analysis of 67 cases

8/6/2019 Analysis of 67 Cases

1/8

Metoclopramide-Associated Tardive DyskinesiaAn Analysis of 67 Cases

Daniel D. Sewell, MD, Dilip V. Jeste, MD

Objective: To summarize information regarding the fre-quency, risk factors, clinical characteristics, treatment, andcourse of metoclopramide hydrochloride\p=m-\associated tar-dive dyskinesia obtained from an analysis of 67 case re-ports.

Data Sources: All the case reports of metoclopramide\x=req-\associated tardive dyskinesia involving human patients inthe literature in English obtained by using Index Medicusand Med-Search. The indexing terms used were as fol-lows: metoclopramide, tardive dyskinesia, dyskinesia, par-kinsonism, and extrapyramidal side effects.

Study Selection: For a patient to be included, the mainpublished research criteria had to be met based on theinformation provided. These criteria included exposureto metoclopramide for at least 30 days before the onset of

dyskinesia. Fifty-two patients met these criteria.

Data Extraction: One author independently extractedthe data.

Data Synthesis: The incidence and prevalence of tar-dive dyskinesia associated with metoclopramide have notbeen well studied. The mean (\m=+-\SD)length of treatmentwith metoclopramide before the onset of symptoms was20\m=+-\15months. The most common location of the dys-kinetic movements was the face (28 [60%] of 47) fol-lowed by the tongue (21 [45%] of 47). In 15 (71%) of 21patients on whom long-term follow-up was provided, thesymptoms were still present 6 months or more after dis-continuation of metoclopramide.

Conclusion: Persistent tardive dyskinesia is a serious po-tential side effect associated with metoclopramide treatment.

(Arch Farn Med. 1992;1:271-278)

Tardivedyskinesia (TD), an

involuntary movement dis-order that may appear af-

ter several months of treat-ment with antipsychotic

drugs, is characterized by a variable com-bination of orofacial dyskinesia, chorea, athe-

tosis, dystonia, tics, stereotypies, and fa-cial

grimacing,and may be transient or

persistent.1 The most widely described triadofsymptoms is sucking and smacking move-ments of the lips, puffing of the cheeks withtongue thrusting or rolling, and lateraljawmovements.2 Several factors have made TD

a serious problem. First, the reported prev-alence of TD has increased progressivelysince 1960 and is currently about 24%among patients treated with neurolepticdrugs.3'4 Second, in approximately 60% ofpatients with TD, the disorder persists for

longer than 3 months, even after theneu-

roleptic treatment has been discontin-

ued.5 Third, there are no well-established,effective treatments for TD,although somesuccess has been reported with tetrabena-zine, a dopamine-depleting drug availableonly with an investigative new drug ap-plication.7 Last, there is a growing numberof court cases against clinicians and hos-pitals filed by patients with TD.8

Although metoclopramide hydrochlo-ride has been in clinical use for over twodecades, its pharmacologie actions are com-plex and still not completely understood.It is known that among its other actions,metoclopramide is a dopamine receptorblocker and has some antipsychotic activ-ity. Like other drugs that are known to bedopamine receptor blockers, metoclopra-mide has been associated with extrapyra-midal side effects including TD. We re-viewed the literature in English and found21

articles containing67 cases

ofmeto-

clopramide-associateddyskinesia. We pro-

From the Department ofPsychiatry, University ofCalifornia, San Diego, and the

PsychiatryService, the San

Diego Veterans Affairs MedicalCenter.

on June 14, 2011www.archfammed.comDownloaded from
http://www.archfammed.com/http://www.archfammed.com/http://www.archfammed.com/http://www.archfammed.com/


2/8

vide herein an overview of what isknown about metoclopramide-associated TD, and summarize per-tinent information from an analysisof the 52 cases that met our inclu-sion criteria.

Metoclopramide was first intro-duced into clinical use in France9 andin 1967, it became available for clin-ical use in the United Kingdom10 asan antiemetic in a wide range of sit-uations. Metoclopramide did not be-come available in the United Slatesuntil 1979." According to Wyeth-Ayerst Laboratories (Philadelphia, Pa),

the companythat

upuntil

May 1990had exclusive rights to the manufac-ture and sale of the drug in the UnitedStates, the mean number of singleunits (capsules or tablets) sold to drugstores and hospitals in the UnitedStales per year from 1987 to 1989was 1 470 066 (personal communi-cation, Wyeth-Ayerst LaboratoriesMedical Services Division, 1990). In

May 1990, the drug became avail-able in generic form. It is projectedthat in 1990, 73.2% of the metoclo-pramide prescriptions written in theUnited Slates will be by physiciansin family practice (28%), internal med-icine (26.8%), or general practice phy-sicians (10.8%), or gastroenlerolo-gists (7.6%) (wrilten communication,IMS America, December 5, 1991).

Metoclopramide is generally clas-sified as a specific D2 dopamine re-ceptor blocker.12 There is evidence,

however, that it also actson

dopa-mine autoreceptors. '3 Metoclopramidecan be administered orally, intramus-cularly, and intravenously. In patientsother than those receiving chemother-apy for cancer, the usual dose for anyroute ofadministration is 1.0 mg threeor four times daily before meals or be-fore symptoms are likely to occur.

Metoclopramide treats vomit-ing associated with narcotic analge-sia, radiation therapy, and preg-nancy as well as or better than

phenothiazine antiemetics. It dimin-

ishes postoperative vomiting when itis given near the end of surgery, andis superior to placebo in a variety ofgastrointestinal disorders associatedwith delayed gastric emptying,11* suchas gastroesophageal reflux, gastric sta-sis following vagotomy, and dia-betic gastroparesis. A few studies sug-gest that a dose of 60 mg/d mayfacilitate healing of local inflamma-tion.15 Metoclopramide providessymptomatic improvement in dia-betic gastroparesis, although objec-tive improvement has been difficultto demonstrate.14 Metoclopramide isequal to or superior to anticholin-ergic drugs, phenothiazine antiemet-ics, and antacids in the treatment of

dyspepsia. No evidence exists to sup-

portthat

metoclopramide promoteshealing of peptic ulcer.14Initial studies using relatively

small doses of metoclopramide failedto demonstrate that it had signifi-cant tranquilizing or antipsychotic ac-tivity,"1'17 but now there is some ev-idence showing that it is an effectiveantipsychotic drug when adminis-tered in 500- to 1000-mg doses.18"20The effect of metoclopramide on invivo dopamine turnover suggests thatits potency as an antipsychotic drugon a milligram per kilogram basis maybe similar to that of chlorprom-azine.18 Stanley et al19,20 successfullytreated psychotic symptoms in eightpatients with chronic schizophreniausing metoclopramide.

The reported frequency of ex-trapyramidal side effects associatedwith metoclopramide varies from 1%21to 5%22 to 9%14 to 25%.l Acute dys-

tonic reactionsare

themost com-

mon type of extrapyramidal side ef-fect associated with metoclopramide.The approximate frequency of dys-tonic reactions is 0.2% when a doseof 30 to 40 mg/d is used. In patientsreceiving markedly higher doses, therate of dystonic reactions is consid-erably higher. For example, in youngpatients receiving chemotherapy forcancer, the frequency of dystonic re-actions is 25% or greater if dysloniaprophylaxis (eg, intravenous diphen-hydramine) is not administered.

Parkinsonian side effects also oc-cur but the frequency is unknown.23Bateman et al24 reported 20 cases ofmetoclopramide-induced parkin-sonism. In seven cases, the parkin-sonian side effects appeared within7 days of treatment, and in seven othercases, the symptoms appeared after

28 days. In two of these latter cases,the symptoms appeared after 5 yearsof treatment, which makes their as-sociation with metoclopramide treat-ment rather tenuous.

MATERIALS AND METHODS

We searched the literature in En-

glish for studies of the frequency ofmetoclopramide-associated TD and

found onlya

few such studies.We

found no systematic studies of TDin patients treated with metoclopra-mide. Because of this, we includedall the studies that provided infor-mation on frequency regardless of thequality of the methods used or of themanner of reporting results. We alsosearched the literature for case re-

ports of metoclopramide-associatedTD. We found 21 articles contain-

ing information on 67 patients with

metoclopramide-associated dyskine-sia. We also found an article byBoard" with summary data from 26cases of involuntary movement dis-orders "consonant with TD" that were

reported to the A. H. Robins Co (Rich-mond, Va) in the first 5 years duringwhich metoclopramide was avail-able in the United States. We did not

include these data in our analysis be-cause we could not determine which

of these caseswere

previously de-scribed in the published casereports.

To be included in our analysis,a patient had to meet the main TDdiagnosis criteria of Schooler andKane25 and Jeste and Wyatt5 (basedon the description provided), ex-cept for the criterion regarding theminimum length of exposure. We de-creased this requirement to 1 monthbased on the definition of TD pro-

posed in the DSM-1V Sourcebock? Aminimum duration of exposure of



3/8

1 month is necessary to avoid in-

cluding acute and subacute dyski-nesias that are clearly not "tardive"or late-onset.3 In addition, patientscould not have a history of anothermovement disorder.

Fifty-two cases met these crite-ria. Of the patients included in the

study, exposure to metoclopramidewas documented to be 3 months ormore in 47, between 1 and 3 monthsin two, and was undocumented be-fore the onset of TD in three cases.To decrease the likelihood of bias,we report findings from all 52 in-cluded cases (group 3) and findingsfrom only the 47 case reports withdocumented exposure to metoclo-

pramide of more than 3 months

(group1). For purposes of discus-

sion, patients with less than 3 months'exposure to metoclopramide and pa-tients with undocumented exposurebefore the onset of symptoms are re-ferred to as group 2 patients. Onlyone of 52 patients had a history ofprevious exposure to neurolepticdrugs. This was a patient in group 2who had received 10 mg of triflu-

operazine hydrochloride twice a dayfor an unspecified length of time fortreatment of an episode of depres-sion that occurred 2'/2 years beforetreatment with metoclopramide.26

Fifteen of the original 67 wereexcluded. We excluded four cases be-

cause the symptoms ofdyskinesia ap-peared within the first 30 days of treat-ment with metoclopramide.27"30 Tencase reports that did not contain enoughinformation to determine whether the

cases met inclusionand exclusion cri-

teria were excluded.2431 We

attemptedto contact the authors of the reportsto obtain the missing information butour efforts did not yield any additionalinformation. We excluded one case

because written communication with

the author revealed that the dyskine-sia resolved "within 1 hour of its

onset."22Ten of the articles contained case

reports that clearly met criteria forinclusion in the analysis but, none-

theless, contained cases with miss-ing data. We attempted to contact the

authors of these articles to obtain the

missing data. Grimes et al32 re-ported 12 cases with only mean val-ues for the entire group. For the pur-poses of our analysis, we used themean values as if they were the spe-cific values for each of these cases. Twocases reported by Miller and Jankov-

ic33 didnot

include information aboutthe length of exposure; however, weassumed that the researchers who madethese diagnoses were aware of the cri-teria regarding minimum requiredlength of exposure and we includedthese two cases. It should be noted

that MillerandJankovic33 reportedmax-imum daily metoclopramide dose. Theauthors of all the other case reportsincluded in our analysis reported dailydoses, mean daily doses, or dose ranges.When dosage ranges were given, weused the midpoint value of the rangeto calculate the mean daily dose forour sample.

Our results are summarized in theTable3436

FREQUENCYTo our knowledge, the incidence andprevalence ofTD associated with me-toclopramide have not been studied.Indo and Ando28 found 10 cases of

metoclopramide-induced parkinsonismamong 282 cases of parkinsonismtreated during a 10-year period from1970 through 1979 at a hospital in

Gifujapan, and noted that four of these

patients had dyskinesia. They did not

provide information on the size of thepopulation from which these cases wereidentified. Wiholm et al37 found 11

cases of TD associated with metoclo-

pramide that had been reported to theSwedish Adverse Drug Reactions Ad-

visory Committee from 1977 to 1981.

Using total drug sales and prescrip-tion statistics, these authors calculated

the incidence of TD during treatmentwith metoclopramide to be one in 2000to

2800treatment

years.Bateman et

al24 found four cases recorded in an

adverse reactions register in the UnitedKingdom during a 15-yearperiod whenan estimated 15.9 million prescriptionshad been written. Miller and Jankov-ic33 assessed lmetoclopramide-treatedpatients with dyskinesia who had beenselected from a database of 3000 pa-tients with various extrapyramidal dis-

ordersseen

from 1977 throughJune1989. The most common movementdisorder in this group wasTD, observedin 10 patients. The frequency ofmetoclopramide-associated TD has notbeen well studied. Ganzini et al38 re-

cently reported a study of 51 patientstreatedwithmetoclopramidefor a meanof 2.5 years. Twenty-four of these pa-tients had diabetes. In the diabetic

group, 42% met research criteria for

TD. In the group without diabetes, 19%met criteria for TD. We did not findany study of the prevalence of TD inlarge cohorts ofmetoclopramide-treatedpatients.

RISK FACTORS

Wiholm et al37 found that the incidenceofTD with long-term metoclopramideuse in patients aged 70 years or olderwas greater than one in 1000. This in-

cidencewas two to

threetimes

higherthan that in a group representing allthe patients treated with metoclopra-mide. These authors concluded that

long-term metoclopramide treatmentwas associated with a significant riskof developing TD, especially amongthe elderly.

The mean (SD) age of the pa-tients in group 3 was 7010 years.The mean age of patients in groups1 and 2 were 7010 and 71 13,

respectively. Group 1 was com-posed of 24 women, 10 men, and13 patients whose gender was notspecified. Group 2 was composed offive women. In group 3, the female-male ratio was approximately 3:1. Six-teen of the 47 patients in group 1and two of the five patients in group2 had parkinsonian symptoms alongwith TD. Four patients in group 1and one patient in group 2 were de-scribed as having diabetes. Al-though depression has been pro-



4/8

Case Reports of Metoclopramide-Associated Tardive Dyskinesia (TD)

Source, yNo. ofCases

Mean Age(Range), y

Sex, miUncertain

Mean Dose(Range), mg/d

Mean Length(Range) ofUse Prior

toTDTD

Localization'1

TD Status AfterMetoclopramide

Withdrawal at Last

Follow-up, mofKataria et al,431978

Lavy et a!,491978

Grimes et al,3219824:Indo and Ando,281982Shearer et al,"01984Wholmetal,3'1984

Beauclair and Fontaine,261986Breitbart,41 1986Lazzara et al,601986Patel and Louis,341986Sirota et al,451986Samie et al,421987Yamamoto et al,351987Miller and Jankovic,331989||Lang,441990Sewell et al36

12

3

75 (66-84)

48

72

67(64-71)5076 (69-86)

74

45

83

84

65

66

81

68 (53-85)74

46

0/3

1/0

?

1/1/1

1/0

0/11

0/1

1/0

0/1

0/1

0/1

1/0

1/0

2/90/1

2/0

28 (15-40)

20-40 (6 y)

80 (4 y)2930

3031 (10-60)

30-60

2 mg/kg30

?

30

30-40

30

33 (20-40)30

40 (30-60)

34 (12-66)

72

26

15 (6-27)4

17 (4-44)

?

11

88

2

12

9

18 (2.5-48)4

11.5(11-12)

J, Tr (n=2);J, Lp (n=1)

F, J; Lp, To

F, Lp, ToLp; Lp, To; JTr

F, J, To (n=2);F, J, To, Tr (n=9)

E, ToF, J, Lp, Lm, TrF, J, To

Lp, ToF

F, Lm, Lp, To, TrLp, To

F, Lp (n=9); ? (n=2)F, To

J, Lp; To

R 0.75, P 9, P 6

R 0.75 (3), P 18 (8)R 0.75, P 0.75, R 3R27

P 17, P, 2, R ?

P6

P3

P1

P6

R0.25

P3P0.32

?

P27

P 18, P3

* J indicates jaw; Tr, trunk; Lp, lips; F, face; To, longue; and Lm, limbs.fP indicates persistence of symptoms, and R, complete resolution of symptoms at time of latest reported follow-up.Authors reported mean values from 12 cases of metoclopramide-associated TD. Numbers in parentheses indicate sample sizes studied for TD status at

follow-up.Authors did not specify exact number of months but used the word "several.

"

\\Authors reported maximum (not mean) metoclopramide doses.

posed as a risk factor for antipsychoticdrug-induced TD,39 only two case re-ports of metoclopramide-associatedTD (one each in groups 1 and 2) spec-ified the presence or history of de-pression.2640 Metoclopramide was pre-scribed for a number of differentindications and five patients had twoindications for metoclopramide. Theindications for metoclopramide treat-ment were nausea and vomiting(n=10); esophageal reflux/esophagitis (n=5); hiatal hernia (n=4);

diabetic gastroparesis (n=3); epigas-tric discomfort (n=2); ulcer (n=2); an-orexia (n=2); chronic gastritis (n=2);diverliculitis (n=l); ileus (n=l); par-tial gastrectomy (n=l); and unspec-ified (n-24).

CLINICAL CHARACTERISTICS

Thirty-one of the 52 case reports an-alyzed described the temporal rela-

tionshipbetween the treatment

withmetoclopramide and the onset of TDsymptoms. In 25 cases (81%), the TD

Symptoms developed during treat-ment with metoclopramide and in six

(19%), the symptoms developed af-ter treatment was discontinued. In

group 1, 23 patients (79%) devel-oped symtoms during treatment andsix (21%) after treatment was dis-continued. This information was re-

ported for only two of the five pa-tients in group 2, both of whom

developed TD symptoms during treat-ment. The mean length of treatmentwith metoclopramide before the on-

set of TD symptoms for the 49 casesfor which this information was pro-vided was 20 15 months and themean daily dose was 32 7 mg. Inpatients in group 1, mean treatmentduration was 2015 months and

mean daily dose was 33 10 mg. In20 (38%; 18 from group 1 and twofrom group 2) of 52 cases, parkin-sonian symptoms were reported tocoexist with TD.

In47 (44 from group

1and threefrom group 2) of the 52 case re-

ports, the topographic location of the

dyskinetic movements was de-scribed. All but five cases had dys-kinetic movements affecting more thanone region. The most common lo-cation of the dyskinetic movementswas the face (n=28), followed by thetongue (n=21), the lips (n=19), the

jaw (n=19), the trunk (n=9), and thelimbs (n=3). Three patients had move-ments affecting the eyelids.37,41,42 Threehad movements affecting the pelvicregion,43,44 of whom one had con-tractions of the anal and vaginal mus-

culature.43 Two patients had dia-phragmatic dyskinesia42'44 and one ofthese resulted in severe breathingand swallowing difficulties.42 TheFigure compares the topographic lo-calization of metoclopramide-associated TD with that of TD in-duced by traditional antipsychoticagents.

TREATMENT AND COURSE

In all 52 cases, treatment was ulti-

mately discontinued once the pos-



5/8

sible causative relationship betweenmetoclopramide and TD was recog-nized. Thirty-two case reports (30 ingroup 1 and two in group 2) in-cluded information regarding thecourse of the TD symptoms after with-drawal of the drug. Twenty-one ofthese provided information 6 monthsor longer after onset of TD and 11either did not specify the length ofthe follow-up or had a follow-up ofless than 6 months. The mean fol-

low-up provided for all cases for whichthis information was available was166.5 months, with a range of 6to 27 months. It is noteworthy that15 (71%) of the 21 case reports withfollow-up information at 6 monthsor more had persistent TD.

In fivegroup

1 cases and in one

group 2 case, the TD symptoms werenoted to have resolved completely atsome point during the first 6 monthsafter discontinuation of metoclopra-mide. The mean duration until theresolution of the symptoms for all case

reports that included this informa-tion was 2.02.2 months and the

range was 1 week to 6 months.In the remaining 11 case re-

ports that included follow-up infor-mation, 10 patients were describedas havingpersistent TD symptoms andin one patient the symptoms werereported to resolve. The duration offollow-up was not specified in twoof the 10 patients with persistent TDand in the one in whom the symp-toms did not resolve. In the remain-

ing eight patients, the follow-up pe-riod reported was less than 6 months.The mean follow-up for these eightwas 1.4 1.4 months and the rangewas 6 days to 4 months.

One patient in group 1 was de-scribed as achieving significant im-provement with treatment with cho-

line; however, the dose and durationof treatment were not specified.43 An-other patient28 in group 1 was treatedwith levodopa and benserazide;1 week later, the tremor diminishedbut the authors did not comment on

whether these medicationsde-

creased the symptoms of dyskine-sia. Grimes et al32 reported that in

100-

90-

80-

I 60" ^ 50" ^ BL B^i

s *_

m b^ m .^

Face Lips Jaw Tongue Limbs Trunk

Topographic distribution of tardive dyskinesia (TD) in patients with metoclopramide-associated TD(hatched bars) vs patients with TD associated with traditional antipsychotic medications (solid bars).Percentages for patients with TD associated with traditional antipsychotic use were from an analysisby Jeste and Wyatt,5 who calculated and reported mean values using percentages obtained from fivestudies of the localization of dyskinetic movements. Although the topographic distribution of TDassociated with metoclopramide resembles that associated with traditional antipsychotic drugs, oneapparent difference is the lower frequency of limb dyskinesia in cases of metoclopramide-associated TD.

three of 12 patients (all group 1) with

TD,the

dyskineticmovements re-

solved within 3 weeks after the ter-mination of metoclopramide treat-ment. One patient died, and in theremaining eight, the movements per-sisted between 6 and 36 months(mean, 18 months). In addition, theauthors reported that two patients hadmild to moderate improvement in TDwith treatment with 30 g of lecithin

granules per day, and in one pa-tient, treatment with haloperidol de-creased the TD but worsened par-kinsonian symptoms.32 Anotherpatient41 (group 1) initially treatedwith anticholinergic drugs that madethe movements more pronounced,was then treated with diazepam,which led to moderate improve-ment but the symptoms persisted un-til the patient died of cancer 3 monthslater. Another patient (group 2) ex-perienced great improvement in fa-

cial dyskinesia 1 day after receivingtreatment with diphenhydramine hy-drochloride.45 One patient42 (group1) receiving "carbidopa/levodopa" (25mg/100 mg) at the time the TD symp-toms developed, continued receiv-

ing this medication unchanged withthe addition of 2 mg of trihex-

yphenidyl hydrochloride twice a day.One month later, the patient's TD

symptoms were much worse.Both

medications were discontinued and

treatment with 5 mg of diazepam fourtimes daily, 0.1 mg of reserpine four

times daily, and 0.5 mg of clon-

azepam every4 hours was begun. The

patient improved several months later,but grimacing and blepharospasmpersisted.

COMMENT

One of the limitations of a reviewbased on case report analysis is thevariable omission of pertinent infor-mation. Other limitations of this re-

view are the modest sample size of52 cases and the fact that few of the

case reports included the systematicapplication of published criteria todiagnose TD.

On the other hand, to our

knowledge this is the largest reviewofmetoclopramide-associated TD andthere have been no published pro-spective studies. In addition, we sys-tematically tried to apply published

criteria for the diagnosisof TD and

excluded cases that did not meet those

criteria. Our survey does not provethat a causal relationship exists be-tween treatment with metoclopra-mide and the development of TD. Toobtain direct evidence, a prospec-tive study needs to be conducted com-

paring the frequency of dyskinesiain a group of patients whose only ex-

posure to neuroleptic drugs has beentreatment with metoclopramide, with

a group of well-matched control pa-tients never exposed to metoclopra-



6/8

mide or any other neuroleptic med-ication. Such a study will need totake into account the rate of spon-taneous dyskinesia46 and the im-pact of other possible causes ofdyskinesia.

I(71%) of the 21 case reportswith follow-up information

at 6 months or more had |persistent TD.

The results of our literature re-view are of interest for several rea-

sons. Although the findings here donot provide direct evidence that me-toclopramide causes TD they do pro-vide highly suggestive indirect evi-dence. This evidence takes threeforms; (1) the number of publishedcase reports; (2) the known phar-macologie similarities between me-toclopramide and the traditional an-tipsychotic medications; and (3) thesimilarities we found between the syn-drome of TD associated with meto-

clopramide and that caused by thetraditional antipsychotic agents.

We found 52 case reports in theliterature in English of metoclopra-mide-associated TD that met our se-lection criteria. This may be too largea number to be simply a coinci-dence. There have been more case

reports of TD associated with meto-

clopramide treatment than with anyother class of medication except thetraditional antipsychotic drugs. Fur-

thermore, in 47% of the cases thatwe analyzed, TD symptoms were re-ported to persist for at least 6 months.To our knowledge, except for neu-roleptic drugs, no other drugs pro-duce persistent dyskinesia.5

Metoclopramide shares a num-ber of characteristics with the com-

monly used antipsychotic drugs: (1)it blocks dopamine receptors12; (2)it has tranquilizing and antipsy-chotic

effects18"20;and

(3)the rate ex-

trapyramidal side effects associatedwith metoclopramide, although not

as high as with traditional antipsy-chotic drugs, is much higher than withother psychotropic medications.

The syndrome of TD in patientstreated with metoclopramide resem-bles that seen in patients treated with

traditional antipsy-chotic medications.The mean

lengthof

metoclopramide treat-ment prior to the on-set of dyskinesia was20 months, clearly in-dicating the "tardive"nature of the move-ment disorder. The ab-

normal movements observed in the52 cases ofmetoclopramide-associatedTD are phenomenologically similar tothe antipsychotic drug-induced TD.47In 15 (47%) of the 32 cases that pro-vided long-term follow-up informa-tion, TD was still present after 6 months.This is similar to the proportion ofpa-tients in whom symptoms of TD per-sist after exposure to traditional an-

tipsychotic drugs.48 In all cases, theTD symptoms developed either dur-ing treatment or within a few weeksof discontinuation. In addition, manyof the patients who developed TD

while taking metoclopramide expe-rienced temporary worsening imme-diately after discontinuation of thetreatment,44,49,50 a phenomenon of-ten seen after traditional neuroleptictreatment withdrawal. Last, anticho-

linergic medications worsened symp-toms41 but resumption of treatmentwith metoclopramide37,49 or initia-tion of treatment with a traditional

antipsychotic37,51 improved symp-toms. Thus, there are striking simi-larities between TD associated withmetoclopramide and that inducedby antipsychotic drugs. Nonethe-less, the comparison presented in the ,Figure suggests that metoclopramide-associated TD may have at least onedifference from antipsychotic drug-induced TD. Specifically, TD asso-ciated with metoclopramide maymore frequently involve the face and

lipsand less

frequentlythe limbs than

that associated with antipsychoticdrugs.

There are a number of reasons

why studying metoclopramide-associated TD is important: (1) ourreview suggests that the syndromemay be persistent; (2) it can lead toserious complications such as respi-ratory compromise42; (3) it appearsto be relatively resistant to treat-ment

justlike the TD associated with

traditional antipsychotic treatment;(4) metoclopramide is a rather com-monly used medication and there-fore a significant number of patientsmay be at risk; and (5) the fre-

quency of TD associated with meto-

clopramide is probably underesti-mated. Although few of the authorsof case reports of metoclopramide-associated TD provided quantitativedata regarding TD severity, the de-scriptions provided and the distressand complications reported suggestthat the overall severity of the casesin our analysis was high. It is likelythat reports of only the more severecases of TD were published. It isknown that a majority of the casesof antipsychotic drug-induced TD aremild. If the literature on TD associ-

ated with traditional antipsychoticdrugs is any guide, there may be a

large number of cases of mild TD as-sociated with metoclopramide.

It should also be noted that re-

ports on frequency that obtained datafrom adverse drug reaction registersare limited by the rate of reportingof adverse drug reactions. What per-centage of adverse drug reactions areactually reported is not known, butcertainly it is reasonable to assumethat this percentage is not high.

Determining the frequency ofmetoclopramide-associated TD re-quires accurate recognition of TD. Ina letter to the editor in 1983, Ca-

sey52 alerted physicians to the drug'spotential for neurologic side effectsincluding TD. Until recently, how-ever, TD has been a medication sideeffect almost exclusively diagnosedin psychiatric settings. Just as psy-chiatrists were initially slow to rec-

ognizeand

diagnoseTD

duringthe

period following the first descrip-tion of TD,48 clinicians outside of the



7/8

psychiatric field currently may besomewhat less likely to recognize TD,especially when it is mild. The factthat patients with TD frequently arecompletely unaware of it until it hasbeen brought to their attention bysomeone else probably also de-creases the likelihood that TD will

be diagnosed promptly.The published literature sug-

gests that TD is more frequent in olderpatients and in women.53 Severalgroups of investigators have re-ported a frequent coexistence of TDand parkinsonism54,55 with some in-dicating that neuroleptic drug-induced parkinsonism may be a riskfactor for TD.56,57 In addition, a re-

cently published controlled study byGanzini el al58 suggests that diabetesmellitus may be a risk factor for TD.Our results suggest that these risk fac-tors could also characterize metoclo-

pramide-associated TD.It is not known how the risk oi

metoclopramide-associated TD com-pares with the risk of TD from other

antipsychotic drugs. Except for pa-tients undergoing chemotherapy, mostpatients receiving metoclopramide

generally receive doses much lowerthan those required to treat psycho-sis, which suggests that the risk forthese patients may be lower. On theother hand, metoclopramide is of-ten used for long periods in patientswho might have a higher frequencyof putative TD risk factors such asadvanced age and diabetes. In addi-tion, it is unclear why there are sofew cases in the literature of TD as-sociated with

prochlorperazinemale-

ate than with metoclopramide when,except for patients receiving chemo-therapy for cancer, both are often pre-scribed at doses much lower thanthose required to treat psychosis. Itis possible that metoclopramide ismore likely to produce TD than otherdrugs that block dopamine. Until re-cently, TD was not associated withmedications used outside of the prac-lice of psychiatry. The case reportssummarized above suggest that TDis no longer an iatrogenic disorderunique to psychiatry.

RECOMMENDATIONS

Our review suggests a number ofclin-ical recommendations: (1) metoclo-pramide should be used with cautionin patients who are at high risk of de-veloping TD, eg, elderly or diabetic

patients; (2) patientswho

requireme-

toclopramide should be monitored atregular intervals for signs of TD;(3) if signs of TD appear, even if theyare very mild, metoclopramide shouldbe discontinued, if feasible.

Accepted for publication June 10,1992.From the Department of Psychi-

atry, University of California, San Di-ego, and the Psychiatry Service, the San

DiegoVeterans

AffairsMedical Cen-

ter.

This work was supported, in part,by grant 1R01MH45131 from the Na-tional institutes of Menial Health andby the Department of Veterans Affairs.

We thank IMS America, Ply-mouth Meeting, Pa.

Reprint requests to Psychiatty Ser-vice V-J16A, San Diego Veterans Af-fairs Medical Center, 3350 La folia Vil-lage Dr, San Diego, CA 92161 (DrSewell).

REFERENCES J1. Baldessarini RJ, Cole JO, Davis JM, et al. Tar-

dive Dyskinesia:A Task Force Reportof the Amer-ican Psychiatric Association. Washington, DC:American Psychiatric Association; 1980:23.

2. Perry PJ, Alexander B, Liskow Bl. PsychotropicDrug Handbook. 5th ed. Cincinnati, Ohio: Har-

vey Whitney Books; 1988:28.3. Jeste DV, Yassa R. Neuroleptic-induced tardive

dyskinesia.In: Francis

A,ed. DSM-IV Source-

book. Washington, DC: American Psychiatric As-sociation. In press.

4. Kane JM, Smith JM. Tardive dyskinesia: preva-lence and risk factors, 1959 to 1979. Arch Gen

Psychiatry. 1982;39:473-481.5. Jeste DV, Wyatt RJ. Understanding and Treating

Tardive Dyskinesia. New York, NY: Guilford Press;1982.

6. Jeste DV, Lohr JB, Clark K, Wyatt RJ. Pharma-

cologic treatment of tardive dyskinesia in the 1980s.J Clin Psychopharmacol. 1988;8(suppl):38-48.

7. Jankovic J, Orman J. Tetrabenazine therapy of

dystonia, chorea, tics, and other dyskinesias. Neu-

rology. 1988;38:391-394.8. Tancredi LR. Malpractice and tardive dyskine-

sia: a conceptual dilemma. J Clin Psychophar-macol. 1988;8(suppl):71-76.

9. Casteels-Van Daele M, Jaelen J, Van Der Schueren

P, Zimmerman A, Van Den Bon P. Dystonic re-actions in children caused by metoclopramide.Arch Dis Child. 1970;45:130-133.

10. Robinson OPW. Metoclopramide: side effects andsafety. Postgrad Med J. 1973;49(suppl):77-80.

11. Board AW. Tardive dyskinesia associated withhigh-dose intravenous metoclopramide. N EnglJ Med. 1986;315:518-519.

12. Kebabian JW, Calne DB. Multiple receptors for

dopamine. Nature. 1979;277-93-96.13. Alander T, Anden N, Grabowska-Anden M. Me-

toclopramide and sulpiride as selecting blockingagents of pre- and postsynaptic dopamine re-

ceptors. Naunyn Schmiedebergs Arch Phama-col. 1980;312:145-150.

14. Harrington RA, Hamilton CW, Brogden RN,Linkewich RA, Romankiewicz, Heel RC. Meto-

clopramide: an updated review of its pharma-cologic properties and clinical use. Drug Eval.1983;25:451-494.

15. Schutz E. Fortechritte in der Therapie des Ulcus

duodeni. Munch Med Wochenschr. 1976:118:

231-234.

16. Borenstein PP, Bles G. Effets cliniques et elec-troencephalographiques du metoclopramide enpsychiatrie. Therapie. 1965;20:975-995.

17. Nakra BRS, Bond AJ, Lader MH. Comparative

psychotropic effects of metoclopramide andprochlorpromazine in normal subjects. J Clin Phar-macol. 1975;15:449-454.

18. Stanley M, Wilk S. Striatal DOPAC elevation pre-dicts antipsychotic efficacy of metoclopramide.Life Sci. 1979;24:1907-1912.

19. Stanley M, Lautin A, Rotrosen J, Gershon S. An-

tipsychotic efficacy of metoclopramide: do DA/neuroleptic receptors mediate the action of anti-

psychotic drugs? IRCS MedSci Clin MedClin Phar-

macolNervousSys PsycholPsychiatry. 1979;7:332.20. Stanley M, Lautin A, Rotrosen J, Gershon S, Klein-

berg D. Metoclopramide: antipsychotic efficacyof a drug lacking potency in receptor models.

Psychopharmacology. 1980;71:219-225.21. Borda IT. Drug treatment of gastrointestinal dis-

orders. In: Swift CG, ed. Clinical Pharmacologyin the Elderly. New York, NY: Marcel Dekker Inc;1987:529-579.

22. Craig JB, Powell BL, White DR, Capizzi RL. An-tiemetic therapy in patients treated with high\x=req-\dose cytosine. Oncology. 1987;44:90-92.

23. Physicians' Desk Reference, 45th Edition. Orad-ell, NJ: Medical Economics Books; 1991.

24. Bateman DN, Rawlins MD, Simpson JM. Ex-

trapyramidal reactions with metoclopramide.BMJ. 1985;291:930-932.

25. Schooler NR, Kane JM. Research diagnoses for

tardive dyskinesia. Arch Gen Psychiatry. 1982;39:488-489.

26. Beauclair L, Fontaine R. Tardive dyskinesia as-

sociated with metoclopramide. Can Med Assoc

J. 1986;134:613-614.27. Melmed S, Bank H. Metoclopramide and facial

dyskinesia. BMJ. 1975;8:331.28. Indo T, Ando K. Metoclopramide-induced par-

kinsonism: clinical characteristics of ten cases.

Arch Neurol. 1982;39:494-496.29. Leopold NA. Prolonged metoclopramide\x=req-\

induced dyskinetic reaction. Neurology. 1984;34:238-239.30. Jankovic J, Glass JP. Metoclopramide-induced



8/8

phantom dyskinesia. Neurology. 1985;35:432-435.31. Miller LG, Jankovic J. Neurologic approach to

drug-induced movement disorders: a study of125 patients. South Med J. 1990;83:525-532.

32. Grimes JD, Hassan MN, Krelina M. Long-termfollow-up of tardive dyskinesia due to metoclo-pramide. Lancet. 1982;2:563.

33. Miller LG, Jankovic J. Metoclopramide-inducedmovement disorders: clinical findings with a re-view of the literature. Arch Intern Med. 1989;149:2486-2492.

34. Patel M, Louis S. Long-term neurologic compli-cations of metoclopramide. N Y StateJMed.1986;86:210.

35. Yamamoto M, Ujike H, Ogawa N. Metoclopramide\x=req-\induced parkinsonism. Clin Neuropharmacol. 1987;10:287-289.

36. Sewell DD, Yoshinobu BH, Caligiuri MP, JesteDV. Metoclopramide-associated tardive dyski-nesia in hemodialysis patients with diabetes mel-litus: two case reports. Gen Hosp Psychiatry. Inpress.

37. Wiholm BE, Mortimer 0, Boethius G, HaggstromJE. Tardive dyskinesia associated with metoclo-pramide. BMJ. 1984;288:545-547.

38. Ganzini L, Casey DE, Hoffman WF, Heintz RT.Tardive dyskinesia in neuroleptic-treated diabet-ics. Presented at the 30th Annual Meeting of theAmerican College of Neuropsychopharmacol-ogy; December 10,1991; San Juan, Puerto Rico.

39. Casey DE. Tardive dyskinesia and affective dis-orders. In: Gardos G, Casey DE, eds. Tardive Dys-kinesia andAffective Disorders. Washington, DC:

American Psychiatric Association; 1984:1-20.40. Shearer RM, Bownes IT, Curran P. Tardive akathi-

sia and agitated depression during metoclopra-mide therapy. Ada Psychiatr Scand. 1984;70:428-431.

41. Breitbart W. Tardive dyskinesia associated withhigh-dose intravenous metoclopramide. N EnglJ Med. 1986;315:518.

42. Samie MR, Dannenhoffer MA, Rozek S. Life\x=req-\threatening tardive dyskinesia caused by meto-

clopramide.Mov Disord.

1987;2:125-129.43. Kataria M, Traub M, Marsden CD. Extrapyrami-dal side-effects of metoclopramide. Lancet. 1978;2:1254-1255.

44. Lang AE. Clinical differences between metoclo-pramide- and antipsychotic-induced tardive dys-kinesia. Can J Neurol Sci. 1990;17:137-139.

45. Sirota RA, Kimmel PL, Trichtinger MA, DiamondBF, Stein HD, Yudis M. Metoclopramide\x=req-\induced parkinsonism in hemodialysis patients:report of two cases. Arch Intern Med. 1986;146:2070-2071.

46. Casey DE, Hansen TE. Spontaneous dyskine-sias. In: Jeste DV, Wyatt RJ, eds. Neuropsychi-atric Movement Disorders. Washington, DC: Amer-

ican Psychiatric Association; 1984:67-95.47. Casey DE. Tardive dyskinesia. In: Meltzer HY, ed.

Psychopharmacology: The Third Generation ofProgress. New York, NY: Raven Press; 1987:1411-1419.

48. Jeste DV, Wyatt RJ. Changing epidemiology oftardive dyskinesia: an overview. Am J Psychia-try. 1981;138:297-309.

49. Lavy S, Melamed E, Penchas S. Tardive dyski-nesia associated with metoclopramide. BMJ. 1978;1:77-78.

50. Lazzara RR, Stoudemire A, Manning D, PrewittKC. Metoclopramide-induced tardive dyskine-sia: a case report. Gen Hosp Psychiatry. 1986;8:107-109.

51. Grimes JD, Hassan MN, Preston DN. Adverseneurologic effects of metoclopramide. Can MedAssoc J. 1982;126:23-25.

52.

CaseyDE.

Metoclopramideside effects. Ann In-

tern Med. 1983;98:673-674.53. Kane JM, Woerner M, Lieberman J, Kinon B. Tar-

dive dyskinesia. In: Jeste DV, Wyatt RJ, eds. Neu-ropsychiatric Movement Disorders. Washing-ton, DC: American Psychiatric Association; 1984:97-118.

54. Hoffman WF, Labe SM, Casey DE. Neuroleptic-induced parkinsonism in older schizophrenics.Biol Psychiatry. 1987;22:427-439.

55. Gardos G, Perenyi A, Cole JO, Samu I, Kocsis E,Casey DE. Seven-year follow-up of tardive dys-kinesia in Hungarian outpatients. Neuropsycho-pharmacology. 1988;1:169-172.

56. Crane GE. Pseudoparkinsonism and tardive dys-kinesia. Arch Neurol. 1972;27:426-430.

57. Kane JM, Woerner M, Lieberman J. Tardive dys-kinesia: prevalence, incidence, and risk factors.J Clin Psychopharm. 1988;8(suppl):52-56.

58. Ganzini L, Heintz RT, Hoffman WF, Casey DE.The prevalence of tardive dyskinesia in neuroleptic-treated diabetics: a controlled study. Arch Gen

Psychiatry. 1991;48:259-263.

Practice Commentary

Metoclopramide,an effective drug in the treatment of emesis and gastrointestinal disturbances, is frequently pre-

scribed by generalists. Seventy-one percent of all prescriptions are written by family physicians, internists, andgeneral practitioners, and it is commonly used in our practice. Sewell and Jeste remind us of potentially serious side

effects of the drug. It is important to note that the risk for TD increases with the duration of therapy and with age and thatthe incidence is higher in diabetics. It is discouraging that many cases were permanent. As practitioners, we can minimize therisk by carefully selecting patients, especially among the elderly, and limiting the duration of therapy. We should all be awareof the mild forms of TD, including those visible to the examiner but unrecognized by the patient, and discontinue tine treatmentif feasible when the symptoms appear.

This study again illustrates how poorly we understand the risks and benefits of many of our interventions. In spite of anexhaustive literature search and conscientious efforts on the part of the authors to obtain missing data, the incidence, prev-alence, and duration of metoclopramide-induced TD is still poorly understood because of the limitations of the studies re-viewed and the small sample size.

Thomas L. Speros, MDWashington, NC

From the Washington (NC) Family Medicine Center.


analysis of 67 cases

Documents