermann EM, Kimbrell TA,et al. Mood improvement following dailyleft prefrontal repetitive transcranial magneticstimulation in patients with depression: aplacebo-controlled crossover trial. Am J Psy-chiatry. 1997;154(12):1752–1756.

ExtrapontineMyelinolysis

ResemblingNeurolepticMalignant Syndrome

To the Editors:

Early recognition of neurolepticmalignant syndrome (NMS) is crucialin preventing morbidity and mortalityby allowing for discontinuation of trig-

gering drugs and rapid implementationof treatment.1,2 However, a prematurediagnosis of NMS can have significantadverse consequences. Overdiagnosis ofNMS could result in delayed recog-nition and treatment of other seriousmedical disorders that present with sim-ilar symptoms.1–3 Furthermore, an in-correct diagnosis of NMS may inhibitfuture antipsychotic treatment becauseof unwarranted concerns about recur-rent episodes.

The need to carefully consider al-ternative diagnostic causes of fever andencephalopathy is heightened in pa-tients receiving second-generation anti-psychotics (SGAs) because these agentsare less likely to induce NMS.4,5 In ad-dition, SGAs may cause a milder or‘‘atypical’’ form of NMS, which is lessspecific and therefore more difficult todistinguish from other disorders.4,5 Toillustrate these points, we describe aunique case in which NMS was ini-tially considered in a patient receivingquetiapine, who subsequently proved tohave extrapontine myelinolysis (EPM).To our knowledge, this is the first reportof EPM mimicking NMS.

CASE HISTORYA 14-year-old girl, with pan-hypopi-

tuitarism and diabetes insipidus, presentedwith diarrhea, lethargy, and a serum sodiumof 106 mEq/L. She had been receiving nasaldesmopressin 10 Ag twice a day, hydrocor-tisone 10 mg daily, and levothyroxine 150Ag daily. Significant psychiatric historyincluded pervasive developmental disorderwith associated behavioral symptoms. Trialsof haloperidol, risperidone, methylpheni-date, ziprasidone, and chlorpromazine werewithout benefit, but she improved afterquetiapine 100 mg twice a day was initiated2 years previously.

On admission, she received hypertonicsaline; desmopressin was held, and hydro-cortisone 30 mg was administered intrave-nously every 4 hours. Her sodium wascorrected to 140 mEq/L, and she becamealert and fully oriented. Several hours later,she became confused and agitated withintermittent motor stiffening, prompting sus-picion for seizure activity. All medicationswere held except for quetiapine, which wasfinally discontinued after the patient spiked atemperature to 1018F, raising concerns aboutNMS. Physical examination was unremark-able except for cogwheel rigidity of theupper extremities and immobility. Laborato-ry studies revealed a serum creatine kinase

of 580 IU and a white blood cell count of9.7 � 103/mm3. She received lorazepamfor agitation. A head computed tomographyscan was negative for acute changes, andblood and urine cultures did not showabnormalities.

Although the differential diagnosis atthis point included infection, central pontinemyelinolysis (CPM), and postictal confu-sion, NMS became the working diagnosisbecause of the change in mental status,extrapyramidal symptoms, and elevations intemperature and creatine kinase. She re-ceived dantrolene, carvidopa/levodopa, andlorazepam, but her level of consciousnessand neurological status deteriorated. Fourdays later, her creatine kinase increased to2800 IU, but she became afebrile. A serumquetiapine level was nontoxic.

One week later, a magnetic reso-nance imaging of the brain showed bilat-eral high-signal intensity in the caudateand putamen with restricted diffusion. Com-bined with the history of hyponatremia,these magnetic resonance imaging findingsconfirmed the diagnosis of EPM. She even-tually regained normal motor and sensoryfunction. However, treatment of reemergingbehavioral problems with antipsychoticswas complicated by concerns over recur-rence of NMS due to the possibility ofNMS having occurred with quetiapine inthis patient.

DISCUSSIONWe describe a case in which NMS

associated with quetiapine was initiallyconsidered in a patient subsequentlyshown to have EPM. In retrospect, mostof her symptoms could be attributed toEPM.

CPM typically presents with obtun-dation, quadriplegia, and pseudobulbarpalsy after correction of hyponatremia.6

Classic signs of CPM, which reflectdemyelination of the base of the pons,can be missed because of the limitedextent of the lesions or obscured byassociated medical conditions. In addi-tion, variants of the syndrome are in-creasingly recognized with myelinolysisoccurring outside the pons, includingthe basal ganglia (EPM).6 For example,there are several case reports in whichEPM presented with levodopa-responsiveparkinsonism or catatonia.7 – 12 Thesecases showed a consistent clinical pat-tern that may be helpful in diagnosingEPM; each patient deteriorated after avariable period of improvement in mentalstatus after recovery from hyponatremia.




Familiarity with the diagnoses ofEPM and CPM is relevant to psychiatricpracticebecausewater intoxication leadingto hyponatremia is a well-known phenom-enon that requires treatment in a subgroupof patients with chronic psychiatric dis-orders.13

In retrospect, NMS was an unlikelycause of the features in our patient. She didnot exhibit generalized rigidity, extremehyperthermia, labile autonomic signs, orother secondary symptoms of NMS.1,2

The elevation of creatine kinase was non-specific and does not confirm the diag-nosis of NMS.1,2 Although SGAs mayproduce a mild form of NMS, we believethat quetiapine was not a factor in thiscase.4,5 She received quetiapine for 2 yearswithout incident and previously receivedmore potent antipsychotics without NMS.In fact, there are still few unequivocalreports implicating quetiapine monother-apy in causing NMS.14 Theoretically,the low affinity and fast dissociation ofquetiapine in relation to dopamine recep-tors should diminish the risk of NMS withthis agent.15

We previously proposed that theliability of SGAs in causing extrapyrami-dal symptoms, including NMS, may be sig-nificant only when they are administeredin high-risk populations.16 Risk factorsfor the development of NMS may in-clude hyponatremia and preexisting ba-sal ganglia dysfunction.1,2,17,18 It followsthat patients who sustain basal gangliadamage associated with EPM afterhyponatremia and who are also receivingSGAs could be at increased risk ofdeveloping NMS. Hence, it can be dif-ficult to distinguish EPM from NMS inthese patients, and we cannot excludeentirely the possibility of a transient NMS-like effect of quetiapine in our case.

In conclusion, we report the firstcase of EPM mimicking NMS. Al-though these are rare conditions, EPMand CPM have been associated previ-ously with catatonia and parkinsonismand should also be included in thedifferential diagnosis of NMS. Further-more, this case underscores the impor-tance of identifying serious medicaldisorders as well as NMS in patientspresenting with fever and neurologicalchanges during treatment with antipsy-chotic drugs. Errors in excluding eithercategory of illness prematurely can befatal.

Karen E. Groff, MD*Christopher Lam, MD*yStanley N. Caroff, MD*z

*Department of Psychiatry, University of

Pennsylvania School of Medicine;

yChildren’s Hospital of Philadelphia and

zDepartment of Veterans Affairs

Medical Center, Philadelphia, PA

REFERENCES

1. Caroff SN, Mann SC. Neuroleptic malignantsyndrome. Med Clin North Am. 1993;77(1):185–202.

2. Caroff SN. Neuroleptic malignant syndrome. In:Mann SC, Caroff SN, Keck PE Jr, et al. eds.Neuroleptic Malignant Syndrome and RelatedConditions. 2nd ed. Washington, DC: AmericanPsychiatric Press, Inc; 2003:1–44.

3. Sewell DD, Jeste DV. Distinguishing neuro-leptic malignant syndrome (NMS) from NMS-like acute medical illnesses: a study of 34cases. J Neuropsychiatr Clin Neurosci. 1992;4:265–269.

4. Caroff SN, Mann SC, Campbell EC. Aytpicalantipsychotics and neuroleptic malignant syn-drome. Psychiatr Ann. 2000;30:314–321.

5. Hasan S, Buckley P. Novel antipsychotics andthe neuroleptic malignant syndrome: a re-view and critique. Am J Psychiatry. 1998;155:1113–1116.

6. Victor M, Ropper AH. Adams and Victor’sPrinciples of Neurology. 7th ed. New York:McGraw-Hill; 2001:1193–1195.

7. Stam J, van Oers MHJ, Verbeeten B. Recov-ery after central pontine myelinolysis. JNeurol. 1984;231:52–53.

8. Dickoff DJ, Raps M, Yahr MD. Striatal syn-drome following hyponatremia and its rapidcorrection. Arch Neurol. 1998;45:112–114.

9. Tinker R, Anderson MG, Anand P, et al.Pontine myelinolysis presenting with acuteparkinsonism as a sequel of corrected hypo-natremia. J Neurol Neurosurg Psychiatry.1990;53:87–89.

10. Sadeh M, Goldhammer Y. Extrapyramidalsyndrome responsive to dopaminergic treat-ment following recovery from central pontinemyelinolysis. Eur Neurol. 1993;33:48–50.

11. Nagamitsu S, Matsuishi T, Yamashita Y, et al.Extrapontine myelinolysis with parkinsonismafter rapid correction of hyponatremia: highcerebrospinal fluid level of homovanillic acidand successful dopaminergic treatment. JNeural Transm. 1999;106:949–953.

12. Koussa S, Nasnas R. Catatonia and parkin-sonism due to extrapontine myelinolysisfollowing rapid correction of hyponatremia.J Neurol. 2003;250:103–105.

13. Illowsky BP, Kirch DG. Polydipsia andhyponatremia in psychiatric patients. Am JPsychiatry. 1988;145:675–683.

14. Sing KJ, Guilaume MGI, Ramaekers MA, et al.Neuroleptic malignant syndrome and quetia-pine. Am J Psychiatry. 2002;159:149–150.

15. Kapur S, Seeman P. Does fast dissociationfrom the dopamine D2 receptor explain theaction of atypical antipsychotics? A new hy-pothesis. Am J Psychiatry. 2001;158:360–369.

16. Caroff SN, Mann SC, Campbell EC, et al.Movement disorders associated with atypicalantipsychotic drugs. J Clin Psychiatry. 2002;63(suppl 4):12–19.

17. Looi JCL, Cubis JC, Saboisky J. Hyponatre-mia, convulsions and neuroleptic malignantsyndrome in a male with schizoaffective dis-order. Aust N Z J Psychiatry. 1995;29:683–687.

18. Elizalde-Sciavolino C, Racco A, Proscia-LietoT, et al. Severe hyponatremia, neuroleptic ma-lignant syndrome, rhabdomyolysis and acuterenal failure: a case report. Mt Sinai J Med.1998;65:284–288.

Impaired GlucoseHomeostasis After

Imipramine Intake in aDiabetic Patient

To the Editors:

The tricyclic antidepressant (TCA)imipramine is primarily indicated forthe treatment of depression, but in prac-tice, it is also used for other disorders.We describe a patient with type II dia-betes mellitus, in whom changes in in-sulin need were closely associated withthe use and dose changes of imipramine,which was prescribed for the treatmentof urinary incontinence.

CASE DESCRIPTIONA 62-year-old woman was treated for

several years for type II diabetes mellituswith the oral hypoglycemic agent glimepir-ide (4 mg daily). Since November 27, 1998,she additionally used NPH insulin beforethe night, but the correction of the bloodglucose was only moderately successful be-cause the average HbA1C level was 9.0%(normal value, 4.4%–6.1%). On July 13,2000, a urologist prescribed imipramine(25 mg at bedtime) because of urinary in-continence. On September 12, 2000, thediabetologist switched the oral hypoglyce-mic treatment in combination with bedtimeNPH insulin to intensive insulin therapy(multiple daily injection regimen) with bloodglucose self-monitoring and algorithm-basedadjustment of insulin dose. Glucose mea-surements as well as the amount of injectedinsulin were monitored and registered by thepatient on a daily basis in a ‘‘diabetes diary.’’From September 12, 2000, to May 29, 2002,the average insulin dose was 81 IU/d, and theHbA1C level was 8.4% on August 16, 2001,and 6.8 % on March 13, 2002. On May 30,2002, the dose of imipramine was increasedfrom 25 mg once daily to 25 mg twice daily




(in the morning and at bedtime). After 1month, the need for insulin increased. Be-tween May 30, 2002, and January 10, 2003,the average insulin dose was 95 IU/d, cor-responding to an increase of 17% of in-sulin dose relative to the period that she usedimipramine 25 mg daily. The HbA1C levelwas 7.5% on October 9, 2002. In agree-ment with her urologist, the patient taperedthe use of imipramine between January 11,2003, and February 5, 2003. On February 6,2003, the use of imipramine was discontin-ued completely. During the tapering period,the average insulin dose dropped to 91 IU/d,and after complete discontinuation of imip-ramine on February 6, 2003, the average in-sulin dose decreased further to 48 IU/d. Thiscorresponded to a decrease of 51% relativeto the period when imipramine 50 mg dailywas used. On March 14 and July 23, 2003,the HbA1C level was 6.8% and 7.4%, re-spectively. Further analysis revealed thatthe changes in the amounts of insulin ad-ministered particularly concerned the nightlydose. Concomitantly used medication in-cluded atenolol, lisinopril, simvastatin, andclodronic acid which did not, besides glime-piride (discontinued on September 12, 2000),essentially change during the entire period.During the entire period, no substantialchanges in body weight occurred, and thepatient was not aware of changes in eatingbehavior and pattern.

DISCUSSIONIn our patient, there were 3 inter-

ventions regarding imipramine: (1) start-ing imipramine, (2) dose increase ofimipramine, and (3) discontinuation ofimipramine. All interventions were fol-lowed by changes in the daily amountsof insulin requirements. Especially re-garding the second and third interven-tions, the time relationship between theintervention and the change in insulinrequirement is highly suggestive of apharmacological effect of imipramine.We did not identify other factors thatcould explain the change in require-ments of insulin dose. For the first in-tervention, the time relationship is lessobvious. Nine weeks after the first in-tervention, the diabetologist switchedfrom a fixed regimen of an oral agentcombined with bedtime NPH insulin toan intensive insulin-dosing scheme withself-monitoring and adjustment of in-sulin dose on measured blood glucoselevel. The measured HbA1C level wasmoderately high before the start of imip-ramine. Therefore, the switch to the in-

tensive insulin-dosing scheme probablywas not caused by the use of imipra-mine. Earlier evidence in literature sug-gests that imipramine, as well as otherantidepressants, may affect glucose ho-meostasis.1–13 A strong feature of thiscase is that imipramine was not used fordepressive disorder, which itself may beassociated with changes in food intakeand altered glucose homeostasis, but forurinary incontinence. Moreover, thiscase report is unique because we wereable to illustrate in detail the changes ininsulin requirements, which are a verysensitive marker for altered glucose ho-meostasis. In addition, we could show adose-response relationship, as well as adechallenge. In at least 2 of 3 interven-tions regarding imipramine, we founda strong time relationship between theuse and dose of imipramine and the in-sulin requirement (Fig. 1). Theoretically,glucose homeostasis could be affectedby a direct effect on blood glucose lev-els and/or insulin levels and/or insulinsensitivity. Several mechanisms havebeen described in literature that maybe involved in imipramine-induced glu-cose deregulation. Like most TCAs,imipramine inhibits the synaptic reup-take of both norepinephrine and seroto-nin (5-hydroxytryptamine [5-HT]) atnerve terminals. Norepinephrine maystimulate glycogenolysis and gluconeo-genesis resulting in raised blood glu-cose levels1 or reduced insulin release.2

Because these effects occur in a shorttime span, these mechanisms could not

explain the time gap between the inter-ventions with imipramine and effectson glucose homeostasis. Another mech-anism encompasses a blockade of TCAsof M3 receptors in beta cells, resultingin suppression of insulin secretion andincreased leptin levels, also inhibiting in-sulin secretion by the pancreas.4,14 Inrats, imipramine induced a dose-depen-dent decrease in glucose-stimulated in-sulin secretion which appears to bemediated by inhibition of voltage-sensi-tive Ca2+ channels5. In mice studies,imipramine-induced hyperglycemia hasbeen related to inhibition of the central5-HT2C receptor.7 Blockade of the 5-HT2C receptor may result in craving for‘‘sweets’’ and ‘‘carbohydrates’’ and in-crease body weight.15 TCAs also blockhistaminic-, M3-, and a1-adrenergic re-ceptors, causing adverse drug reactionssuch as dry mouth, followed by drinkinglarge quantities of soft drinks. Botheffects on food intake will complicatethe diabetic’s ability to follow a con-trolled diet. In this patient, however, nochanges in body weight occurred. Final-ly, TCAs may enhance tissue sensitivityto insulin, thereby reducing compensa-tory beta-cell hypersecretion of insulinand, therefore, an increase of externalinsulin need.4 Tissue sensitivity doesnot necessarily go together with weightchanges and is altered by delay whichcould explain the time gap betweenthe interventions with imipramine andchanges insulin need without changes inbody weight.

FIGURE 1. Time relationship between insulin dose and imipramine dose.




This case report illustrates thatimipramine can affect glucose homeo-stasis in diabetic patients. Although themechanism is still unclear, physicianshave to be conscious that imipramine,and probably other antidepressants, canimpair diabetes control in some sensitivepatients. Further research is needed toelucidate the mechanism for this effectand to find out which patients are at risk.

Hieronymus J. Derijks, PharmD*yFred H.P. De Koning, PhD*z

Ronald H. Meyboom, MD, PhD*xEibert R. Heerdink, PhD*

Pieter F.M.J. Spooren, MD, PhDkAntoine C.G. Egberts, Prof*{

*Department of Pharmacoepidemiology and

Pharmacotherapy, Utrecht Institute for

Pharmaceutical Sciences, Utrecht

University, Utrecht, The Netherlands;

yDepartment Clinical Pharmacy,

Maaslandziekenhuis, Sittard,

The Netherlands; zInstitute of Kring

Pharmacies, Den Bosch, The Netherlands;

xThe Uppsala Monitoring Centre,

WHO Collaborating Centre for

International Drug Monitoring, Uppsala,

Sweden; kTweeSteden Hospital,

Department of Internal Medicine, Tilburg,

The Netherlands and {Hospital Pharmacy

Midden Brabant, TweeSteden

Hospital, and St Elisabeth Hospital,

Tilburg, The Netherlands

[email protected]

REFERENCES

1. Goodnick PJ, Henry JH, Buki VM. Treatmentof depression in patients with diabetes mellitus.J Clin Psychiatry. 1995;56(4):128–136.

2. Goodnick PJ. Use of antidepressants in treat-ment of comorbid diabetes mellitus and de-pression as well as in diabetic neuropathy.Ann Clin Psychiatry. 2001;13(1):31–41.

3. Ghaeli P, Shahsavand E, Mesbahi M, et al.Comparing the effects of 8-week treatmentwith fluoxetine and imipramine on fastingglucose of patients with major depressivedisorder. J Clin Psychopharmacol. 2004;24(4):386–388.

4. Moosa MY, Panz VR, Jeenah FY, et al.African women with depression: the effectof imipramine and fluoxetine on body massindex and leptin secretion. J Clin Psycho-pharmacol. 2003;23(6):549–552.

5. Antoine MH, Gall D, Schiffmann SN, et al.Tricyclic imipramine reduces the insulinsecretory rate in islets cells of Wistar albinorats through a calcium antagonistic action.Diabetologia. 2004;47(5):909–916.

6. Lustman PJ, Griffith LS, Clouse RE, et al.Effects of nortriptyline on depression and

glycemic control in diabetes: results of adouble-blind, placebo-controlled trial. Psy-chosom Med. 1997;59(3):241–250.

7. Sugimoto Y, Inoue K, Yamada J. Involve-ment of 5-HT(2) receptor in imipramine-induced hyperglycemia in mice. Horm MetabRes. 2003;35(9):511–516.

8. Marley J, Rohan A. Mianserin-induced hyper-glycaemia. Lancet. 1993;342(8884):1430–1431.

9. Chen JL, Spinowitz N, Karwa M. Hyper-triglyceridemia, acute pancreatitis, and dia-betic ketoacidosis possibly associated withmirtazapine therapy: a case report. Pharma-cotherapy. 2003;23(7):940–944.

10. Warnock JK. Nefazodone-induced hypogly-caemia. Am J Psychiatry. 1997;154:288–289.

11. Petty KJ. Hyperglycemia associated withparoxetine. Ann Intern Med. 1996;125(9):782.

12. Sansone RA, Sansone LA. Sertraline-inducedhyperglycemia: case report. Int J PsychiatryMed. 2003;33(1):103–105.

13. Tavakoli SA, Arguisola MS. Diabetic keto-acidosis in a patient treated with olanzapine,valproic acid, and venlafaxine. South Med J.2003;96(7):729–730.

14. Gilon P, Henquin JC. Mechanisms and phys-iological significance of the cholinergic con-trol of pancreatic beta-cell function. EndocrRev. 2001;22(5):565–604.

15. Hoyer D, Hannon JP, Martin GR. Molecular,pharmacological and functional diversity of5-HT receptors. Pharmacol Biochem Behav.2002;71(4):533–554.

Diplopia WithCitalopram

A Case Report

To the Editors:

Citalopram is an antidepressantwhich effects serotonergic neurotrans-mission through potent and selectiveinhibition of serotonin reuptake. Neuro-endocrine studies suggest decrease ofserotonergic responsively in patientswith major depression. Citalopram is(1) superior to placebo in the treatmentof depression; (2) is similar to that of thetricyclic and tetracyclic antidepressantsand to other selective serotonin reup-take inhibitors; and (3) is safe and welltolerated in the therapeutic dose range of20 to 60 mg/d.1 Data from 3107 patientsfrom 24 clinical trials2 showed that nau-sea, dry mouth, somnolence, increasedsweating, tremor, diarrhea, and ejacula-tion failure, mostly of mild to moderateseverity, occurred with a significant fre-quency with citalopram.

However, bruxism, cutaneous re-actions, torsade de pointes, bradycardia

and hypotension, hyponatermia second-ary to syndrome of inappropriate secre-tion of antidiuretic hormone, priapism,panic attacks, palpebral twitching, pho-topigmentation, galactorrhea, and hyper-triglycemia have been reported rarelywith citalopram.

Diplopia is an unusual phenome-non that may occur after citalopramingestion.3 Diplopia is defined as dou-ble vision. Binocular diplopia is a typeof double vision that is eliminated wheneither eye is occluded.4 Causes of bin-ocular diplopia are isolated third, fourth,and sixth cranial nerve palsy, orbital dis-eases, cavernous sinus/superior orbitalfissure syndrome, posttraumatic status,intranuclear ophthalmoplegia, vertebro-basilar artery insufficiency, other cen-tral nervous system lesions, and spectacleproblem.4

Here, we report a 28-year-old manwho developed diplopia after citalopramingestion. To our knowledge, there is noreport on the interval after which thisside effect is eliminated after discontin-uation of the drug.

CASE REPORTMZ was a 28-year-old, single, medi-

cal student, who lived in a university dor-mitory, and met Diagnostic and StatisticalManual of Mental Disorders, Fourth Editionnonpsychotic major depressive disorder cri-terion for 6 months before his referral. Therewas no history of past psychiatric problemin childhood and adolescence. His conditiondeteriorated day by day that led him to dropseveral courses. He was urged by the univer-sity to visit our clinic.

After his referral, he was examinedin our clinic. The patient’s score on 21-ItemHamilton Depression Scale was 34. Therewas a positive history of major depressivedisorder in his father.

Citalopram, 20 mg/d in the morning,was prescribed to him. After 12 days, inca-pacitating diplopia appeared. This conditionwas accompanied with diarrhea and mem-ory complaints. He was frightened and de-scribed the condition as ‘‘It is terrible. I seeeverything in 2. I don’t know which is realand which is artificial. If I concentrate andclose one eye, I will be able to differentiatethe real one.’’

There was no positive finding in neu-rological and ophthalmologic consultations.Citalopram, the only medication he wastaking, was discontinued. After 60 hours, thediplopia disappeared completely, and he feltrelaxed.




DISCUSSIONAlthough citalopram has been

reported to be a safe and well-toleratedselective serotonin reuptake inhibitorantidepressant,1,2 sometimes, rare phe-nomenon and dangerous adverse reac-tions may occur. Therefore, citalopramshould be prescribed with caution.

In case of diplopia, neurologicalas well as ophthalmologic causes shouldbe considered. Here, our patient wasexamined by a neurologist and an oph-thalmologist separately. The pupil sizeand the reaction to light were normal. Noweakness was noted in the extraocularmuscles when each eye was examinedindividually. The rest of the neurologicaland ophthalmologic examinations wasnormal. Brain computed tomographyand magnetic resonance imaging werenormal. The blood chemistry analyseswere also normal.

Sometimes, patients with conver-sion disorder may present with symptomssuch as diplopia, but here, the diplopiawas drug-induced and disappeared aftercitalopram was discontinued.

Arash Mowla, MD

Ahmad Ghanizadeh, MD

Hamid Ashkani, MDDepartment of Psychiatry, Shiraz University

of Medical School, Shiraz, Iran

[email protected]

REFERENCES

1. Keller MB. Citalopram therapy for depres-sion: a review of 10 years of European ex-perience and data from U.S. clinical trials. ClinPsychiatry. 2000;61(12):896–908.

2. Muldoon C. The safety and tolerability ofcitalopram. Int Clin Psychopharmacol. 1996;11(1):35–40.

3. Dorell K, Cohen M, Huprikar SS, et al.Citalopram-induced diplopia. Psychosomatics.2005;46(1):91–93.

4. Rhee D, Pyfer M. The Wills Eye Manual. 3rded. Philadelphia, PA: Lippincott Williams &Wilkins; 1999:2–3.

How Safe Is Long-termBenzodiazepine

Pharmacotherapy?

To the Editors:

One of the more contentious issuesin clinical practice concerns the long-

term use of benzodiazepines. Manyauthorities have cautioned against suchuse because of concerns about depen-dence, cognitive impairment, and risk offalls in the elderly and have recommen-ded substituting selective serotonin re-uptake inhibitors and related agents asfirst-line treatment for long-term man-agement of anxiety disorders and non-pharmacological treatment for chronicinsomnia.1–4 Other authorities have ex-pressed the contrary point of view thatbenzodiazepines are relatively safe inlong-term use if patients are carefullymonitored.5 Despite the voluminous lit-erature on benzodiazepines, there havebeen, to my knowledge, no systematicstudies on the frequency and type of ad-verse events occurring in patient pop-ulations receiving long-term (ie, >6months) benzodiazepine pharmacother-apy. The authors of a major reviewarticle have said, ‘‘While we have animmense body of knowledge aboutmany of the effects of benzodiazepines,we have very little information abouteither the benefits or the risks of long-term use.’’6

I recently reviewed the records ofmy last 22 years of general psychiat-ric practice in a multispecialty privateclinic affiliated with a university hospi-tal health system and wish to share thehighlights of my experience with respectto the safety of this form of treatment.� A total of 836 patients received con-

tinuous benzodiazepine pharmacother-apy for 6 months or longer. Duration ofuse ranged from 6 months to 21 years.Median duration was 3.5 years.

� Five hundred sixty-one patients re-ceived benzodiazepines for daytimetreatment of anxiety symptoms and80 for nighttime treatment of insom-nia, and 195 received treatment forboth anxiety and insomnia.

� All but 85 patients received concom-itant psychiatric medication.

� In my opinion, the most importantfinding was that only 7 patients (0.8%)had to have benzodiazepines discon-tinued because of adverse events,6 patients for alcohol abuse–relatedevents, and 1 patient for abusing mul-tiple drugs. None of these resultedin death or serious injury.

� One hundred forty-four patients (17%)had adverse events not serious enoughto require discontinuation of benzodi-

azepine treatment. Of these, 116 had aconcomitant substance abuse diagno-sis, and 28 had a psychiatric diagnosisalone at the time of initial evaluation.

� In the dual diagnosis group, the com-monest types of adverse events wereepisodes involving abuse of alcoholand street drugs. Other adverse eventsincluded motor vehicle accidents (noserious bodily injuries) and abuse oroverdose of benzodiazepines. Therewas an incident each of the following:prescription forgery, fall, suicide, sud-den death (no autopsy), early refill ofbenzodiazepine prescription, and fall-ing asleep with lighted cigarette (noinjury).

� In patients without a substance abusediagnosis, the commonest adverseevents were abuse or overdose ofbenzodiazepines or other drugs, orseeking early prescription refill (19),motor vehicle accidents (6), falls (4),and arrests for driving under the in-fluence (2). Some of these patientsprobably were incorrectly diagnosedas not having a substance abuse prob-lem at the time of initial evaluation.

� In most cases, it was impossible todetermine whether the adverse eventswere caused by benzodiazepines orwere related to the underlying psy-chopathology (particularly substanceabuse disorders) and would haveoccurred anyway had the patient notbeen receiving long-term benzodiaz-epine therapy.

� Three percent of patients had beenprescribed at some time during theirtreatment a dose higher than the man-ufacturer’s recommended upper limitfor daytime use. Fifteen percent ofpatients had been prescribed a dosehigher than the manufacturer’s rec-ommended upper limit for hypnoticuse. None of these latter were higherthan twice the upper limit, for exam-ple, 60 mg of flurazepam.

Assuming that the experience inmy practice is typical of clinical prac-tices in general (which admittedly maynot be the case), I conclude that long-term benzodiazepine therapy is relative-ly safe. In choosing a long-term treat-ment for anxiety symptoms, for manypatients, the relative safety of benzo-diazepines may compare favorably withthat of selective serotonin reuptake in-hibitors and related drugs, taking into




consideration the well-known side ef-fects of the latter such as weight gainand sexual dysfunction, as well as theirlack of effectiveness in many cases.With respect to the treatment of chronicinsomnia, the relative safety of benzo-diazepines needs to be balanced againstthe frequent ineffectiveness of non-pharmacological treatment and the rel-atively high rate of accidents andinjuries occurring in patients havinguntreated chronic insomnia.7

Richard A. Schwartz, MDDepartment of Psychiatry, Case Western

Reserve University School of Medicine

Cleveland, OH

[email protected]

REFERENCES

1. American Psychiatric Association. Practiceguidelines for the treatment of patients withpanic disorder. Am J Psychiatry. 1998;155(5 suppl):1–34.

2. Ballenger JC, Davidson JR, Lecrubier Y, et al.Consensus statement on generalized anxietydisorder from the International ConsensusGroup on Depression and Anxiety. J ClinPsychiatry. 2001;62(suppl 11):53–58.

3. Olfson M, Marcus SC, Wan GJ, et al. Nationaltrends in the outpatient treatment of anxietydisorders. J Clin Psychiatry. 2004;65:1166–1173.

4. National Institutes of Health Consensus Devel-opment Conference. Drugs and insomnia: theuse of medications to promote sleep. JAMA.1984;251:2410–2414.

5. Uhlenhuth EH, Balter MB, Ban TA, et al.International study of expert judgment ontherapeutic use of benzodiazepines and otherpsychotherapeutic medications: IV. Therapeu-tic dose dependence and abuse liability ofbenzodiazepines in the long-term treatmentof anxiety disorders. J Clin Psychopharmacol.1999;19(6 suppl 2):23–29.

6. Woods JH, Winger G. Current benzodiazepineissues. Psychopharmacology. 1995;118:107–115. [p112].

7. Balter MB, Uhlenhuth EH. New epidemiologicfindings about insomnia and its treatment. J ClinPsychiatry. 1992;53(12 suppl):34–39.

UnderstandingToxidromes:

Serotonin Toxicity

A Commentary onMontanes-Rada et al

To the Editors:

There have been some interestingnew articles from toxicology research

group (hunter area toxicology service) ofWHYTE1,2 of providing data from morethan 2200 prospectively documented over-doses of serotonergic drugs. These havebeen published in a series of articles1–4

that add greatly to our understanding ofserotonin syndrome (SS) or serotonintoxicity (ST), as some authors now preferto call it. These may be considered, inaddition to the discussion by Montanes-Rada et al,5 who have made some state-ments that can be further clarified in thelight of this additional data.

The key concept that Whyte’sdata confirm is that SS is better con-ceptualized as a form of poisoning (ie,ST), because progressively increasingserotonergic effects are an inevitableresult of the ingestion of overdoses ofselective serotonin reuptake inhibitorsand serotonergic drug combinations(see Refs 6 and 7 for a full discussionof the spectrum concept of ST and theanimal and human evidence that sup-ports it). Whyte’s data from humanoverdoses demonstrate that approxi-mately 15% of selective serotonin reup-take inhibitors–alone overdoses (whichare only ever mild to moderate in se-verity), increasing to 50% for combinedoverdoses of monoamine oxidase inhib-itors + selective serotonin reuptake in-hibitors (which are also much moresevere), exhibit ST (ie, have sufficientserotonergic symptoms to cross the ar-bitrary ‘‘diagnostic’’ threshold). Thereis a dose-related increase in severity ofserotonergic symptoms. This is whytoxicology authors are referring to thecondition as ST; that emphasizes itsinevitable dose-related relationship tothe ingestion drugs (ie, poisoning),whereas the term SS implies that it isidiosyncratic, like neuroleptic malig-nant syndrome (NMS). It may be notedthat this does not exclude the observa-tion that some serotonergic effects thatcan occur with therapeutic doses mayhave idiosyncratic characteristics, asclinical experience suggests.

However, the criteria for SSresolve the problem excluding SSif an antidopaminergic drug isadministered.

There is a strong argument that afundamental difference between NMSand ST exists. NMS is a rare idiosyn-

cratic reaction that is quite different tothe usual toxic effects of neurolepticoverdose.8 Conversely, ST, after inges-tion of serotonergic drugs, is a commonprogression of serotonergic effects,merging into toxicity. ST is much morestrongly statistically associated withserotonergic drugs than NMS is withneuroleptic drugs. The Bayesian theoryof ‘‘prior probability’’ indicates thatknown ingestion of serotonergic drugsis the more important piece of data thatwould help predict the correct diagnosisin doubtful cases.9

The original article of Sternbach10

made suggestions about diagnostic cri-teria, which although helpful for ourinitial understanding, were neverthelessunvalidated proposals derived from arather selected sample. He duly notedthe likelihood of reporting bias in hisseries, but that qualification has some-times received insufficient subsequentattention. In the light of the quality datafrom Whyte’s hunter area toxicologyservice series, it is now also appropriateto review his other suggestion—that pre-vious ingestion of a neuroleptic shouldbe an exclusion criteria for a diagnosisof ST. A special argument is necessaryto logically justify assigning a hierar-chical precedence to one feature overanother, such as the previous ingestionof a particular drug: such a case ofprior probability is stronger for seroto-nergic drugs, especially mixtures ofmonoamine oxidase inhibitors and selec-tive serotonin reuptake inhibitors, whichwe now know have approximately a50% prior probability for inducingST. This is reflected in the diagnosticdecision (CART) rules derived fromWhyte’s data3: rule 1 is if serotonergicdrugs have been ingested, and clonus ispresent, a definite diagnosis of ST ishighly probable (sensitivity, 84%; andspecificity, 97%).

There is no established criteriafor serotonin syndrome. . .the dif-ferential diagnosis between SS andNMS is difficult as there is consid-erable overlap in the symptoms.

Whyte’s data have now establishedreliable criteria and diagnostic decisionrules for ST,3 as above. Other evidenceindicates that ST and NMS are distinct,not only in their in clinical symptoms, but




also in etiology, course, and treatmentresponse: these differences have been dis-cussed in detail elsewhere.7,11–13 In sum-mary, they are as follows:(a) ST is caused by serotonergic drugs

(frequently, predictably, and dose-related). NMS occurs in associationwith neuroleptics (rarely, idiosyn-cratically, and not dose-related).

(b) ST, rapid onset, and progression(hours). NMS, slow onset, and pro-gression (days).

(c) ST, hyperkinesia, and hyperreflexia/clonus, pyramidal rigidity. NMS, bra-dykinesia, and extrapyramidal rigidity.

There will always be situations inclinical medicine where confusion ispossible, but that should not obscure thefact that, in a great majority of cases,the distinction can be made clearly andconfidently.

P. Ken Gillman, MRC Psych, MBDepartment of Clinical Neuropharmacology,

Pioneer Valley Private Hospital,

Queensland, Australia

[email protected]

REFERENCES

1. Whyte IM. Neuroleptic malignant syndrome.In: Dart RC, ed. Medical Toxicology. 3rd ed.Baltimore, Md: Lippincott Williams & Wilkins;2004:101–103.

2. Whyte IM. Serotonin toxicity (syndrome). In:Dart RC, ed. Medical Toxicology. 3rd ed.Baltimore: Lippincott Williams & Wilkins;2004:103–106.

3. Dunkley EJC, Isbister GK, Sibbritt D, et al.Hunter Serotonin Toxicity Criteria: a simpleand accurate diagnostic decision rule for sero-tonin toxicity. Q J Med. 2003;96:635–642.

4. Isbister GK, Hackett LP, Dawson AH, et al.Moclobemide poisoning: toxicokinetics andoccurrence of serotonin toxicity. Br J ClinPharmacol. 2003;56:441–450.

5. Montanes-Rada F, Bilbao-Garay J, de Lucas-Taracena M, et al. Venlafaxine, serotoninsyndrome, and differential diagnoses. J ClinPsychopharmacol. 2005;25(1):101–102.

6. Gillman PK, Whyte IM. Serotonin syndrome.In: Haddad P, Dursun S, Deakin B, eds.Adverse Syndromes and Psychiatric Drugs.Oxford, UK: Oxford University Press; 2004:37–49.

7. Gillman PK. Serotonin syndrome: history andrisk. Fundam Clin Pharmacol. 1998;12(5):482–491.

8. Isbister GK, Whyte IM, Smith AJ. Olanzapineoverdose. Anaesthesia. 2001;56(4):400–401.

9. Buckley NA, Whyte IM, Dawson AH.Diagnostic data in clinical toxicology—should we use a Bayesian approach? J ToxicolClin Toxicol. 2002;40(3):213–222.

10. Sternbach H. The serotonin syndrome. Am JPsychiatry. 1991;148:705–713.

11. Gillman PK. The serotonin syndrome and itstreatment. J Psychopharmacol (Oxf). 1999;13(1):100–109.

12. Whyte IM. Serotonin uptake inhibitors. In:Dart RC, ed. Medical Toxicology. 3rd ed.Baltimore, MD: Lippincott Williams & Wilkins;2004:843–851.

13. Boyer EW, Shannon M. The serotoninsyndrome. N Engl J Med. 2005;352(11):1112–1120.

Comments on Articleby Albers et al

‘‘Low-Dose Fluvoxamineas an Adjunct to ReduceOlanzapine Therapeutic

Dose Requirements:A Prospective

Dose-Adjusted DrugInteraction Strategy’’

To the Editors:

We read with great interest thearticle by Albers et al.1 The authors sug-gest that a 26% reduction of olanzapinedose requirement may be achieved bycoadministration of 25 mg/d fluvox-amine. In the section of combiningantidepressant with olanzapine therapy,there are some issues that should beaddressed.

In pharmacokinetic aspect, fluvox-amine alters the olanzapine metabolismthrough the cytochrome P450 1A2 inhi-bition, and its inhibitory effect seems tobe dose-dependent.2,3 We recently con-ducted a study addressing the dose-dependent alternations in the olanzapinepharmacokinetics by fluvoxamine coad-ministration.4 Ten male schizophrenicsmokers were administrated a singledose of olanzapine 10 mg at baseline,followed by 2 weeks of fluvoxamine 50mg/d and another 2 weeks of fluvox-amine 100 mg/d. Olanzapine 10 mg wasgiven at day 10 during each fluvoxaminetreatment. After pretreatment of fluvox-amine, the area under the curve, maximalplasma concentration, and half-life ofolanzapine were significantly increasedby 30% to 55%, 12% to 64%, and 25%to 32%, respectively. Therefore, fluvox-amine showed a dose-dependent inhibi-tion of olanzapine N-demethylation.

Several pilot studies reported thataddition of fluvoxamine to antipsychotictreatment was well tolerated and couldimprove the psychopathology of schizo-phrenic patients.5,6 Recently, we con-ducted a study regarding the inhibitoryeffect of fluvoxamine on the clozapine-related weight gain and metabolicdisturbances.7 Plasma levels of N-desmethyclozapine, but not clozapine,are associated with the increases inweight and serum glucose and triglycer-ide level. Olanzapine treatment was alsoassociated with weight gain and elevatedlevels of insulin, leptin, and blood lipidsas well as insulin resistance. Metabolicabnormalities and insulin resistancewere associated with both clozapineand olanzapine treatments. Levels ofinsulin and triglycerides increased byincreasing clozapine serum concentra-tion and by increasing ratio of olanza-pine to N-desmethylolanzapine wererevealed recently.8 It is suggested thatthe metabolite N-desmethylolanzapine,but not olanzapine, has a normaliz-ing effect on the metabolic abnormali-ties.9 In the study of Albers et al,1 the40-N-desmethylolanzapine/olanzapinemetabolic ratio decreased by 44% at week6. In spite of no significant change inmetabolic indices in this 6-week study,the effects on lipid metabolism may bemore apparent in the long run. Therefore,this comedication strategy might carrythe risk of aggravating the olanzapine-induced metabolic disturbances.

Although the addition of fluvox-amine to olanzapine medication is welltolerated and critical side effects areabsent, this augmentation strategy shouldbe applied only after careful evaluationof risks versus benefits.

Chih-Chiang Chiu, MD*Mong-Liang Lu, MD, MSy

*Department of Psychiatry, Songde Branch,

Taipei City Hospital and yDepartment of

Psychiatry, Taipei Medical University–Wan

Fang Hospital, Taipei, Taiwan

[email protected]

REFERENCES

1. Albers LJ, Ozdemir V, Marder SR, et al. Low-dose fluvoxamine as an adjunct to reduceolanzapine therapeutic dose requirements: a pro-spective dose-adjusted drug interaction strategy.J Clin Psychopharmacol. 2005;25:170–174.




2. Hiemke C, Peled A, Jabarin M, et al. Fluvox-amine augmentation of olanzapine in chronicschizophrenia: pharmacokinetic interactions andclinical effects. J Clin Psychopharmacol. 2002;22:502–506.

3. Jeppesen U, Gram LF, Vistisen K, et al. Dose-dependent inhibition of CYP1A2, CYP2C19and CYP2D6 by citalopram, fluoxetine, flu-voxamine and paroxetine. Eur J Clin Pharma-col. 1996;51:73–78.

4. Chiu CC, Lane HY, Huang MC, et al. Dose-dependent alternations in the pharmacokineticsof olanzapine during coadministration of flu-voxamine in patients with schizophrenia. J ClinPharmacol. 2004;44:1385–1390.

5. Silver H, Shmugliakov N. Augmentation withfluvoxamine but not maprotiline improves neg-ative symptoms in treated schizophrenia: evi-dence for a specific serotonergic effect from adouble-blind study. J Clin Psychopharmacol.1998;18:208–211.

6. Lu ML, Lane HY, Chen KP, et al. Fluvox-amine reduces the clozapine dosage needed inrefractory schizophrenic patients. J Clin Psy-chiatry. 2000;61:594–599.

7. Lu ML, Lane HY, Lin SK, et al. Adjunctivefluvoxamine inhibits clozapine-related weightgain and metabolic disturbances. J Clin Psy-chiatry. 2004;65:766–771.

8. Melkersson KI, Hulting AL, Brismar KE. Ele-vated levels of insulin, leptin, and blood lipidsin olanzapine-treated patients with schizophre-nia or related psychoses. J Clin Psychiatry.2000;61:742–749.

9. Melkersson KI, Dahl ML. Relationship be-tween levels of insulin or triglycerides andserum concentrations of the atypical antipsy-chotics clozapine and olanzapine in patients ontreatment with therapeutic doses. Psychophar-macology (Berlin). 2003;170:157–166.

Does N-DesmethylolanzapineIncrease, or Reduce,

the Risk forAntipsychotic-InducedMetabolic Syndrome?

To the Editors:

We appreciate the thoughtful com-ments by Drs Chiu and Lu on our studydealing with olanzapine and low-dosefluvoxamine combination treatment toreduce the therapeutic dose requirementsfor olanzapine.1 They note that this com-bination requires further monitoring ofits clinical benefits and risks, with afocus on metabolic syndrome. Althoughthis is consistent with our study conclu-sion that our results form a basis forlonger term evaluations of olanzapine

and adjunct fluvoxamine therapy,1 thecurrent treatment consensus statementfor atypical antipsychotics already indi-cates the need for routine clinicalmonitoring for metabolic side effectsbefore and during maintenance therapy.2

Drs Chiu and Lu suggest thatcoadministration of olanzapine and flu-voxamine may improve negative symp-toms of schizophrenia,3 while potentiallyraising the risk for antipsychotic-inducedmetabolic syndrome by increasing theolanzapine/N-desmethylolanzapine con-centration ratio. Specifically, they indicatethat a reduction in the N-desmethylolan-zapine metabolite concentration may in-crease the risk for metabolic side effects. Insupport of their hypothesis, they cite 2cross-sectional studies of patients treatedwith olanzapine.4,5 We note that the latterwork by Melkersson et al4 representscarefully conducted and detailed charac-terizations of metabolic endpoints inpatients with a psychotic illness. The studysamples were, however, relatively small(N = 14–16), and the cross-sectional de-sign does not permit an evaluation of thedegree of change in metabolic parameters.In one study, for example, only 3 patientshad elevated fasting blood glucose levels,and the other 11 patients were normogly-cemic. On the other hand, recent prospec-tive studies with a larger sample sizesupport the alternative hypothesis that anincrease, rather than a reduction, inolanzapine or N-desmethylolanzapineplasma concentration may lead to an

elevated risk for metabolic side effects.6,7

For example, Perry et al6 reported, in theJune 2005 issue of the Journal, a thresholdolanzapine plasma concentration of 20.6ng/mL being associated with an increasedlikelihood of clinically significant weightgain (7% baseline weight). This associ-ation remained significant after adjustingfor age, sex, baseline symptom severity,body mass index, and symptom improve-ment (odds ratio, 10.1; 95% confidenceinterval, 1.3–75.0; P = 0.024; N = 39). Inthe latter study, N-desmethylolanzapineplasma concentration was not measured.We emphasize, however, that the plasmaolanzapine and N-desmethylolanzapineconcentration exhibit a positive correlation(rs = 0.75, P = 0.001).5 Hence, it isreasonable to anticipate that, for example,the patients with a low olanzapine plasmaconcentration (eg, less than the 20.6 ng/mLconcentration threshold) in the previ-ous prospective study by Perry et al6 wouldalso tend to display a low N-desmethylo-lanzapine concentration. Notably, thesepatients with a low parent drug and(estimated) metabolite concentrationwere also found to have a lowerlikelihood of weight gain. In addition, aprospective study by Lu et al7 found asignificant positive correlation betweenN-desmethylclozapine plasma concen-tration and antipsychotic-induced in-crease in body weight, blood glucose,and triglyceride concentrations. Althoughit is plausible that there may be differ-ences in the biologic direction in which

FIGURE 1. Time course of fluvoxamine steady state plasma concentration duringco-administration with olanzapine in 10 patients with major psychosis. The data arepresented in individual patients (open circles) and as the mean value (bold line) ateach time point (solid triangles). In subject 10, fluvoxamine concentration data wasnot available at week 6. As a reference point, the broken line depicts the in vivoinhibition constant11 of fluvoxamine for CYP1A2. 1.0 ng/mL fluvoxamine = 3.16nmol/L.




the structurally related N-desmethylclo-zapine and N-desmethylolanzapine mayinfluence the risk for metabolic side ef-fects, the study by Lu et al7 would supportthe hypothesis that the risk for metabolicsyndrome may increase at higher con-centrations of the N-desmethylated me-tabolites of clozapine or olanzapine.

Dose-dependent inhibition of olan-zapine clearance by increasing doses offluvoxamine (50–100 mg/d) was reportedby Chiu et al.8 On the other hand, fluvox-amine displays nonlinear pharmacokinet-ics at doses 50 mg/d or higher.9,10 Figure 1shows that a 25-mg daily oral dose offluvoxamine administered in our studyproduces adequate and sustained concen-trations in the plasma that are at or higherthan the in vivo inhibition constant11 forKiiv for CYP1A2. We note, therefore, thatfluvoxamine doses higher than 50 mg/dshould be avoided to reduce uncertainty inthe time course and extent of CYP1A2inhibition during adjunctive treatmentto reduce olanzapine therapeutic doserequirements.

In summary, we would like toemphasize that the cost factor can limitadequate treatment and access ofpatients to atypical antipsychotics. Weagree that longer term and prospectivestudies are warranted to evaluate thesafety and efficacy of the olanzapineand adjunct fluvoxamine combinationtherapy. The available prospective dataappear to suggest that decreasing N-desmethylolanzapine concentration maybe associated with a reduced risk formetabolic side effects. However, itremains to be determined whether andto what extent olanzapine, N-desmethy-lolanzapine, or the N-desmethylolanza-pine/olanzapine ratio may serve as amarker of the antipsychotic-inducedmetabolic syndrome.

Lawrence J. Albers, MD*yVural Ozdemir, MD, PhD*y

Stephen R. Marder, MDyzMaria Augusta Raggi, PhDxManickam Aravagiri, PhDz

Laszlo Endrenyi, PhDkChristopher Reist, MD*y

*VA Long Beach Healthcare System and

Department of Psychiatry and Human

Behavior, College of Medicine, University

of California Irvine, Irvine, CA

yMental Illness Research, Education,

Clinical Center, VISN 22, Los Angeles, CA

zVA Greater Los Angeles Healthcare

System and Department of Psychiatry,

University of California, Los Angeles, CA

xDepartment of Pharmaceutical Sciences,

Faculty of Pharmacy, Alma Mater

Studiorum–University of Bologna, Via

Belmeloro 6, 40126 Bologna, Italy and

kDepartment of Pharmacology, Faculty of

Medicine, University of Toronto,

Ontario, Canada

[email protected]

REFERENCES

1. Albers LJ, Ozdemir V, Marder SR, et al. Low-dose fluvoxamine as an adjunct to reduceolanzapine therapeutic dose requirements: aprospective dose-adjusted drug interaction strat-egy. J Clin Psychopharmacol. 2005;25:170–174.

2. American Diabetes Association, AmericanPsychiatric Association, American Associa-tion of Clinical Endocrinologists, North Amer-ican Association for the Study of Obesity.Consensus development conference on anti-psychotic drugs and obesity and diabetes.Diabetes Care. 2004;27:596–601.

3. Silver H, Shmugliakov N. Augmentation withfluvoxamine but not maprotiline improvesnegative symptoms in treated schizophrenia:evidence for a specific serotonergic effect froma double-blind study. J Clin Psychopharma-col. 1998;18:208–211.

4. Melkersson KI, Hulting AL, Brismar KE.Elevated levels of insulin, leptin, and bloodlipids in olanzapine-treated patients withschizophrenia or related psychoses. J ClinPsychiatry. 2000;61:742–749.

5. Melkersson KI, Dahl ML. Relationship be-tween levels of insulin or triglycerides andserum concentrations of the atypical antipsy-chotics clozapine and olanzapine in patientson treatment with therapeutic doses. Psycho-pharmacology (Berlin). 2003;170:157–166.

6. Perry PJ, Argo TR, Carnahan RM, et al. Theassociation of weight gain and olanzapineplasma concentrations. J Clin Psychopharma-col. 2005;25:250–254.

7. Lu ML, Lane HY, Lin SK, et al. Adjunc-tive fluvoxamine inhibits clozapine-relatedweight gain and metabolic disturbances. JClin Psychiatry. 2004;65:766–771.

8. Chiu CC, Lane HY, Huang MC, et al. Dose-dependent alternations in the pharmacokinet-ics of olanzapine during coadministration offluvoxamine in patients with schizophrenia.J Clin Pharmacol. 2004;44:1385–1390.

9. Hiemke C, Peled A, Jabarin M, et al. Fluvox-amine augmentation of olanzapine in chronicschizophrenia: pharmacokinetic interactionsand clinical effects. J Clin Psychopharmacol.2002;22:502–506.

10. Spigset O, Granberg K, Hagg S, et al. Non-linear fluvoxamine disposition. Br J ClinPharmacol. 1998;45:257–263.

11. Yao C, Kunze KL, Kharasch ED, et al.Fluvoxamine-theophylline interaction: gap be-tween in vitro and in vivo inhibition constantstoward cytochrome P4501A2. Clin PharmacolTher. 2001;70:415–424.

Modafinil-InducedIrritability andAggression?

A Report of 2 BipolarPatients

To the Editors:

Modafinil is a novel drug withstimulant like properties but withouthaving the abuse potential of stimulants.The drug has been found to be effectivefor the treatment of narcolepsy.1 Re-search hints at the efficacy of modafinilin countering the fatigue and sleepinessassociated with depression.2 A recentstudy exploring the role of this agentin schizophrenia has found it to behelpful in countering the symptoms offatigue and sleepiness. Improvementsin cognitive functions and global func-tioning were also found.3 Case reportson the use of modafinil as a therapeuticagent for antipsychotic-associated se-dation have appeared in the literature.4

However, there has been concern aboutthe psychosis-inducing property of thisdrug.5 We hereby report 2 cases ofirritability and aggression related tomodafinil use in bipolar disorder.

Case 1Mr R.K., a 26-year-old single man

diagnosed with bipolar affective disorder,with a family history of suicide in a first-degree relative, presented to our tertiary carepsychiatric hospital with complaints offeeling low and decreased interest in work.On mental status examination, the patientrevealed ideas of helplessness and occa-sional worthlessness. At the time of presen-tation, the patient was taking clozapine 100mg/d and sertraline 50 mg/d. Given the riskof potential manic switch associated withsertraline, it was stopped, and the patientwas started on bupropion 150 mg/d. Withbupropion, he reported getting easily angryand aggressive, and hence, he stopped itwithin 3 days; after which, he felt alright. Hereported back after 2 weeks with complaintof excessive sedation during daytime, whichhad started since the initiation of clozapineand was impairing his work. No depressivesymptoms or cognitions were found on ex-amination. To counter the sedative effectof clozapine, he was prescribed modafinil100 mg/d with his consent. He reported after




another week with complaints of feelingirritable and having difficulty in controllingaggressive outbursts leading to verbal alter-cations in his office. He had no problems withhis sleep, appetite, libido, or personal care.Considering that the patient may be having animpending manic switch, modafinil wasstopped, and he was advised to continueclozapine in divided doses. The irritabilityand aggressive outbursts disappeared within 2days of discontinuing modafinil. The pa-tient has been doing well for 3 months afterstopping modafinil, although the problem ofsedation continues.

Case 2Mr V., diagnosed with bipolar af-

fective disorder, presented to us with com-plaints of daytime sedation with medications.The patient was on valproate 3500 mg/dwith injection zuclopenthixol decanoate 200mg every 10 days. Further interview revealedthat the sedation was worse since the additionof zuclopenthixol to valproate. Physicalexamination revealed no extrapyramidal orcerebellar sign. There were no symptomssuggestive of depression. The serum val-proate level was found to be 124 Ag/mL. Withhis consent, he was started on modafinil 100mg/d to counter the medication-inducedsedation. After 1 week of taking modafinil,the patient started feeling angry with minorevents. He felt irritable most of the time. Athis workplace, he felt as if he was going topick up fights with his colleagues every nowand then. Seeing his irritable mood, his wiferequested him to stop going to work and seekconsultation. With his psychiatrist’s advice,he stopped modafinil, and after the next 2days, he felt normal and remained euthymicafter that. He recalled that the irritability afterstarting modafinil was like his past experi-ence with the beginning of manic episodes.The problem of sedation continued, whichsubsequently led to his valproate dose beingreduced gradually to 2500 mg/d and injectionzuclopenthixol decanoate 200 mg beingspaced out to be administered every 15 days

over the next 2 months. The sedationhas come down since then without relapseof manic symptoms.

DISCUSSIONModafinil has been used success-

fully as an adjunctive agent in thetreatment of residual fatigue and sleep-iness in depressed patients.2 Reports ofmodafinil use to treat valproate-inducedsedation in bipolar disorder have ap-peared recently.6 The mechanism of itsaction is unknown, although it is thoughtto alter the balance of g-aminobutyricacid (GABA) and glutamate, resultingin activation of the hypothalamus. It isalso claimed to act on excitatory hista-minergic neurons and to increase thedopamine level in nucleus accumbensthrough the inhibition of GABA release.7,8

Modafinil has been reported to worsenpsychosis in schizophrenia patientsmaintained on clozapine.5 This may bea reflection of the indirect dopaminergicaction of the drug through inhibition ofGABA secretion. Considering the roleof dopamine in mood regulation,9 thisindirect dopaminergic action of modafi-nil is worth considering as a potentialmechanism underlying the irritabilityand aggression induced by this drug inthe 2 cases cited here. In addition, be-cause the mood stabilizers act throughGABAergic mechanism,10 the inhibi-tion of GABA secretion by modafinilmay well play a role in irritability andaggressive behavior observed here.

To our knowledge, this is the firstreport to illustrate the mood changescaused by modafinil in bipolar patients.Our observations hint at the need for ex-ercising caution while using modafinilin bipolar patients. Further studies maythrow some more light on this issue.

Sanjeev Ranjan, MD

Prabha S. Chandra, MD, MRCPsychDepartment of Psychiatry,

National Institute of Mental

Health and Neurosciences,

Bangalore, India

[email protected]

REFERENCES

1. Broughton RJ, Flemming JA, George CF.Randomized, double blind, placebo-controlledcrossover trial of modafinil in the treatment ofexcessive daytime sleepiness in narcolepsy.Neurology. 1997;49:444–451.

2. DeBattista C, Doghramji K, Menza MA, et al.Adjunct modafinil for the short-term treat-ment of fatigue and sleepiness in patients withmajor depressive disorder: a preliminarydouble blind placebo controlled study. J ClinPsychiatry. 2003;64:1057–1064.

3. Rosenthal MH, Bryant SL. Benefits of adjunctmodafinil in an open-label, pilot study inpatients with schizophrenia. Clin Neurophar-macol. 2004;27:38–43.

4. Makela EH, Miller K, Cutlip WD. Three casereports of modafinil use in treating sedationinduced by antipsychotic medication. J ClinPsychiatry. 2003;64:485–486.

5. Narendran R, Young CM, Valenti AM, et al.Is psychosis exacerbated by modafinil? ArchGen Psychiatry. 2002;59:291–292.

6. Berigan T. Modafinil treatment of excessivesedation associated with divalproex sodium.Can J Psychiatry. 2004;49:72–73.

7. Ferraro L, Tanganelli S, O’Connor WT, et al.The vigilance promoting drug modafinilincreases dopamine release in the rat nucleusaccumbens via the involvement of a localGABAergic mechanism. Eur J Pharmacol.1996;306:33–39.

8. Schatzberg AF, Cole JO, DeBattista C. Manualof Clinical Psychopharmacology. Washington,DC: American Psychiatric Press; 2003.

9. Willner P. Dopaminergic mechanisms indepression and mania. In: Bloom FE, KupferDJ, eds. Psychopharmacology: The FourthGeneration of Progress. 921. New York:Raven Press; 1995:32.

10. Stahl SM. Essential Psychopharmacology.Cambridge, Mass: Cambridge University Press;2000.




respiratory dyskinesia plasma risperidone levels and...

Documents