A Pilot Study of Clomipramine In Young Autistic Children

LAURA E. SANCHEZ, M.D., MAGDA CAMPBELL, M.D., ARTHUR M. SMALL, M.D.,

JEANETTE E. CUEVA, M.D., JORGE L. ARMENTEROS, M.D., AND PHILLIP B. ADAMS, PH.D.

ABSTRACT

Objective: To assess the short-term efficacy and safety of clomipramine in hospitalized young children with autism.

Method: This was an open pilot study; after a 1-week placebo baseline, subjects were treated with clomipramine for

5 weeks. Dosage was individually regulated; starting dose was 25 mg/day; increments were 25 mg/day. Maximum dose

was 250 mg/day or 5.0 mg/kg per day, whichever was less. Multiple raters, under several conditions, used the Children's

Psychiatric Rating Scale, Clinical Global Impressions, Conners Parent Teacher Questionnaire, and the Clinical Global

Consensus Ratings. Results: Eight children, aged 3.5 to 8.7 years, were enrolled in the study; seven of these completed

the study. A 3.5-year-old boy was excluded during the third week of treatment after having urinary retention on two

occasions. At doses ranging from 2.50 to 4.64 mg/kg per day (mean = 3.14), one child improved moderately and

six were rated as worse on the Clinical Global Consensus Ratings. Untoward effects were common. Conclusions:

Clomipramine was not therapeutic and was associated with serious untoward effects in this sample. Young autistic

children may be more prone to experience untoward effects than older patients. J. Am. Acad. Child Ado/esc. Psychiatry,

1996, 35(4):537-544. Key Words: clomipramine, autism, adverse effects.

Pharmacotherapy, in addition to a comprehensive psy­chosocial and educational program (Campbell andSchopler, 1989; Campbell et al., 1995), is often re-

Accepted May II, 1995.

Dr. Sanchez isAssistant Professor ofChild Psychiatry, University ojPennsylua­nia Medical School, Philadelphia. Dr. Campbell is Professor ofPsychiatry and

Dr. Small is Clinical Professor of Psychiatry, New York University MedicalCenter, New York. Dr. Cueva is Assistant Professor of Psychiatry, New YorkMedical College, New York. Dr. Armcnteros is Research Fellotoand Dr. Adams

is Assistant Clinical Professor ofPsychology and Psychiatry, College ofPhysicians

and Surgeons 0/Columbia University, New York. At the time 0/this studyDrs. Sanchez, Cueua, and Armenteros were Research Fellows, Department ofPsychiatry, New York University Medical Center.

Part of this paper was presented at the 147th Annual Meeting of the

American Psychiatric Association, Philadelphia, PA, May 21-26, 1994, andthe 41st Annual Meeting of the American Academy of Child and AdolescentPsychiatry, New York, October 25-.30, 1994.

This work was supported by the Fli Lilly Pilot Research Award for Junior

Faculty and Child Psychiatry Fellowsfrom the American Academy of Childand Adolescent Psychiatrya.s: USPHS grants T32 M H-18915 a.s.. M. C,j.E.C, j.L.A.) and MH-40177 (M.e) from the NIMH; the Hirschell E.and Deanna F Leuinc Foundation; the Beatrice and Samuel A. SeaverFoundation; and the Marion 0. and Maximilian E. Hofftnan Foundation,Ine. (M. C). The authors thank Mr. Thomas 13. CooperfOr analyzing eMI

and DCMI levels in plasma, and the New York Health and HospitalsCorporation, Bellevue Hospital Center, and the Bellevue Hospital Pharmacy

[or their cooperation.

Reprint requests to Dr. Sanchez, PCGc. 34th Street and Civic CenterBoulevard, Philadelphia, PA 19104.

0890-8567/96/3504-0537$03.00/0© 1996 by the American Academyof Child and Adolescent Psychiatry.

]. AM. ACAD. CHIl.D ADOl.ESC. PSYCHIATRY, 35:4, APRIl. 1996

qui red for a subgroup of children with autism whodisplay severe symptoms of aggressiveness toward selfand others, irritability, temper tantrums, and hyper­activity. Of all psychoactive agents, haloperidol hasbeen the only one studied systematically and is themost efficacious (Anderson et aI., 1984, 1989; Camp­bell et aI., 1978). However, not all children with autismwho require pharmacotherapy respond to haloperidol,and the risk of neuroleptic-related dyskinesias limitsthe usefulness of this drug (Campbell et aI., 1988a,1990; Malone et aI., 1991). Recent reports suggestthat clomipramine (CMI), a tricyclic antidepressantand potent serotonin reuptake inhibitor, may be aneffective alternative to neuroleptic treatment in autism(Brasic et al., 1994; Gordon et aI., 1993; McDougleet aI., 1992).

The potential usefulness of CMI in autism is sup­ported by research suggesting that disregulation of theserotonergic system plays a role in the pathogenesis ofautism. One third of autistic children have hyperseroto­nemia (Campbell et aI., 1975; Ritvo et aI., 1970),which appears to be associated with greater cognitiveimpairment, increased stereotypies, and severity of be­havioral disturbance (Campbell et aI., 1975; Ritvoet aI., 1986). In addition, decreased central 5-hydroxy­tryptamine responsiveness has been demonstrated inautistic adults (McBride et aI., 1989).

537

SANCHE/, ET AL.

Serotonergic disregulation has also been postulatedin the pathogenesis of obsessive-compulsive disorder(OCD) (Flament et al., 1985, 1987; Leonard et al.,1989; Mavissakalian et al., 1990; Thoren et al.,1980a,b), a disorder for which CMIhas been shownto be both efficacious and safe in pediatric (DeVeaugh­Geiss et al., 1992; Flament et al., 1985; Leonard et al.,1989) and adult (DeVeaugh-Geiss et al., 1989, 1990)populations, It has been suggested that stereotypies,ritualistic behaviors, and resistance to change, whichare common in autistic individuals and may accompanyor give way to obsessive-compulsive-like symptoms inadolescence (Wing and Attwood, 1987) may belongto the same behavioral spectrum as OCD and may,like OCD, respond to treatment with a serotonergicagent, e,g" CMI (Gordon et al., 1993). CMI wasreported to be therapeutic in subjects with autisticdisorder both in open and in double-blind, placebo­controlled studies; the positive effects included reduc­tion of obsessive-compulsive-like symptoms as well asstereotypies (Brasic et al., 1994; Gordon et al., 1993;McDougle et al., 1992). These positive findings werebased on the administration of CMI to school-agechildren, adolescents, and adults, all of whom wereoutpatients. There is some supportive evidence based onlarge number ofsubjects that response to a psychoactiveagent is more modest in younger children with autismthan in older ones (Campbell et al., 1978; Locascioet al., 1991) and that younger children are more proneto experience serious side effects (Campbell et al., 1971,1991; Petti and Campbell, 1975).

The objective of the present study was to explorethe efficacy and safety of CMI under careful clinicaland laboratory monitoring in a sample of hospitalizedyoung children with autistic disorder, with a relativelynarrow age range.

METHOD

Subjects

All subjects were inpatients in the psychiatric nursery of anacute psychiatric hospital setting. Autistic disorder with onset before36 months of age was diagnosed according to DSM-III-R criteria(American Psychiatric Association, 1987); the subjects also metthe DSM-IV criteria (American Psychiatric Association, 1994) forautistic disorder. The diagnosis was made by consensus of rwoclinicians (child psychiatrists) and independently by two seniorresearch child psychiatrists (M.e. and A.M.S.). Inrcrrarcr reliabilitywas 100%. One of the research child psychiatrists (A.M.S.) saw

538

the children only during the ratings and was blind to the drugunder study.

Exclusion criteria were as follows: autism with an identifiablecause (e.g., congenital rubella, inborn errors of metabolism, ctc.):history or clinical evidence of cardiac disease; seizure disorder orhistory of seizure disorder, including febrile seizures; history orclinical evidence of systemic diseases (e.g., renal, liver, endocrinolog­ical, or vascular); tardive or withdrawal dyskinesia; Tourerrc'sdisorder; and concurrent adminisuarion of any psychoactivemedication.

The eight subjects (pertinent data shown in Table 1) werereferred to the Psychiatric Nursery of Bellevue Hospital Center,Children's Inpatient Unit, with the presumptive diagnosis ofautisticdisorder. Their behavioral symptoms included withdrawal, deviantspeech, stereotypies, ritualistic behaviors, temper tantrums, aggres­siveness against others, self-injurious behavior, and hyperactivity.Before the study, the patients underwent a complete psychiatricand psychological evaluation, as well as a medical and neurologicalworkup. All subjects were free of medication (including antibiotics,psychoactive drugs, aspirin, etc.) at least 2 weeks before enrollmentin the study.

Design and Medication

This was an open drug trial: after a I-week placebo baseline,subjects were treated with clomipramine hydrochloride (CMI) for5 weeks. Since it was not possible to obtain matching placebofrom the manufacturer (Ciba-Geigy), subjects received I mL ofsimple syrup daily at 8:00 A.M. during placebo baseline. Themedication dose was individually regulated with increments of 25mg/day every 2 to 3 days until optimal therapeutic effects oruntoward effects (behavioral or other) were observed, but a maxi­mum dose of 250 mg/day or 5.0 mg/kg per day, whichever wasless, was not exceeded. The starting dose was 25 mg/day. CMIwas administrated in divided doses; at 25 mg/day it was given ina single 8:00 A.M. dose, at 50 mg/day it was administered rwicea day, and starting with 75 mg/day it was administered three timesa day. Upon completion of the study, the medication was taperedover a period of 10 days.

Assessments

ClinicalBehavioral Ratings, Ratings ofAbnormalMovements, andUntoward Effects. Behavioral ratings were conducted twice, onbaseline and at the end of the 5-weelc treatment period, as shownin Figure 1, on fixed days (Thursdays) and fixed times (8:00 A.M.)

by multiple trained raters in a variety of settings. In addition, aconsensus rating by the clinical staff concerning the child's responseto treatment was obtained on the Clinical Global Consensus Ratingsat the end of the treatment period. On the Clinical GlobalConsensus Ratings, children are rated as markedly, moderately, orslightly improved, no change, or worse. Assessment measures arcsummarized in Figure I.

Two senior research child psychiatrists (M.e. and A.M.S.)independently rated the children in a playroom setting, I hourafter drug administration, on measures developed by the Psycho­pharmacology Research Branch of the National Institute of MentalHealth (Psychopharmacology Bulletin, 1985)-the Children's Psychi­atric Rating Scale (CPRS), the Clinical Global Impressions (CGI),and the Abnormal Involuntary Movement Scale-and, in addition,on the Timed Stereotypies Rating Scale (Campbell, 1985). Asemistructured interview was used and videotaped (Campbell, 1985;Campbell and Palij, 1985). On the CPRS items are rated as 0 ~

J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 35:4, APRIL 1996

C LO M IPRA MI NE IN YOUNG AUT IST IC C H ILD REN

TABLE 1C ha racteristics or the Eigh t Subjects

SubjectNo .

2

3

4

5

6

7

Sex(CN)

M ale(4.0)

Male(3.5)

Male(R.7)

M ale(7.6)

Male(7.1)

Femal e(5 .2)

Male(8.2)

Male(6.8)

Level of Int ellectualFunctioning"

Borderline retarded; no nver bal

Mild ly reta rded; non verbal

Severely retarded; few wo rds

Mod erately reta rded;few words

Severely retarded; nonverbal

Mild ly retarded; nonverbal

M od eratel y retarded ; few wor ds

Mod erately retarded ; few words

T arget Symptoms

Stereo typ ies, hyperactivity, aggression,tan trums

Stereotypies, aggression, tantrum s. hyperact ivity

Stereo typies, ritu als (sn iffing hair, Rush ing toi let),aggression, tan tr u ms, liyperacriviry

Stereo typies, SIB," tant rums, labile mo od

Stereotypies, SIB, aggression, hyperactivity

Stereotypies, ritual s (linin g up obje cts), aggression,tant rums, hyperactivity

Stereoty pies. rituals (lining up objects),SIB. aggression. tant rum s

Stereotypies, ritua ls (lining up objects),SIB, aggression , tantrums

" CA ~ ch rono logical age, in years.f, Intellectual fun ct ionin g based on Adaptive D evelopm ent al Quotient, Gesell D evelopmental Sched ules (Gesell and

Amarr uda, 1947).r SIB ~ self-inju rious behavio r.

not assessed , I ~ not present. 2 ~ very mild, 3 ~ mi ld. 4 ~

mod erate, 5 ~ modera tely severe, 6 ~ severe, and 7 ~ extremelysevere. O nly the first 2R item s of the C PRS were rated , and ofthese 14 selected items appropri ate for aut istic child ren (Camp belland Palij, 1985 ) were analyzed , as were 4 factors der ived fromthese 14 items (Autism, Anger/Uncoop erativen ess, H yperactivity,and Speech Dev iance) (Overall and Ca mpbe ll, 1988 ). On theCG I, severi ty of illness is rated on a 7-po inr scale as 0 ~ no tassessed, 1 ~ normal, not at all ill, 2 ~ bor de rline mentally ill, 3 ~

mi ld ly ill, 4 ~ moderately ill, 5 ~ markedly ill, 6 ~ severely ill,and 7 ~ among the most extremely ill pa tients . Improvement asmea sured on the C G I is also rated as 0 ~ not assessed , I ~ verym uch improved. 2 ~ mu ch improved , 3 ~ mini mally improved,4 ~ no change , 5 ~ mi nimally worse. 6 ~ m uch wo rse. and 7 ~

very mu ch wo rse. T he nursery school teacher rated the child renin the classroom on the Conners Paren t Teacher Q uestionnaire

..__. . . . Dose Titration --· ·- Oplimotl Dose( 'LOM) I'I{ AMI NE )'I{EATMI:NT

{+ ) ( +) (+) (-t-) (+

Fig. 1 Study design and assessments. Respiratory rate, pulse rate, bloodpressure. and untoward cflecrs were assessed daily. + ~ electrocard iogram :(+) ~ electrocardiogram prior to dose increment; • = weighr; • ~ laboratorystudies, Children's Psychiatric Rating Scale, Clinical Global Impressions,Parent Teacher Questionnaire, Abnormal Involuntary Movement Scale,T imed Stereotypies Rating Scale; • = Clinical Global Consensus Ratings.

J. AM . AC AD. C H Il. D ADO I.ESC. PSYCHI ATRY, 35 :4 , APRIl. 19 96

(PTQ) (Psychopharmacology Bulletin, 1985). PTQ sco res arc 0 (no tat all), I (just a lit tle), 2 (pre tty mu ch), and 3 (very mu ch).Untoward effects were rated daily on a check list develo ped for C M !.

Physical and Laboratory Measures. Height was measu red onadm ission and wcigh t was measured weekly, Wednesdays at 10:00A.M ., fo llowing a pro tocol. Respirator y rate, supine and stand ingblood pressurc, and heart rates were moni tored daily at 10:0 0 A.M.

Sleep, appetite, urine output, and bowel movements were record eddaily on a 24-hour basis by the nursing staff. T welve-lead electrocar­d iograms (EC Gs) were obtained at baseline, prior to cach do sagechange , 10 days after optim al dose, last day of the optimal dose,and at end of the study after C M ] was tapered. T he EC Gs werereviewed by a pediatr ic card iologist and the PR, Q RS, Q T interval,and the amplitude of the T wave in leads V5 and II were measured ,Laborator y studies (complete blood cell co unt with differen tia l,liver funct ion tests, electro lytes, and uri ne ana lysis) were repeateddu ring the last week of C M f rrcarmc nr and after discontinu at ionof C M!. Veno us blood was d rawn on T uesdays at 7:00 A.M ., pr iorto medication ad mi nistration.

Clomipramine and N-Desmethylcfomipmmine Levels in Plasma.During the fourth an d fifth week of C M ! treatm ent, J6 hoursafter the administrat ion of the previous do se of d rug, venous bloodsam ples of 6 mL were drawn in to a syringe. im mediately transferredto an oxalatcd glass tube, and centrifuged at 1,70 0 rpm for 10minutes. Plasm a was asp irated and frozen at 20°C unt il analyzedfo r C M f and N-dcsmeth ylclomiprami nc (DCM I) levels.

C M I and D CM I were quan titated using a gas-liquid chromatog­raphy systcm fitted with a nitroge n detecto r. C M I, DC M I, thein terna l standards im ipram ine, and dcsmcthyli mi prami ne were

539

SANCHEZ ET AL.

extracted from a basic solution with a 1.5% isoamyl alcohol!heptane solution. The organic layer was then back-extracted with0.1 mol!L hydrochloric acid. This acid extract was then madebasic and rccxrracred into a small volume of 15% isoamyl alcohol!toluene. Small aliquots of this organic phase were injected into a15-meter DB 17 megabore column (]&W Scientific) at 80°C,programmed at 300/minute to 260°C. The injector and detectorwere heated at 265°C and 275°C, respectively. This procedure hashigh sensitivity in that as little as 4 ng/mL (I mL plasma extracted)can be quantitated.

Statistical Analysis

The effectiveness of CMI on behavioral measures was assessedwith a matched-pairs t test of the difference between ratings atoptimal dose and baseline. The effects of CMI on blood pressureand heart rates were analyzed using a matched-pairs t test of thedifference between the mean of all optimal and baseline bloodpressure values for each child. Analyses of systolic and diastolicblood pressures and heart rates were performed separately for supineand standing assessments.

RESULTS

Sample

Characteristics including behavioral target symptomsof the eight children enrolled in the study are summa­rized in Table 1. The children's ages ranged from 3.5to 8.7 years. Intellectual functioning ranged from aborderline to a severely retarded level as assessed by theGesell Developmental Schedules (Gesell and Amatruda,1947). Four children were Hispanic, two were African­American, and two were Caucasian. Three childrenbelonged to social class II, 2 to social class IV, and 3 tosocial class V on the Hollingshead index (Hollingshead,1965). Only two of the eight children had a historyof treatment with other psychoactive agents, one withnaltrexone and one with haloperidol. Seven children,six boys and one girl, aged 4.0 to 8.7 years (mean =

6.8), completed the study; one 3.5-year-old boy wasexcluded during the third week of treatment with CMIafter having acute urinary retention on two separateoccasions within 8 hours.

Dosage, Untoward Effects, and Physical Data

Untoward effects and the lowest and highest doseof CMI at which they occurred are summarized inTable 2. Optimal therapeutic doses of CMI rangedfrom 50 to 175 mg/day (mean = 103.57) or 2.50 to4.64 mg/kg per day (mean = 3.43). The highest doseexplored ranged from 50 to 175 mg/day (mean =

103.12) or 2.83 to 4.65 mg/kg per day (mean = 3.14).

540

All eight children experienced at least one untowardeffect and five children experienced three or more.Acute urinary retention requiring catheterization oc­curred on the 16th day of CMI administration, twicewithin an 8-hour period, in the youngest subject, a3.5-year-old boy who was receiving 50 mg/day (3.15mg/kg per day). Severe constipation occurred in thefirst four children treated with daily doses of 50 mg(1.45 mg/kg) to 75 mg (3.15 mg/kg) of CMI. Afterthe protocol was changed to increase fluid intake andto include a high-fiber diet and Metamucil (one table­spoon, three times a day), for all subjects, no childexperienced constipation. Four children had insomniawhile receiving 100 mg/day (3.39 mg/kg) to 175 mg/day (4.54 mg/kg). Four children exhibited behavioraltoxicity which included increases in aggression againstothers, irritability, temper tantrums, self-injurious be­havior (biting fingers and toes, picking skin), and cryingspells while receiving 75 mg/day (3.39 mg/kg) to125 mg/day (4.37 mg/kg). Behavioral toxicity wasfrequently associated with insomnia; neither behavioraltoxicity nor insomnia occurred at doses below 3.0 mg/kg per day. The most common untoward effect wasdrowsiness, which occurred in seven children at dailydoses of 25 (1.06 mg/kg) to 175 mg (4.54 mg/kg).Other untoward effects were enuresis, licking lips, anddiarrhea. Both systolic and diastolic standing and supineblood pressures were higher during CMI treatment thanat baseline, but not significantly. Laboratory studies andECGs remained unchanged from baseline.

Behavioral Assessments

ClinicalGlobal Consensus Ratings. Of the seven chil­dren who completed the study, one was rated moder­ately improved and six were rated as worse on thismeasure.

Clinical Global Impressions. The mean baseline CGISeverity of Illness ratings for the seven children whocompleted the study was 5.85 (Table 3). CGI Severityof Illness ratings did not differ significantly at end oftreatment from baseline (Table 3).

No child experienced significant clinical improve­ment with CMI treatment as rated on the CGl. Infact, two of the seven children were rated as worse atend of treatment.

Children s Psychiatric Rating Scale. Mean baselineand end-of-treatment ratings and the average differencebetween these scores for all seven children for the sum

J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 35:4, APRIL 1996

CLOMIPRAMINE IN YOUNG AUTISTIC CHILDREN

TABLE 2Untoward Effecrs in Eight Children Treated with Clomipramine (CMI)

CMI Daily Dose

No. of Subjects Lowest Dose Highest Dose

Untoward Effects with Untoward Effects mg/day mg/kg/day mg/day mg/kg/day

Drowsiness 7 25 1.06 175 4.54Insomnia 4 100 3.39 175 4.54Constipation 4 50 1.45 75 3.15Diarrhea I 25 1.06 50 2.04Urinary retention (twice within 8 hr

in a 3.5-yr-old boy) I 50 3.15 50 3.15Enuresis I 50 1.78 lOa 3.50Licking lips I 75 2.62 75 2.62Behavioral toxicity" 4 75 3.39 125 4.37

" Aggression against others, temper tantrums, irritability, self-injurious behavior, crying spells.

of 14 selected items and the 4 CPRS factors (Autism,AngeriUncooperativeness, Hyperactivity, Speech Devi­ance) are reported in Table 3. The sum of 14 selectedCPRS items score and the 4 CPRS factor scores didnot differ significantly from baseline at the end oftreatment. In addition, CPRS item ratings did notdiffer significantly from baseline to end of treatment.However, some reduction in stereotypies was noted.Stereotypies decreased in four, increased in two, andremained unchanged in one child as measured on item28 (stereotypies) of the CPRS (Fig. 2).

Parent Teacher Questionnaire. Mean baseline andend-of-treatment PTQ total scores are reported inTable 3. The PTQ total score is the sum of individualratings on the 10 items of this instrument. Analysis

of this measure was limited to four children becauseof missing data. No significant difference betweenbaseline and end of treatment was detected on thismeasure.

Plasma Drug Levels

Because of difficulties obtaining blood samples, onlytwo of the seven children who completed the studyhad venous blood drawn on weeks 4 and 5 for plasmadrug levels. The remaining five children had only onesample drawn on either week 4 or 5 of treatment withCMI. Children were compliant with medication asevidenced by plasma levels. Plasma levels varied widely,ranging from 23 to 73 ng/mL (mean = 39.4) for CMIand 32 to 132 ng/mL (mean = 89.4) for DCMI anddid not correlate with behavioral ratings.

7

•6 DISCUSSION

5

Fig. 2 Mean scores of stereotypies (item 28, CPRS) on baseline and atend of treatment with clomipramine. Children's Psychiatric Rating Scale(CPRS): 0 = not assessed, 1 = not present, 2 = very mild, 3 = mild, 4 =

moderate, 5 = moderately severe, 6 = severe, 7 = extremely severe.

In this open study, involving a small sample ofyoung children with autism, CMI was not therapeuticand untoward effects were serious and common.

Our findings differ from those of the only double­blind, placebo-controlled study (Gordon et al., 1993)and from two open studies (Brasic et al., 1994;McDougle et al., 1992). The discrepancy betweenthe present report and previous reports may reflectdifferences in samples. The sample in the present studydiffers from that in the Gordon et al, (1993) reportwith respect to age, intellectual functioning, presenceof OCD symptoms, and patient status. The samplein the present study is significantly younger than the

TreatmentBaselineo

2

]. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 35:4, APRIL 1996 541

SANCHEZ ET AL.

TABLE 3Behavioral Ratings: Mean Baseline and End-of-Treatment Scores for Seven Children

Mean Score AverageBehavioral Ratings Baseline End of Treatment Difference p

Sum of 14 CPRS items 3.27 3.26 -.01 .95CPRS Autism factor 5.04 4.81 -.22 .33CPRS Anger/Uncooperativeness 2.10 2.44 .33 .53Cl'RS Hyperactivity factor 4.21 3.85 -.35 .45Cl'RS Speech Deviance 2.57 2.32 -.25 .54CG I Severity rating 5.85 6.14 .28 .32l'TQ Total Score (n = 4) 18.0 18.75 .75 .88

Note: CPRS = Children's Psychiatric Rating Scale; CGI Clinical Global Impressions; PTQ = Parent TeacherQuestionnaire.

one in the Gordon et al. (1993) report (mean age 6.40versus 9.67 years; p = .01) and younger than those inother reports, as shown in Table 4. In addition, whenthe Gordon er al. (1993) and the present sample arecontrolled for age and subsamples consisting ofchildrenbetween the ages of 6 and 9 years are compared,intellectual functioning in the present study is signifi­cantly lower (p = .028). All subjects in the Gordonet al. (1993) as well as in the McDougle et al. (1992)reports had OCD symptoms, whereas subjects in thepresent study had stereotypies (Table 1 and Fig. 2) butdid not exhibit prominent OCD symptoms. Subjects inprevious reports were outpatients, while subjects in thepresent study were inpatients (Table 3), suggesting thatthey may have had greater severity of symptoms.

Therapeutic trials of other agents affecting the sero­tonergic system, in particular t-dopa and fenfluramine,

have met with mixed success (Campbell et al., 1976,1988b; for review, see Campbell and Schopler, 1989),and these drugs are not believed to be efficacious inthe treatment of autism.

Untoward effects were common in this sample ofeight young autistic children who were under close24-hour monitoring. Among the most disturbing wasthe development of acute urinary retention, twicewithin an 8-hour period, in the youngest subject, a3.5-year-old boy. Difficulty with urination has beenreported by Flament et al. (1985) to occur in 11%(N = 2/19) and by Leonard et al. (1989) in 4% (N =

2/45) of children and adolescents with OCD treatedwith CMI, although the severity of the urinary difficul­ties is unclear. Acute urinary retention occurred twicein a 15-year-old male treated with CMI for OCD(Hermesh et al., 1987). There is supportive evidence

TABLE 4Reports of Clomipramine in Autism

Intellectual Functioning

Average-Mildly Moderately-SeverelyRetarded RetardedStudy

Gordon ct al., 1993

McDougle cr al., 1992

Brasic er al., 1994

Sanchez cr al,(present study)

rr NA = not available.* p = .01 (t test).

No. of Age in Years:Subjects Range

Study Design (Status) (Mean)

Double-blind, 24 6-18placebo-con trolled, (Outpatients) (9.67)*crossover

Case reports 5 13-33(Outpatients) (25.20)

Open trial 5 6-12(Outpatients) (NA)"

Open trial 8 3.50-8.70(Inpatients) (6.40)*

16

4

o

3

8

5

5

542 ]. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 35:4, APRIL 1996

that age may be an important factor, not only inthe response to treatment with a psychoactive agent(Campbell et al., 1978; Locascio et al., 1991), but alsoin the development of side effects (Campbell er al.,1991). Data indicate that acute urinary retention ismore common in the youngest age group than inadults. Seven percent of 46 children and adolescentsand only 2% of 322 adults treated with CMI hadurinary retention, while no adult (N = 319) or childor adolescent (N = 44) receiving placebo had urinaryretention (Physicians'Desk Reftrence, 1995). In a studyof imipramine hydrochloride involving 10 childrenwith autism, aged 2 to 6 years, half of the sampleexperienced behavioral worsening and a 3-year-old boyhad seizures and status epilepticus (Campbell et al.,1971; Petti and Campbell, 1975).

While no definite conclusions can be drawn giventhe methodological limitations of the present study,our data suggest that further studies involving largersamples of young children are needed before CMI canbe recommended as a treatment modality for youngchildren with autism.

REFERENCES

American Psychiatric Association (1987), Diagnostic and StatisticalManual~(Memtl!Disorders, 3rd cdiiion-reoised (DSM.!!!.R). Washington, DC:American Psychiatric Association

American Psychiatric Association (1994), Diagnostic and StatisticalManualofMenttl! Disorders, 4th edition (DSM·!V). Washington, DC: AmericanPsychiatric Association

Anderson LT, Campbell M, Adams 1', Small AM, Perry R, Shell J (1989),The cflccrs of haloperidol on discrimination learning and behavioralsymptoms in autistic children. J Autism Dev Disord 19:227-239

Anderson LT, Campbell M, Grega DM, Perry R, Small AM, GreenWH (1984), Haloperidol in infantile autism: effects on learning and

behavioral symptoms. Am J Psychiatry 141: 1195-1202Brasic JR, Barnett JY, Kaplan D er al. (1994), Clomipramine ameliorates

adventitious movements and compulsions in prepubertal boys withautistic disorder and severe mental retardation. Neurology 44: 1309-1312

Campbell M (1985), Timed Srcrcorypies Rating Scale. PsychopbarmncolBII1121:1082

Campbell M, Adams 1', Perry R, Spencer EK, Overall JE (l988a), Tardiveand withdrawal dyskinesia in autistic children: a prospective study.Psychopharmacol Bull 24:251-255

Campbell M, Adams 1', Small AM et al. (I 988b), Efficacy and safety offenAuramine in aurisric children, JAm Acad ChiM Adolesc l',ychiat'y27:434-439

Campbell M, Anderson LT, Meier M er al. (1973), A comparison ofhaloperidol, behavior therapy and their interaction in autistic children.JAm Acad cuu Psychiatry 17:640-655

Campbell M, l'ish B, Shapiro '1', Floyd A (1971), Imipramine in preschoolautistic and schizophrenic children. J Autism Child Schizophr1:267-282

Campbell M, Friedman E, Green WH, Collins P], Small AM, Breuer I-!(1975), Blood serotonin in schizophrenic children. A preliminary study.1m Pharmncopsycbiatry 10:213-221

J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 35:4, APRIL 1996

CLOMIPRAMINE IN YOUNG AUTISTIC CHILDREN

Campbell M, Locascio JJ, Choroco MC et al. (1990), Stereotypies andtardive dyskinesia: abnormal movements in autistic children. P'yC!Jop/Jflr­mncol 15/111 26:260-266

Campbell M, Palij M (1985), Behavioral and cognitive measures usedin psychopharmacological studies of infantile autism. 1'.,ychoplJtlrmacolBII1121: I047-1 053

Campbell M, Schoplcr E (1989), Pervasive developmental disorders. In:Treatments of Psychiatric Disorders, Vol I. A Task Force Report of theAmerican Psychiatric Association, Karasu '1'13 (Chairperson). Washing.ton, DC: American Psychiatric Association, PI' 179-294

Campbell M, Schoplcr E, Mesibov GB, Sanchez LE (1995), Pervasivedevelopmental disorders. In: Treatment of Psychiatric Disorders: TheDSM·1V Edition, Gabbard GO, ed. Washington, DC: American Psychi­atric Press, PI' 151-178

Campbell M, Silva RR, KafanrarisV cr al. (1991), Predictors of side effectsassociated with lithium adminisrr.uion in children. PsyC!JojJ}}{lrnMco!

Bull 27:373-380Campbell M, Small AM, Collins P], Friedman E, David R, Gcnicscr

NB (1976), I.evodopa and levoamphetamine: a crossover study inschizophrenic children. Curr Ther Res 18:70-86

Dc'Vcaugh-Cciss J, Katz R, Landau 1', Goodman W, Rasmussen S (1990),Clinical predictors of treatment response in obsessive compulsive disor­der: exploratory analyses from multicenter trials ofclomipramine. Psycho­pbarmncol Bull 26:54-59

DeVeaugh.Geiss J, Landau P, Katz R (1989), Preliminary results from a

multicenter trial of clomipramine in obsessive-compulsive disorder.PsychophttrlnacolliuI125:%-40

DcVcaugh-Cciss J, Moroz G, Biederman J ct al. (1992), Clomipraminehydrochloride in childhood and adolescent obsessive-compulsive disor­del': a multicenter trial. J Alii Acad Child Adolesc Psychiatry 31 :45-49

Flament MF, Rapoport ]L, Berg CJ cr al. (1985), Clomipramine treatmentof childhood obsessive-compulsive disorder: a double-blind controlledstudy. Arch Gen Psychiatly 42:977-983

Flarnent MF, Rapoport JL, Murphy DL, Berg CJ, Lake CR (1987),Biochemical changes during clomipramine treatment of childhoodobsessive-compulsive disorder. Arch Gen Psychiatry 44:219-225

Gesell A, Arnatruda CS (1947), Developmental Diagnosis, 2nd cd. NewYork: Harper & Row

Gordon CI', State RC, Nelson JE cr al. (1993), A double-blind comparisonof clomipramine, desipramine, and placebo in the treatment of autisticdisorder. Arch Gl'II Psychiatry 50:441-447

Hermesh H, Aizenberg D, Weizman A, Lapidot M, Muuirz H (1987),Clomipramine-induced urinary dysfunction in an obsessive-compulsiveadolescent. Drug Intell Clin Pbarm 21:877-879

Hollingshead AB (1965), Tioo-Factor Index ofSocial Position (1957). NewHaven, CT: Yale University Department of Sociology

Leonard FlL, Swede SE, Rapoport JL cr al. (1989), Treatment of obsessive­compulsive disorder with clomipramine and desipramine in childrenand adolescents: a double-blind crossover comparison. Arch GenPsychia­tly 46:1088-1092

Locascio JJ, Malone Rl', Small AM ct al. (1991), Factors related tohaloperidol response and dyskincsias in autistic children. Psychopbnrma­colliuI127:119-126

Malone RI', Ernst M, Godfrey KA, Locascio JJ, Campbell M (1991),Repeated episodes of neuroleptic. related dyskincsias in autistic children.Psychopbarmncol 151111 27: 113-117

Mavissakalian MR, Jones 13, Olson S, Pcrcl JM (1990), Clomipramine inobsessive-compulsive disorder: clinical response and plasma levels. J ClinPsychopharmacol 10:261-268

McBride PA, Anderson GM, I-Iertzig ME ct al. (1989), Serotonergicresponsiviry in male young adults with autistic disorder. Arch GenPsychiatry 46:213-221

McDougle CJ, Price LH, Volkmar I'll. er al, (1992), Clomipramine inautism: preliminary evidence of efficacy. J Am ACtld Child AdolescPsychiatry 31 :746-750

Overall JE, Campbell M (1988), Behavioral assessment of psychopathologyin children: infill1tile autism. J Clin PsychoI44:708-716

Petti TA, Campbell M (1975), Imipramine and seizures. Am J Psycbia­try 132:538-547

543

SANCHEZ ET AI..

Physicians 'DeskReference 49th cd (1995), Montvale, NJ: Medical EconomicsData Production Company

Psychopharmacolof!J1 Btllletin (1985), Special Feature: Rating Scales andAssessment Instruments for Usc in Pediatric Psychopharmacology Re­search, 2 I(4)

Rirvo ER, Freeman BJ, Yuwilcr A ct al, (1986), FenAuramine therapy forautism: promise and precaution. Psychopharmacol13t11122:133-140

Ritvo ER, Yuwilcr A, Geller E, Ornirz EM, Saeger K, Plotkin S (1970),Increased blood serotonin and platelets in early infantile autism. ArchGen Psychiatry 23:566-572

544

Thoren P! Asberg M, Bertilsson 10, Mcllsrrorn B, Sjoqvisr F, Traskman 10(l980a), Clomipramine treatment of obsessive-compulsive disorder: II.Biochemical aspects. Arch Gen Psychiatry 37:1289-1294

Thoren P! Asberg M, Cronholm B, Jornestedt 10, Traskrnan 10 (I 980b),Clomipramine treatment of obsessive-compulsive disorder: 1. A con­trolled clinical trial. Arch Gen Psychiatry 37:1281-1285

Wing 10, Attwood A (1987), Syndromes of autism and atypical development.In: Handbook ofAutism and Pervasive Developmental Disorders, CohenDJ, Donnellan AM, cds. New York: Wiley & Sons, pp 3-19

J. AM. ACAD. CHILD ADOLESC. PSYCHIATRY, 35:4, APRIL 1996