neuroblastoma and atxia

7/28/2019 Neuroblastoma and Atxia

http://slidepdf.com/reader/full/neuroblastoma-and-atxia 1/15

2002;109;86-98Pediatrics

Elvira Burger, Susan B. Turkel, Mark S. Borchert, Susan Hollar and Sonia PadillaWendy G. Mitchell, Yolanda Davalos-Gonzalez, Virdette L. Brumm, Sonia K. Aller,

Neurologic SequelaeOpsoclonus-Ataxia Caused by Childhood Neuroblastoma: Developmental and

http://www.pediatrics.org/cgi/content/full/109/1/86located on the World Wide Web at:

The online version of this article, along with updated information and services, is

reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.Village, Illinois, 60007. Copyright © 2002 by the American Academy of Pediatrics. All rightstrademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove

andpublication, it has been published continuously since 1948. PEDIATRICS is owned, published,PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

. Provided by Indonesia:AAP Sponsored on April 5, 2011www.pediatrics.orgDownloaded from

http://www.pediatrics.org/cgi/content/full/109/1/86



http://www.pediatrics.org/





Opsoclonus-Ataxia Caused by Childhood Neuroblastoma:Developmental and Neurologic Sequelae

Wendy G. Mitchell, MD*#**; Yolanda Davalos-Gonzalez, BA*; Virdette L. Brumm, PhD*#**;

Sonia K. Aller, PhD, CCC‡; Elvira Burger, BArb§; Susan B. Turkel, MDʈ‡‡; Mark S. Borchert, MD¶#§§;Susan Hollar, MS, CCCʈʈ; and Sonia Padilla, MA*

ABSTRACT. Objective. Opsoclonus-ataxia, also called“dancing eye syndrome,” is a serious neurologic condi-tion that is often a paraneoplastic manifestation of occultneuroblastoma in early childhood. Despite resection oftumor and immunosuppressive therapy, outcome gener-ally includes significant developmental and behavioralsequelae. There is controversy about how treatment al-ters outcome. The goals of this study were to understandthe ongoing neurologic and developmental deficits ofchildren who are treated for opsoclonus-ataxia with as-

sociated neuroblastoma; to relate treatment history tooutcome; and to quantify objectively the acute changes inmotor function, speech, mood, and behavior related tointravenous immunoglobulin (IVIg) treatment.

Methods. Patients were children with opsoclonus-ataxia caused by neuroblastoma, regardless of intervalsince diagnosis. Records were reviewed, and childrenunderwent comprehensive evaluations, including neuro-logic examination and tests of cognitive and adaptivefunction, speech and language, and fine and gross motorabilities. Psychiatric interview and questionnaires wereused to assess current and previous behavior. In 6 chil-dren, a videotaped standardized examination of eyemovements was performed. Additional examinationswere performed immediately before and 2 to 3 days after

treatment with IVIg in 5 children.Results. Seventeen children, ages 1.75 to 12.62 years,

were examined. All had a stage I or II neuroblastomaresected 3 months to 11 years previously. None receivedany other treatment for the tumor. All but 1 had receivedat least 1 year of either oral corticosteroids or cortico-tropin (ACTH); 12 had received 1 or more courses of IVIg,2 g/kg. Three had received other immunosuppressivetreatment, including cyclophosphamide. Cognitive de-velopment and adaptive behavior were delayed or abnor-mal in nearly all children. Expressive language was moreimpaired than receptive language. Speech was impaired,including both intelligibility and overall output. Fineand gross motor abilities were impaired. Increased agewas strikingly associated with lower scores in all areas.Behavioral problems early in the course included severe

irritability and inconsolability in all; later, oppositionalbehavior and sleep disorders were reported. Opsoclonusabated in all, but abnormalities in pursuit eye move-ments were found in all 6 children cooperative withstandardized examination. Outcome did not differ inchildren who were treated with ACTH versus oral ste-roids. Three children who had received cyclophospha-mide fared poorly. Immediate versus delayed treatmentwas not associated with better outcome. IVIg improvedboth gross and fine motor and speech function acutely,

but we could not confirm long-term benefit of IVIg. Totalnumber of courses of IVIg was not associated with out-come.

Conclusions. Opsoclonus-ataxia caused by neuroblas-toma causes substantial developmental sequelae that arenot adequately prevented by current treatment. The in-creased deficits in older children raise concern that thisrepresents a progressive encephalopathy rather than atime-limited single insult. Although the study is cross-sectional and neither randomized nor blinded, we wereunable to confirm a purported advantage of either ACTHover corticosteroids or of cyclophosphamide. A random-ized study is needed but is difficult for this rare condi-tion. Pediatrics 2002;109:86–98; opsoclonus, ataxia, neuro-blastoma, paraneoplastic disorders, developmental sequelae.

ABBREVIATIONS. IVIg, intravenous immunoglobulin; CHLA,Childrens Hospital Los Angeles; OKN, optokinetic; SD, standarddeviation; MLU-m, mean length of utterance in morphemes.

Paraneoplastic neurologic syndromes are neuro-nal degenerative disorders that occur as anindirect result of malignancies.1–5 In adults,

paraneoplastic cerebellar degeneration, with or with-out opsoclonus, occurs primarily in association with

breast or ovarian cancer or small cell carcinoma of the lung. In children, the syndrome of opsoclonus,

cerebellar ataxia, and myoclonus occurs as a para-neoplastic neurologic syndrome related to occult,low-grade neuroblastomas.6–11 The syndrome occursalmost exclusively in young children (6 months to 3years of age) and is generally thought to occur inapproximately 2% to 3% of children with neuroblas-toma.12,13 The syndrome occurs primarily in childrenwith localized, small neuroblastomas (stage I or II),with favorable histologic features, and without am-plification of the N-myc gene.14,15 Survival in chil-dren with neuroblastoma presenting with opsoclo-nus-ataxia is substantially higher than in childrenwith neuroblastoma without neurologic presenta-tion.7,9,16–19 A recent review of children registered in

From the *Neurology Division, ‡University Affiliated Program, §Division of

Physical and Occupational Therapy, ʈPsychiatry Program, and ¶Ophthal-

mology Division, Childrens Hospital Los Angeles, Los Angeles, California;

Departments of #Neurology, **Pediatrics, ‡‡Psychiatry, and §§Ophthalmol-

ogy, Keck School of Medicine, University of Southern California, Los An-

geles, California; and ‡‡California State University Northridge, Northridge,

California.

Received for publication Mar 19, 2001; accepted Jun 25, 2001.

Reprint requests to (W.G.M.) Childrens Hospital Los Angeles, Neurology

Division, Box 82, 4650 Sunset Blvd, Los Angeles, CA 90027. E-mail:

[email protected]

PEDIATRICS (ISSN 0031 4005). Copyright © 2002 by the American Acad-

emy of Pediatrics.

86 PEDIATRICS Vol. 109 No. 1 January 2002. Provided by Indonesia:AAP Sponsored on April 5, 2011www.pediatrics.orgDownloaded from






the Children’s Cancer Group Neuroblastoma Regis-try found 100% survival in children with stage I or IIneuroblastoma presenting with opsoclonus-ataxia.13

Children who have higher stage (metastatic or recur-rent) neuroblastoma with N-myc amplification andwho have opsoclonus-ataxia, either as a presentingfeature or later in their course, do not seem to havethis good prognosis, although the number of casesreported is very small.20

It is commonly assumed that paraneoplastic neu-

rologic syndromes occur when there is antigenic sim-ilarity between tumor cells and neurons. Althoughantibodies may provide some biological control of the tumor, they promote progressive damage to tar-get neurons.21 In several adult syndromes of para-neoplastic cerebellar degeneration, identifiable anti-

bodies are associated with the syndrome (anti-Yo ingynecologic malignancies, anti-Hu in small cell car-cinoma of the lung).22,23 Recently, Albert et al24 re-ported that cytotoxic T-lymphocytes were present in

blood and cerebrospinal fluid of adults with para-neoplastic cerebellar degeneration. For one syn-drome in which the tumor antigen has been cloned,

the cytotoxic T-cells were directed against this anti-gen.In children with opsoclonus-ataxia, antineuronal

antibodies are less consistently found and their asso-ciated antigen is still uncertain, although Connolly etal25,26 recently reported finding antibodies to variouscomponents of cerebellar neurons in at least somechildren with opsoclonus-ataxia. In a study of

banked serum samples from children with neuroblastoma with and without opsoclonus-ataxia, an-tineuronal antibodies were found more frequently inthe children with opsoclonus-ataxia but were foundin some children with neuroblastoma alone and were

highly variable.27

Anti-Hu and Anti-Yo antibodieshave been found only rarely.28,29 T-cell activation inchildhood opsoclonus-ataxia has not yet been dem-onstrated but is suspected to be involved in themechanism of both tumor suppression and neuronaldamage.

Adult paraneoplastic cerebellar degeneration isneuropathologically characterized by severe loss of Purkinje cells.30 In children with opsoclonus-ataxiaand neuroblastoma, tumors are small and indolentand rarely metastasize or recur after resection, soneuropathological examinations are not available.However, it is generally assumed that similar mech-anisms apply to pediatric and adult paraneoplasticsyndromes.

Despite treatment with corticosteroids, cortico-tropin (ACTH), intravenous ␥ globulin (intravenousimmunoglobulin [IVIg]), and a variety of other im-munosuppressant medications, most children whohave had paraneoplastic opsoclonus-ataxia have sig-nificant residual neurologic and developmental def-icits.13–15,19,31 Ongoing treatment, usually with corti-costeroids, continuously or intermittently, is oftenrequired to maintain function such as ambulation,speech, and continued developmental gains.19,31

Children with nonparaneoplastic (presumablypostinfectious) opsoclonus-ataxia are also noted to

have severe neurologic sequelae despite prolongedtreatment with corticosteroids and/or ACTH.32

Other immune-modulating treatments including in-travenous ␥ globulin (IVIg),33,34 cyclophosphamide,azathioprine, and plasmapheresis have been used,with variable reported success. As recently as the1980s, it was assumed that the neurologic syndromeoften remitted spontaneously after surgical removalof the tumor.35 However, several reports have con-cluded that neurologic symptoms commonly con-

tinue after tumor resection and that developmentrarely returns fully to normal.14,19,31 Currently, mostpediatric neurologists and oncologists who are expe-rienced with opsoclonus-ataxia begin treatment of the neurologic symptoms immediately. However,duration of treatment varies widely, as do goals of treatment, ranging from complete suppression of allneurologic symptoms to more modest goals such asreestablishing independent ambulation.

The role of chemotherapeutic agents is uncertain.One study reported that children who were treatedwith chemotherapy had better neurologic out-comes.15 Treatment was presumably for the tumor

but with secondary effects of significant immuno-suppression. A single patient, reported in 1977, whodeveloped ataxia without opsoclonus after detectionof neuroblastoma received extensive chemotherapy,with outcome reported to be “good” with only mildmental retardation.36 One patient, published in aprevious series at this institution, who had apparentspontaneous resolution of opsoclonus-ataxia imme-diately after resection of a stage III neuroblastomareceived multiagent chemotherapy and reportedlyhad a good neurologic outcome.19

IVIg has been used as an immunosuppressant ther-apy for a variety of neurologic conditions with pre-

sumed autoimmune causes, including chronic immunedemyelinating peripheral neuropathy, Guillain-Barresyndrome,37 adult paraneoplastic cerebellar degenera-tion,38,39 and acute demyelinating encephalomyelopa-thy.40 Only in Guillain-Barre, chronic immune demy-elinating peripheral neuropathy, and multifocalmotor neuropathy has use of IVIg been the subject of formal randomized, blinded, clinical trials.41–43

Nonetheless, IVIg is thought to be useful in at leastsome individuals. No manufacturer of IVIg has yetobtained a labeled indication for any chronic neuro-logic condition. IVIg has been used for opsoclonus-ataxia with varying dosage schedules, in various

combinations with corticosteroids, ACTH, or othermedications.44–47 Although parents often reportnear-immediate changes in neurologic function afteran IVIg treatment, these changes have been difficultto quantify or confirm.

This study was undertaken with several goals: 1)to understand the ongoing neurologic and develop-mental deficits of children who are treated for opso-clonus-ataxia associated with neuroblastomas; 2) toattempt to relate treatment history to outcome; and3) to attempt objectively to quantify improvements inmotor function, speech, mood, and behavior relatedto IVIg treatment.

ARTICLES 87. Provided by Indonesia:AAP Sponsored on April 5, 2011www.pediatrics.orgDownloaded from





METHODS

Patient Recruitment

Patients who were followed in the Neurology Clinic for opso-clonus-ataxia caused by neuroblastomas at Childrens Hospital LosAngeles (CHLA) were invited to participate in the study. In ad-dition, an announcement of the study was placed on the childneurology Internet bulletin board and on the web site of a parent-run support group for opsoclonus-ataxia. Patients who had notpreviously been treated at CHLA contacted the researchers either

by electronic mail or telephone and were prescreened for inclu-sion. All parents signed an informed consent approved by theinstitutional review board of CHLA. Inclusion criteria were as

follows: documented gait ataxia, with or without opsoclonus,preceding the diagnosis of a neuroblastoma, in a child youngerthan 4 years at diagnosis. Children with opsoclonus-ataxia unre-lated to a neuroblastoma were not included, nor were children inwhom neurologic symptoms developed after treatment of a tu-mor.

Record Review

Medical records were reviewed and parents and physicianswere interviewed to obtain the following diagnostic information:age at first presentation of neurologic symptoms, presence of anypreexisting developmental delays before opsoclonus-ataxia, dura-tion of opsoclonus-ataxia before resection of neuroblastoma, loca-tion and size of tumor, histologic classification, and presence of N-myc amplification if available. Treatment information included

time of initiation, duration, and dosage of corticosteroids, ACTH,IVIg, chemotherapy, and other immunosuppressant modalities aswell as medications to control symptoms, such as antipsychotics,antidepressants, anxiolytics, clonidine, or sedatives. Imaging stud-ies of the original tumor were reviewed.

Standardized Testing

The specific tests used and age ranges are listed in Table 1. For4 patients who could not obtain valid basal scores on an age-appropriate test as a result of low level of function, a cognitivefunction test for a younger age group was used. Scores wereconverted by calculating mental age divided by chronological age,then converted to standard scores.48

Overall Cognitive, Adaptive, and BehavioralAssessment

All children were evaluated by a pediatric neuropsychologist toascertain their current developmental status and cognitive func-tion, using standardized, age-appropriate measures, in the child’spreferred language. Parent questionnaires were administered bythe research coordinator in the parent’s preferred language.

Motor Examination

Fine motor and gross motor skills were evaluated by an occu-pational therapist. All of the patients had initial evaluations todetermine their level of gross and fine motor skills. The standard-ized tests that were chosen had comparable gross motor scoresand examined balance, bilateral coordination, and ambulation.The fine motor categories were also comparable. Both tests in-cluded evaluation of eye-hand coordination, upper limb speed,and dexterity.

Speech and Language Assessment

Each child was evaluated by pediatric speech and languagepathologists to determine developmental levels in receptive andexpressive vocabulary and language, articulation/speech intelli-gibility, and oral motor function. A variety of different tests wereused, depending on age, level of functioning, and primary lan-guage. Test scores were converted to standard scores and com-

bined into 3 domains: receptive vocabulary, expressive vocabu-lary, and receptive language. Specific measures used for eachdomain are listed in Table 1. Two additional measures of expres-sive language (mean length of utterance in morphemes andBrown’s 14 grammatical morphemes) were used in all but 3 pa-tients, who were out of the age range.

Articulation patterns were obtained during all assessmenttasks. A formal test of articulation was attempted but could not be

completed in most cases as the majority of the children exhibitedlow tolerance for structured tasks and exhibited difficulties innaming or expressive vocabulary. Intelligibility assessments weremade by the same 2 speech/language pathologists and by paren-tal report.

Psychiatric Examination

The child psychiatrist interviewed the parents with their childand other family members who accompanied them. The parentswere asked to describe their child’s behavior before and coincidentwith the onset of neurologic symptoms, past or current psychiatricsymptoms, social skills and relationships, and academic function.The parents spontaneously answered open-ended questions basedon their memories. The child psychiatrist observed the child’s playand interactions and recorded a mental status examination foreach child.

Neurologic Examination

A standard neurologic examination was performed by the childneurologist. Specific attention was paid to eye movements, gaitpatterns, reflexes, muscle tone, and behavior during the visit.

Examination of Eye Movements

Patients were asked to observe the movement of red lights froma distance of 1 m on a light bar (Tracoustics Inc, Austin, TX).

Saccadic stimuli were flashing lights that alternated from center to10 degrees off center on the vertical and horizontal meridians at afrequency of 1 Hz, 1 ⁄ 2 Hz, and 1 ⁄ 3 Hz. Pendular pursuit stimuli werearrays of light-emitting diodes subtending 0.5 degrees and mov-ing at a sinusoidal velocity either horizontally or vertically over adistance subtending 10 degrees at a frequency of 0.65 Hz, 0.3 Hz,0.17 Hz, and 0.14 Hz. Optokinetic (OKN) pursuit stimuli weresimilar arrays of light-emitting diodes separated by a distance of 5degrees and moving at a constant velocity in one horizontal orvertical direction over a distance subtending 10 degrees. Thespeed of the OKN stimuli were 22, 11, 7, and 5 degrees/sec.

Eye movements of the patients were recorded on videotape.One observer without knowledge of the duration or severity of thepatients’ disease later reviewed this tape. Saccadic eye movementswere scored as “good” if they were accurate and maintained thefrequency of the stimuli, “fair” if there was mild to moderatedysmetria or delay, and “poor” if accuracy and latency wereunreliable. Pendular and OKN pursuit eye movements werescored as “good” if they smoothly maintained the frequency of thestimulus, “fair” if there was mild to moderate jerkiness; lesserquality pursuit was considered “poor.”

Reexamination Before and After IVIg

For each child currently receiving IVIg, a brief examination wasdone before and 2 to 3 days after at least 1 of the treatments. Testselection was determined by results of the detailed baseline ex-amination. Pre- and posttreatment examinations emphasized mo-tor skills, speech, and behavior, as these were the areas thatparents most frequently reported as sensitive to the treatment.

Motor Examination

Sample activities were chosen from the initial test for the pre-and post-IVIg testing sessions. The activities that were chosen forthe pre- and post-IVIg testing were ones that the patient could dopartially but did not fully meet the standards during the initialevaluation.

Because only certain portions of the test were used to score thepatient’s pre- and post-IVIg performance, a standardized scorewas not possible. Despite this, the items used were still scored ina standardized way on a 3-point scale as described by the testdevelopers. Each child’s pre-IVIg performance was measuredagainst his or her post-IVIg performance. The pre- and post-IVIgsessions were also videotaped for later review and comparison.The videotaped material made it possible to detect slight differ-ences in performance that were too subtle to register on thestandardized tests, such as changes in quality of performance,tremors, and time needed to complete tasks.

88 OPSOCLONUS-ATAXIA AND CHILDHOOD NEUROBLASTOMA. Provided by Indonesia:AAP Sponsored on April 5, 2011www.pediatrics.orgDownloaded from






Speech and Language

Post-IVIg assessments of a subgroup of the available studysample consisted primarily of observations of expressive language

based on a brief language sample and parental report and of changes in the child’s speech intelligibility per examiner judg-ment. Because of the short time intervals between pre- and post-treatment conditions, as well as the variable levels of participation

by the very youngest of the children, attempts to readministerformal assessment instruments were limited to the vocabularytests.

Statistical Analysis

To facilitate comparison across age ranges and testing instru-ments, we converted all speech and language and motor test

TABLE 1. Standardized Tests Used

Scores Generated Standardized Age Rangeof Test

Ages of Patients TestedOut of Standard Range,

With Calculated Scoring*

Cognitive functionBayley Scales of Infant Development II Mental Development Index 3 mo through 3 y 3.79 y, 3.94 y

(2 patients)Wechsler Preschool and Primary

Intelligence Scales, RevisedFull Scale IQ, Verbal IQ,

Performance IQ2 y 11 mo through 6 y 7.89 y, 8.65 y (2 patients)

Wechsler Intelligence Scales forChildren, Third Edition

Full Scale IQ, Verbal IQ,Performance IQ

6 through 16 y 11 mo None

Test of Nonverbal Intelligence 2 Nonverbal IQ Used for single patientage 10.27 y NoneAdaptive functioning

Vineland Adaptive Behavior Scales Total, Communication,Daily Living, SocialSkills

Birth through 18 y11 mo

None

Behavior problemsAchenbach Child Behavior Checklist Total, Internalizing,

Externalizing2 through 18 y None

Speech and languagePeabody Picture Vocabulary Test III Receptive Vocabulary 2.5 y through adult NonePeabody Picture Vocabulary Test—R Receptive Vocabulary 2.5 y through adult

(Spanish translation)Expressive Vocabulary Test Expressive Vocabulary 2.5 y through adult NoneMcArthur Communication

Development InventoryExpressive Vocabulary 8 through 30 mo Single patient age

3.95 yExpressive One Word Picture

Vocabulary Test—R

Expressive Vocabulary 2.5 to 12 y (Spanish

translation)Preschool Language Scale—3 Receptive Language Birth through 83 mo NoneClinical Evaluation of Language

Fundamentals—PreschoolReceptive Language 3 through 6.5 y None

Clinical Evaluation of LanguageFundamentals—3

Receptive Language 6 y through adult None

Test of Auditory Comprehension of Language—R

Receptive Language 3 to 10 y Single 10.3-y-old patientwith very limitedspeech

The Rossetti Infant-Toddler LanguageScale

Receptive Language Used for a singlepatient, age 1.9 y

Mean Length of Utterance inMorphemes

Standard score for age 21 mo to 6 y 6.28 through 8.65 y (3patients whose ageequivalent onvocabulary wereunder 6 y)

Brown’s 14 Morphemes Number of morphemesused

21 mo to 6 y 6.28 through 12.62 y(6 patients)

Motor functioningPeabody Developmental Motor Scales:

Fine Motor Scores Fine Motor Total,Grasping, Hand Use,Eye-Hand Coordination,Manual Dexterity

Birth through 5 y None

Gross Motor Scores Total, Balance, Non-Locomotor, Locomotor,Receipt/Propulsion

Same None

Bruininks-Oseretsky Test of MotorProficiency: Fine Motor Scores Fine Motor Total, Eye-

Hand Coordination,Manual Dexterity,Upper Limb

Coordination, ResponseSpeed

5 y through adult None

Gross Motor Scores Gross Motor Total,Balance, Runningspeed/agility, BilateralCoordination, Strength

Same None

* Score ϭ mental age/chronological age ϫ 100.






measurements to z scores, with a mean of 0 and standard devia-tion (SD) of 1. Statistical evaluation included descriptive statistics,Pearson correlation, and Kruskal-Wallis tests, all of which wereperformed using Minitab, Release 12 (Minitab Corp, State College,PA). Because of the descriptive nature of the study and smallsample size, only limited statistical analyses were performed.

RESULTS

Seventeen children were examined: 7 boys and 10girls. Twelve were primarily followed for treatmentof opsoclonus-ataxia at CHLA. This represented all

but 2 of children known to have opsoclonus-ataxiacaused by neuroblastoma seen at CHLA during theprevious 17 years. No family that could be contacteddeclined to participate. During the same time period,only a single child who was younger than 4 yearsand seen at onset of opsoclonus-ataxia at CHLA wasnot found to have a neuroblastoma. She was notincluded in the study group. Five additional childrenwhose source of neurologic care for opsoclonus-ataxia was elsewhere participated. In 1 of the 5 pa-tients, the original diagnosis and tumor resection had

been at CHLA, but all subsequent neurologic treat-ment was provided in another state. Age at exami-nation ranged from 1 to 13 years. The interval be-

tween onset of opsoclonus-ataxia and examinationranged from 1 week to 12 years.

Medical Review

Medical diagnostic and treatment information issummarized in Table 2. Fifteen of the 17 children hadopsoclonus-ataxia; 2 had ataxia alone, without disor-der of eye movements. Neuroblastoma was diag-nosed by a variety of imaging modalities, most fre-quently contrast-enhanced thin-cut computerizedtomography. Thirteen children had abdominal tu-mors, 3 had thoracic tumors, and 1 had a tumor inthe neck. All tumors were stage I or stage II, with

favorable histopathology in 16 of 17. When tested,n-MYC amplification was absent.

All children received immunosuppressant medica-tion, in various dosages, schedules, and combina-tions. All received oral corticosteroids, at least as theinitial treatment. In 6 children, ACTH (Acthar gel,Rhone-Poulenc Pharmaceuticals, Collegeville, PA)was substituted when initial response to oral medi-cation was suboptimal or when oral dosing was nottolerated. IVIg was used in 12 patients at a dose of 2g/kg over 1 or 2 days, ranging from a single courseto Ͼ40 treatments. Two children received cyclophos-phamide for symptoms of opsoclonus-ataxia, 1 im-

mediately after diagnosis and 1 a year later. Onechild received bolus methylprednisolone, cyclophos-phamide, methotrexate, and chlorambucil when hedeveloped steroid-resistant nephrotic syndrome sev-eral years after beginning treatment for opsoclonus-ataxia. Although the treatment was not directed athis neurologic condition, chronic ataxia improvedafter intensive immunosuppression. The interval be-tween the onset of opsoclonus-ataxia and diagnosisof neuroblastoma ranged from 2 days to 20 monthsin the group. The interval between the onset of symptoms and initiation of immunosuppressanttreatment ranged from Ͻ1 week to 29 months, withdelays of Ͼ1 year in 4 children. One additional pa- T

A B L E

2 .

M e d i c a l I n f o r m a t i o n : T u m

o r L o c a t i o n , S t a g e , a n d T r e a t m e n t

P a t i e n t

G e n d e r

L o c a t i o n

S t a g e

N - M y c

A g e a t

E x a m i n a t i o n

Y e a r s

R e c r u i t e d

F r o m

C H L A

V e r s u s

P a r e n t G r o u p

F i r s t

S y m p t o m s

( M o n t h )

A g e N B

D i a g n o s e d

( M o n t h )

I n i t i a l

S t e r o i d

( M o n t h )

A

C T H

(

Y / N )

O t h e r

C h e m o

( Y / N )

O n s e t

t o R X

( M o n t h )

R e s e c t i o n

t o R X

( M o n t h )

F i r s t

I V I g

( M o n t h )

I V I g #

C o u r s e s

O n s e t

t o I V I g

( M o n t h )

1

M

a r p

2

1

1 . 7 5

C H L A

1 8

1 8

1 8

N

N

0

0

1 8

1

0

2

M

a r p

2

1

1 . 9 0

C H L A

1 8

2 1

2 2

N

N

4

1

n d

0

n d

3

F

a a

2

1

2 . 7 4

P a r e n t

2 0

2 1

3 7

N

N

1 7

1 6

3 9

8

1 9

4

F

t

1

1

2 . 2 7

C H L A

1 7

1 8

1 8

N

N

1

0

1 8

4

0

5

M

a a

2

1

3 . 0 5

C H L A

1 8

2 5

2 8

N

N

1 0

3

n d

0

n d

6

F

a a

2

1

3 . 3 0

C H L A

2 1

2 1

2 1

Y

N

0

0

2 1

3

0

7

F

a a

2

1

3 . 6 8

C H L A

1 4

1 4

1 7

N

N

3

3

1 5

1 8

1

8

F

a r p

1

1

3 . 7 9

C H L A

1 9

2 3

2 4

Y

Y

5

1

2 5

4 1

2

9

F

a a

2

1

3 . 9 4

P a r e n t

1 7

1 8

4 7

N

N

3 0

2 9

n d

0

n d

1 0

F

t

1

1

4 . 2 6

P a r e n t

3 1

3 6

3 1

N

N

0

Ϫ 5

3 3

1 5

2

1 1

F

a a

2

n d

5 . 9 5

P a r e n t

1 9

2 8

1 9

Y

N

0

Ϫ 9

3 8

1 1

1 9

1 2

M

a a

2

1

6 . 2 8

C H L A

1 5

1 6

3 5

Y

N

2 0

1 9

n d

0

n d

1 3

F

n

1

n d

7 . 8 9

P a r e n t

1 1

1 3

1 3

Y

Y

2

0

5 0

2

3 9

1 4

F

a r p

2

n d

8 . 6 5

C H L A

1 3

1 6

3 3

N

N

2 0

1 7

3 5

8

2 2

1 5

M

a a

1

n d

1 0 . 2 7

C H L A

1 3

1 7

1 8

Y

N

5

1

1 9

1 2

6

1 6

M

a r p

1

n d

1 0 . 3 3

C H L A

1 8

3 8

4 0

N

N

2 2

2

n d

0

n d

1 7

M

a a

2

n d

1 2 . 6 2

C H L A

1 5

1 8

1 9

N

Y

4

1

1 9

1

4

a r p i n d i c a t e s a b d o m i n a l , r e t r o p e r i t o n e a l ; a a , a b d o m i n a l , a d r e n a l ; t , t h o r a c i c ; n , n

e c k ; n d , n o t d o n e .







tient was treated briefly with steroids, followed byIVIg during the year after diagnosis, then restartedcorticosteroid and IVIg treatment 9 years later. Threechildren began corticosteroids, ACTH, and/or IVIg

before the resection of the neuroblastoma; 1 had adelay of 5 months between recognition of opsoclo-nus-ataxia and beginning therapy and detection of the tumor, during which she received ACTH andoral steroids.

All children received physical and occupational

therapy for periods ranging from a few months toϾ

5years after diagnosis and with frequency of treat-ment sessions ranging from monthly to 4 times aweek. All but 2 of the children received interventionfor speech/language delay. The intensity of the in-tervention ranged from 1 to 4 sessions a week. For-mal hearing test results were not available, but therewere no reports or history of hearing impairment. Allchildren who were older than 3 years at the time of evaluation were enrolled in school programs withvariable levels of special intervention, ranging fromself-contained classes for the multiply handicappedto a few hours a week of resource specialist. Nonewere fully mainstreamed educationally.

Neurologic Findings

One child had a sixth nerve palsy, which devel-oped shortly after onset of opsoclonus-ataxia andhad not cleared as of the examination date, 3 monthsafter onset. One had epilepsy (patient 12). At the timeof the examination, 2 children were ambulatory onlywith a walker (patients 9 and 16). Patient 9 had

begun treatment with prednisone for the first timeshortly before the study visit, 30 months after onsetof opsoclonus-ataxia and resection of her neuroblas-toma. Subsequent to the study visit, while receivingoral prednisone, she continued to have rapid im-

provement in motor function and became fully ambulatory without assistance several months later. Pa-tient 16 was treated for approximately 1 year afterdiagnosis but had been off all immunosuppressivemedications for 7 years at the time of the study visit.

He was ambulatory only with a walker, had severedrooling, and had essentially no speech. Subsequentto the study visit, he restarted oral corticosteroidswith moderate improvement in gait, cessation of drooling, and minor improvements in speech. Pa-tient 1 was not yet fully ambulatory (crawling, pull-ing to stand, cruising) at the time of the study visit,3 months after diagnosis, while on oral corticoste-roids and IVIg. Six patients had mild to moderatelower extremity hyperreflexia (nonsustained clonus,

Babinski responses, and/or crossed adductor reflex-es), but none had significant clinical spasticity ortoe-walking.

Eye Movements

Six patients (patients 11–15 and 17) were coopera-tive for the videotaped examination of eye move-ments. All 6 patients had greater impairment of pur-suit eye movement than saccadic eye movements.Five patients had good horizontal saccades at alltarget speeds. None had good vertical saccades at 1Hz, and only 2 had good vertical saccades at slowertarget speeds. All patients had poor pursuit of thefastest pendular (0.65 Hz) and OKN (22 degrees/sec)targets. Vertical pursuit eye movement was usuallyworse and never better than horizontal eye move-ment. Deficiencies in eye movement usually im-proved with slower stimuli. Often an eye movementwas better in one direction than in the opposite di-rection, and this directional asymmetry was consis-tent for any given patient. There seemed to be arelationship between eye movement scores and IQ.However, the small number of patients with eyemovement examinations made statistical testing im-practical.

Cognitive Functioning

Scores on cognitive and adaptive measures aresummarized in Table 3. The neuropsychologic dataindicate significant deficits in overall cognitive de-velopment. Overall cognitive functioning was nor-mal in only 3 children, whose full scale IQ or Mental

TABLE 3. Cognitive and Adaptive Function

Patient Full Scale IQ/Mental

Development Index

VerbalIQ

PerformanceIQ

Test Used VinelandTotal

VinelandCommunication

VinelandDaily Living

VinelandSocial Skills

1 56 N/A N/A Bayley-II 74 80 77 802 54 N/A N/A Bayley-II 75 79 79 823 86 N/A N/A Bayley-II 76 91 73 84

4 50 N/A N/A Bayley-II 74 76 87 805 61 N/A N/A Bayley-II 75 79 81 766 94 92 97 WPPSI-R 89 98 98 897 91 93 89 WPPSI-R 71 84 83 778 58* N/A N/A Bayley-II* 56 61 60 629 53* N/A N/A Bayley-II* 69 74 90 78

10 70 71 75 WPPSI-R 70 79 78 7911 60 61 66 WPPSI-R 63 70 73 7312 65 55 80 WISC-III 58 61 64 6513 50* 53* 46* WPPSI-R* 56 51 60 7214 44* 35* 52* WPPSI-R* 41 41 41 5015 68 N/A N/A TONI-2 31 33 Ͻ20 4816 48 55 49 WISC-III 53 44 60 5817 46 54 47 WISC-III 52 57 49 62

N/A indicates not applicable.* Calculated using formula mental age/chronologic age ϫ 100, using out-of-age-range test.






Development Index was Ͻ1 SD below the norm. Allother children were functioning in the impairedrange, with IQ or MDI Ͼ2 SD below the mean (seeTable 3). Adaptive functioning was similarly im-paired; only 1 child’s score was within 1 SD of themean, and 6 were in the borderline range ( Ϫ1 to Ϫ2SD below normal). Increasing age was associatedwith increasing deficits in adaptive functioning.

The parents reported more behavioral difficultieson the adaptive functioning measures than on a be-

havioral measure, the Achenbach Child BehaviorChecklist,49 but both of these underestimated behav-ioral disturbances observed during the clinical inter-views. In general, parent responses on the ChildBehavior Checklist did not indicate significant emo-tional/behavioral problems. Only 4 patients had aTotal Score in the clinical range. Only 1 patient had aclinically significant level of Externalizing symp-toms, and no patient had Internalizing symptoms.On the Syndrome Scales, clinically significant eleva-tions were Attention Problems (5), Social Problems(4), and Thought Problems (3).

Motor SkillsDetailed motor testing results are listed in Table 4.In general, younger children were more likely toscore within the age-acceptable range than the olderchildren. Ten children were tested using the PeabodyDevelopmental Motor Scale.50 Seven were testedwith the Bruininks-Oseretsky Test of Motor Profi-ciency.51

The younger children, tested on the Peabody De-velopmental Motor Scale, scored lower on the finemotor tasks than on the gross motor tasks. Five of 10children performed poorly on eye-hand coordina-tion. In contrast, hand usage and grasping werewithin or above the acceptable ranges in 9 of 10patients. The gross motor performance of the chil-dren was slightly better than the fine motor scores.The locomotor, balance, and receipt and propulsioncategories were the most impaired, although only 3of 10 had z scores lower than Ϫ1.5. Nine scoredwithin the acceptable ranges for all of the gross mo-tor categories except for the nonlocomotor category,in which 10 children scored within the acceptableranges.

In contrast to the mostly acceptable scores for theyounger group, the older children had z scores thatwere mostly below the acceptable range for bothgross and fine motor abilities. Two children scored

above Ϫ1.5 on the balance and strength categories, but all of the other gross motor scores were below theacceptable range and even lower than Ϫ2. The finemotor performance of the older group was also be-low the acceptable age range, and only 1 child couldscore above Ϫ1.5 on response speed and visual mo-tor control.

Speech and Language

Detailed scores on speech and language measuresare in Table 5. The majority of the children partici-pated in the assessment tasks with moderate to max-imum support to maintain attention and interest. T

A B L E

4 .

D e t a i l e d T e s t S c o r e s , M o t o r T e s t i n g

P a t i e n t

T e s t

U s e d

F i n e

M o t o r

T o t a l

G r a s p i n g H

a n d

U s e

E y e - H a n d

C o o r d i n a t i o n

M a n u a l

D e x t e r i t y

U p p e r

L i m b


R e s p o n s e

S p e e d

G r o s s

m o t o r

T o t a l

B a l a n c e

N o n -

L o

c o m o t o r

L o c o m o t o r

R u n n i n g

S p e e d /

A g i l i t y

R e c e i p t /

P r o p u

l s i o n

B i l a t e r a l


S t r e n g t h

1

P D M S

0 . 0 5

1 . 5 6

0 . 6 1

Ϫ 0 . 0 2

0 . 3 6

0 . 8 1

1 . 0 4

1 . 1 8

0 . 9 5

Ϫ 0 . 0 5

2

P D M S

0 . 2 8

0 . 0 0

2 . 3 3

Ϫ 0 . 0 1

Ϫ 0 . 3 3

Ϫ 2 . 3 3

Ϫ 0 . 7 1

1 . 1 8

Ϫ 2 . 0 5

Ϫ 0 . 4 4

3

P D M S

0

0 . 0 0

2 . 3 3

Ϫ 0 . 0 1

0 . 1 8

Ϫ 0 . 0 5

Ϫ 0 . 2 8

1 . 1 3

0 . 0 5

0 . 0 5

4

P D M S

Ϫ 2 . 3 3

0 . 6 8

0 . 0 2

Ϫ 2 . 0 5

Ϫ 1 . 0 8

Ϫ 0 . 3 1

Ϫ 0 . 0 8

Ϫ 0 . 3 3

Ϫ 0 . 2 5

0 . 6 8

5

P D M S

2 . 0 5

2 . 3 3

Ϫ

0 . 9 2

Ϫ 0 . 9 2

0 . 3 9

2 . 3 3

0 . 1 3

0 . 5 0

0 . 1 0

1 . 0 8

6

P D M S

0 . 0 2

Ϫ 0 . 7 1

2 . 3 3

Ϫ 0 . 2 3

0 . 9 2

0 . 1 8

0 . 5 5

0 . 5 0

0 . 3 6

Ϫ 0 . 3 3

7

P D M S

Ϫ 1 . 1 3

Ϫ 1 . 2 5

2 . 3 3

Ϫ 1 . 0 0

Ϫ 0 . 1 0

Ϫ 2 . 0 5

Ϫ 1 . 7 5

0 . 4 1

Ϫ 1 . 3 4

0 . 1 5

8

P D M S

Ϫ 1 . 5 6

2 . 3 3

2 . 3 3

Ϫ 1 . 8 8

Ϫ 1 . 1 3

Ϫ 1 . 5 6

Ϫ 0 . 2 3

Ϫ 0 . 3 3

Ϫ 1 . 3 4

Ϫ 1 . 5 6

9

P D M S

Ϫ 2 . 0 2

Ϫ 2 . 0 5

Ϫ

1 . 0 4

Ϫ 1 . 8 8

Ϫ 1 . 7 5

Ϫ 2 . 0 5

Ϫ 2 . 3 3

Ϫ 0 . 3 3

Ϫ 2 . 3 3

Ϫ 2 . 0 5

1 0

P D M S

Ϫ 1 . 7 5

2 . 3 3

2 . 3 3

Ϫ 2 . 0 5

Ϫ 1 . 4 0

0 . 1 5

0 . 5 8

Ϫ 0 . 0 5

Ϫ 0 . 1 0

0 . 7 1

1 1

B O

Ϫ 4 . 4 8

Ϫ 2 . 8 0

Ϫ 2 . 8 0

Ϫ 2

. 8 0

Ϫ 1 . 8 0

Ϫ 2 . 6 0

Ϫ 2 . 8 0

Ϫ 2 . 8 0

Ϫ 2 . 4 0

Ϫ 2 . 4 0

1 2

B O

Ϫ 3 . 3 2

Ϫ 2 . 4 0

Ϫ 2 . 8 0

Ϫ 2

. 8 0

Ϫ 1 . 6 0

Ϫ 3 . 3 5

Ϫ 1 . 4 0

Ϫ 2 . 8 0

Ϫ 1 . 8 0

Ϫ 1 . 0 0

1 3

B O

Ϫ 3 . 5 4

Ϫ 2 . 8 0

Ϫ 2 . 8 0

Ϫ 2

. 8 0

Ϫ 2 . 4 0

Ϫ 2 . 5 5

Ϫ 2 . 8 0

Ϫ 2 . 8 0

Ϫ 2 . 8 0

Ϫ 1 . 8 0

1 4

B O

Ϫ 2 . 4 6

Ϫ 2 . 8 0

Ϫ 2 . 8 0

Ϫ 3

. 0 0

Ϫ 2 . 4 0

Ϫ 3 . 5 6

Ϫ 2 . 8 0

Ϫ 2 . 8 0

Ϫ 2 . 8 0

Ϫ 2 . 8 0

1 5

B O

Ϫ 3 . 5 6

Ϫ 2 . 8 0

Ϫ 2 . 8 0

Ϫ 3

. 0 0

Ϫ 2 . 8 0

Ϫ 5 . 2 4

Ϫ 2 . 8 0

Ϫ 2 . 8 0

Ϫ 2 . 8 0

Ϫ 2 . 8 0

1 6

B O

Ϫ 3 . 4 4

Ϫ 1 . 2 0

Ϫ 2 . 8 0

Ϫ 2

. 8 0

Ϫ 2 . 6 0

Ϫ 3 . 4 6

Ϫ 1 . 4 0

Ϫ 2 . 8 0

Ϫ 2 . 6 0

Ϫ 2 . 4 0

1 7

B O

Ϫ 3 . 3 2

Ϫ 2 . 8 0

Ϫ 2 . 8 0

Ϫ 2

. 6 0

Ϫ 0 . 6 0

Ϫ 3 . 4 5

Ϫ 2 . 6 0

Ϫ 2 . 8 0

Ϫ 2 . 2 0

Ϫ 1 . 4 0

D a t a a r e e x p r e s s e d a s z s c o r e s , w i t h m e a n ϭ

0 , S D ϭ

1 .

P D M S i n d i c a t e s P e a b o d y D e v e l o p m e n t a

l M o t o r S c a l e s ; B O , B r u i n i n k s - O s e r e t s k y

T e s t o f M o t o r P r o f i c i e n c y .







Inconsistency in attending and a tendency for agita-tion and impulsivity were observed and were judgedto be at levels greater than expected in a typicalpopulation of children of the same chronological age.In a number of cases, nonparticipation resulted inmissing scores. All test scores were converted to zscores for comparison across domains, with the ex-ception of Brown’s 14 morphemes, which is pre-sented as raw score. The language sampling mea-sure, mean length of utterance in morphemes (MLU-m), was converted to z scores by using the formula

provided by Miller.52 Language sample sizes ob-tained were of variable length and in most caseswere smaller than the standard 50 utterances onwhich the normative data are based. In comparisonto typically developing children, the low quantity of output noted for most of the study children repre-sented an area of significant clinical deficit.

Abnormalities in vocabulary and language devel-opment were pervasive; several children had suchsevere deficits that they were below the scorablerange. Only two z scores were Ͼ1 SD above themean, obtained by 2 different children on the expres-sive vocabulary measure. Comparing receptive and

expressive vocabulary performance in the 12 patientsfor whom scores on both tests were available, recep-tive vocabulary z score was better than expressivevocabulary in 7, expressive vocabulary was better in4, and scores were equal in 1.

The language sample analysis revealed the follow-ing results. The range of observed spontaneous ut-terance length in words for the ages for which this isa valid measure (12 months to 6 years) was 1 to 7.Generally, the observed and reported quantity of verbal production was low. In terms of syntacticcomplexity referenced by MLU-m, only 5 of 12 chil-dren (45%) in the age range had MLU-m within 1 SDof the mean for age. The number of different gram-

matical morphemes (Brown’s 14) ranged from 0 to 14morphemes in the sample. Within the standard agerange of the measure (27– 60 months), the range was0 to 8. Four children ranging between 30 months and53 months had 1 or none of the 14 grammaticalmorphemes in their language samples. In typicallydeveloping children, the acquisition of the 14 mor-phemes corresponds to language stages starting atapproximately 27 to 30 months and continuingthrough 47 to 60 months. Approximately 50% of themorphemes may be acquired by 43 to 46 months, and

100% of the morphemes are generally acquired by 60months of age. Of 7 patients tested at above 60months of age, when all 14 morphemes are expected,1 was nonverbal, 1 had all 14 morphemes, and theremaining patients had from 0 to 8 morphemes.

All of the children in the sample exhibited misar-ticulations involving single-sound errors, a numberof phonological patterns associated with young de-velopmental levels, and single-sound and soundcombination distortions associated with oral motorcoordination and planning difficulties. Difficulties incoordinating respiration and motor patterns neces-sary for speech were observed. Intelligibility judg-

ments by the 2 evaluators (unfamiliar individualsversus parents) ranged from 5% to 90%, with thehighest intelligibility being observed in a 51-month-old child. Parental reports of intelligibility were inthe 30% to 100% range, with the mode being 65%.

Oral motor examination to assess function of theoral mechanism revealed difficulties in volitionalcontrol of movements involving tongue elevationand lateralization. Inefficiency in imitating soundswas observed, often characterized by groping pos-tures and exhalation before articulation. Varying pat-terns of productions of sounds or words imitatedwere observed. Hypernasal vocal quality and incon-sistent volume modulation were also noted.

TABLE 5. Language and Speech

Patient Receptive Vocabulary Expressive Vocabulary Receptive Language MLU-m Brown’s 14Morphemes

Score Test Used Score Test Used Score Test Used

1 N/A N/A N/A Ϫ1.61 02 N/R N/R Ϫ1.53 Rossetti Ϫ1.77 03 0 PPVT-III ϩ1.13 EVT 1.00 PLS-3 Ϫ1.00 34 N/R Ϫ1.64 MacArthur N/R PLS-3 Ϫ1.64 35 Ϫ0.53 PPVT-III Ϫ2.13 EVT Ϫ0.10 PLS-3 Ϫ2.72 26 0 PPVT-III ϩ1.47 EVT N/R Ϫ0.13 37 Ϫ0.20 PPVT-III Ϫ0.60 EVT N/R Ϫ0.86 08 Ϫ2.33 PPVT-III Ϫ2.47 EVT Ϫ1.60 PLS-3 Ϫ2.91 0

9 Ϫ0.60 PPVT-III Ϫ1.40 EVT Ϫ1.13 PLS-3 Ϫ3.50 110 N/R N/R N/R ϩ0.70 811 Ϫ1.60 PPVT-III Ϫ0.80 EVT Ϫ1.70 PLS-3 Ϫ0.20 612 Ϫ2.40 PPVT-III Ϫ2.94 EVT Ϫ2.07 CELF-P Ϫ2.83 013 Ϫ1.90 PPVT-III Ϫ3.60 EVT ϽϪ3.00 CELF-3 ϽϪ3.50 214 ϽϪ4.0 PPVT-R ϽϪ3.33 EOWPVT-R N/R ϽϪ3.50 115 Ϫ1.60 PPVT-III N/V Ϫ2.33 TACL-R N/V N/V16 Ϫ2.20 PPVT-III Ϫ2.20 EVT Ϫ2.26 CELF-3 N/A 1417 Ϫ2.13 PPVT-III Ϫ2.40 EVT Ϫ3.90 CELF-3 N/A 8

Scores for language are expressed as standardized z scores, with mean ϭ 0, SD ϭ 1. Brown’s 14 morphemes are raw score, range 0 to 14.N/A, not available (due to age); N/R, not responsive; N/V, no verbal output, not scorable; PPVT-III, Peabody Picture VocabularyTest—III; PPVT-R, Peabody Picture Vocabulary Test—R; MacArthur, MacArthur Communicative Development Inventory: Words andSentences; EVT, Expressive Vocabulary Test; EOWPVT-R, Expressive One-Word Picture Vocabulary Test-R; Rossetti, The RossettiInfant-Toddler Language Scale; PLS-3, Preschool Language Scale-3; CELF-P, Clinical Evaluation of Language Fundamentals-Preschool;CELF-3, Clinical Evaluation of Language Fundamentals-3; TACLR-R, Test of Auditory Comprehension of Language-R.






Psychiatric Examination

All children had significant psychiatric and behav-ioral disturbances early in their course. The mostprominent symptom at the onset of illness, coinci-dent with onset of neurologic symptoms, was irrita-

bility, noted in 15 of the 17 patients. Most childrenwere remembered as inconsolable, difficult to soothe,or needing to be held all the time. Night terrors anddisturbed sleep were prominent, reported in 10 chil-dren. Severe tantrums with self-injurious behaviors,

such as head banging, were reported in 7 patientsand were distressing to the families. Mood swingsand poor affective regulation were sometimes recog-nized at onset (4 patients) but were associated laterwith steroid or ACTH therapy in 5 additional pa-tients and persisted into the chronic phase in 8 of 10patients. This pattern represented a marked changefrom premorbid personality or temperament, andalthough these psychiatric symptoms were often themost disruptive for the family, they were not thereason that medical attention was sought.

When the children were seen for psychiatric eval-uation, these symptoms had changed but not disap-peared. Irritable, dysphoric mood and mood swingswere still common problems (10 of 17), and tantrumsand oppositional and aggressive behavior were moreprominent than earlier (10 of 17). Older children hada history of difficulty with toilet training. Three of 10had persistent enuresis, and 1 had encopresis.

Treatment Effects

Six children received ACTH, either as primarytreatment or after they did not tolerate or respond tooral prednisone. Median IQ (Wechsler D. WechslerIntelligence Scales for Children—III. San Antonio,TX: The Psychological Corporation; 1991 or Wechsler

D. Wechsler Preschool and Primary Scales of Intelli-gence—Revised. San Antonio, TX: The PsychologicalCorporation; 1989) or MDI (Bayley [Bayley N. BayleyScales of Infant Development, Second Edition. SanAntonio, TX: The Psychological Corporation; 1993])of the group that received ACTH was not differentfrom the group that received only oral steroids (67without ACTH versus 73 with ACTH; P ϭ .366).Similarly, performance on the Vineland Scales (Spar-row SS, Balla DA, Cichetti DV. Survey Form Manual,Vineland Adaptive Behavior Scales. Circle Pines, MN:American Guidance Service, Inc; 1984) or the grossmotor measures did not differ in the group that

received ACTH versus only oral corticosteroids. Finemotor measures tended to be worse in the group thatreceived ACTH, approaching but not reaching statis-tical significance ( z ϭ Ϫ3.4 versus Ϫ1.75; P ϭ .078).Three children who received chemotherapy werecompared with 14 who did not. No measure showedany advantage for the small group that receivedchemotherapeutic agents, with trends toward lowerfunctioning in the small group that received them.Median IQ or MDI for the 3 children who receivedother chemotherapy or immune modulating agentswas 63, compared with 71 for the other 14 children(P ϭ .101). The effect of total use of IVIg was exam-ined by dividing the patients into 3 groups of ap-

proximately equal size. Five received no IVIg, 7 re-ceived 1 to 8 courses, and 5 received 10 to 41 courses.Age at evaluation of the 3 groups was similar (3.7years for the group with no IVIg, 3.3 years for thegroup with 1 to 8 courses, and 4.3 years for the groupwith 10 or more courses). There were no significantdifferences in cognitive, adaptive, or motor function-ing among treatment groups. However, the groupwith no IVIg contained a disproportionate number of children with long delays between onset of symp-

toms and treatment.There was a striking relationship between age atevaluation and level of function, with older childrenexhibiting significantly more deficits in overall adap-tive behavior and gross and fine motor function thanyounger children. Figure 1 plots age versus IQ orMDI and Vineland total score. Figure 2 plots ageversus z-transformed gross and fine motor function.Pearson correlation of age with Vineland score wasϪ0.812 (P Ͻ .0001), with fine motor function wasϪ0.711 (P Ͻ .001), and with gross motor function wasϪ0.753 (P Ͻ .0001). Correlation of age and IQ/MDIwas Ϫ0.398 (NS).

Median delay from onset to treatment was 4months, with a range of 0 to 30 months. Timing of treatment was unrelated to cognitive, adaptive, ormotor scores and was unrelated to age at evaluation.Patients with very early treatment (within 4 weeks of onset) were compared with the group as a whole.There were no significant differences in any out-come.

Acute Effects of IVIg

Five children (ages 16 months to 55 months) wereexamined immediately before and 1 to 3 days after atreatment with IVIg. Two grams of IVIg per kilogramof body weight were infused over 6 to 8 hours in 4

children. One gram per kilogram daily for 2 doses on2 successive days was infused in 1 child. Premedica-tion with acetaminophen and diphenhydramine wasprovided to prevent headache, nausea, or rash.

Speech and Language

An overall increase in the quantity of verbalizationwas observed and reported after IVIg treatment. Thisobservation by the examiners may have been par-tially attributable to the posttreatment evaluation’s

being the third visit of the child. However, all par-ents reported increased posttreatment verbal interac-tion. There were also reports of increased attentive-

ness to verbal input, as well as increased length of utterance, albeit sometimes in the form of jargon. Theexaminers observed increased attention to task anddecreased agitation during structured activities. Noconsistent parental reports regarding changes inspeech intelligibility were obtained. There was someperception of increased intelligibility by the examin-ers, but this may have been attributable to familiarityfactors, as well as increased verbalization in context.

Motor Skills

Balance, locomotor, and receipt and propulsionskills were the gross motor categories that were af-fected by the IVIg treatments. Two children were







able to climb stairs with an alternate pattern, placingonly 1 foot on each stair after the IVIg session, butcould not do this task a few days earlier during theevaluation before the IVIg. A younger child who wascrawling up the stairs during the pre-IVIg evaluation

was able to climb the stairs, placing both feet on eachstair while holding onto the rail after her IVIg ses-sion. One boy who could not stand on 1 leg beforethe IVIg session was able to do a single limb stancefor 5 seconds. Three of the children tested were ableto bounce a tennis ball against a wall after the IVIgsession but could not do this task before the IVIg.Eye-hand coordination and hand usage were the finemotor categories that improved after the IVIg treat-ments. The children were able to copy basic shapesand cut with scissors after the IVIg sessions withmore skills and precision than before the treatment.

DISCUSSION

It has been repeatedly observed that paraneoplas-tic opsoclonus-ataxia caused by neuroblastoma is achronic neurologic condition associated with signif-icant morbidity attributable to long-term develop-mental deficits. This study confirms previous re-ports. There was a striking relationship of age attesting to degree of deficit in this group. The young-est, most recently treated children were functioningat nearly normal levels, whereas deficits were muchmore striking in the oldest children. Unfortunately,we cannot address a major question regarding thisrelationship: is the apparent worsening of develop-mental level with age attributable to plateauing of

development (ie, a static encephalopathy, presentfully at onset of symptoms, that becomes more ap-parent with increasing demands on the child), or isthere a progressive process that worsens with eachrelapse of neurologic symptoms? This will require

prospective, longitudinal studies to evaluate.We hypothesized that earlier and more intensive

treatment would be associated with better outcome.Findings did not support this hypothesis. However,we cannot determine whether the children who weretreated more aggressively received such treatment

because of more severe symptoms or more frequentrelapses, which might be expected to have a poorerneurologic prognosis.

Only 3 children in this group were treated withchemotherapy, at varying times after the onset of symptoms. One child received cyclophosphamideimmediately before surgery and for 6 postoperative

courses, specifically for opsoclonus-ataxia. A secondchild received 1 year of cyclophosphamide after sherepeatedly relapsed on ACTH and IVIg, with severepsychiatric symptoms and behavioral problems inaddition to ataxia. A third child received multiplecourses of methotrexate, cyclophosphamide, andchlorambucil after he developed an apparently un-related autoimmune condition 3 years after diagno-sis of opsoclonus myoclonus syndrome. In contrastto the experience of Russo et al,15 these 3 childrenwere functioning at low levels, and the additionalchemotherapy did not seem to provide neurologic ordevelopmental protection.

We also did not find a selective advantage of

Fig 1. The relationship of adaptive functioning (measured by total score on the Vineland Adaptive Behavior Scales), cognitive function-ing (Full Scale IQ on Wechsler Intelligence Scales for Children III or Wechsler Preschool and Primary Scales of Intelligence —Revised, orMDI on Bayley II) versus age is plotted.






ACTH over oral corticosteroids, although dosage,length of course, and schedules varied too widely to

be certain. This contrasts with the experience re-ported by Hammer et al,31 who found oral steroids to

be useless and noted responses only to ACTH. In ourgroup, however, several children who had not re-

sponded to high-dose oral corticosteroids (generally2 mg/kg/d of oral prednisone) responded well tointramuscular ACTH gel, giving some indicationthat it may provide superior results in some in-stances. We cannot make any firm conclusion abouteither the possible role of chemotherapy or the ben-efit of ACTH, given the descriptive nature of thestudy. This will require randomized prospective tri-als. Nevertheless, a great deal of attention has beengiven to findings of Russo et al,15 attributing a long-term advantage in developmental outcome to use of cyclophosphamide. In view of our results and thoseof Rudnick et al,13 we caution that a randomized

prospective trial of cyclophosphamide is imperative before it is widely introduced for opsoclonus-ataxia.Treatment with IVIg seemed to confer at least

short-term benefits in all 5 children who were exam-ined both before and 2 to 3 days after an infusion.There were clear gains in motor function; both qual-itative differences in motor abilities and specifictasks that could not be performed before treatmentwere completed posttreatment. Speech quality im-proved, but, more striking, ability to cooperate withthe evaluation and overall mood was observed toimprove post-IVIg. This study does not provide ad-equate measures of the long-term efficacy of IVIg ineither maintaining improvement or decreasing long-

term deficits. We did not find that patients who weretreated with many courses of IVIg benefited in long-term comparison to those who were treated withonly a few course or none, but, again, this may beconfounded by varying disease severity, which dic-tated more treatment in severely affected children.

Although opsoclonus eventually disappeared inall children, abnormalities of pursuit eye movementsand, to a lesser extent, saccadic eye movements per-sisted in all patients who could cooperate with for-mal testing.

Considering the disordered features observed inthe communication of the young children (youngerthan 7 years), morphologic and syntactic compo-nents of language were the most striking. Even inchildren with relatively longer sentences, the mor-phologic markers typically expected in the 26- to48-month age range were frequently absent. The sig-nificantly depressed expressive vocabulary levels,

both in quantity and variety, may be a function of overall cognitive developmental levels, which werelow in nearly all of the patients.

Speech characteristics responsible for the ex-tremely low levels of intelligibility were also signif-icant for being directly associated with oral, phona-tory, and respiratory incoordination. Although somedevelopmentally recognizable misarticulation/pho-nological patterns were noted, the very poor intelli-gibility noted is considered to be motor based.

The significant expressive language impairmentobserved, combined with the severe speech intelligi-

bility problems, results in significantly impairedcommunicative competence. Even the children who

Fig 2. The relationship of fine motor and gross motor versus age is plotted. Motor scores are converted to z scores, with mean of 0 andSD of 1.







had received early and intensive speech and lan-guage intervention exhibited persistent speech andlanguage disorder.

The psychiatric and behavioral aspects of opsoclo-nus-ataxia deserve special mention, as they have not

been emphasized in previous reports and contributesubstantially to overall morbidity. Irritability has

been noted as a presenting symptom of opsoclonus-ataxia syndrome, regardless of cause.25 Irritabilitywas recognized as a presenting symptom in earlier

studies of opsoclonus-ataxia.19,31

Anecdotal reportsdocumented the reduction in the quality of life inpatients with opsoclonus-ataxia, and children wererecognized to have persistent disruption of social-emotional stability.53 Extensive psychosocial evalua-tions of these patients noted “emotional dysregula-tion” with heightened emotional lability and

behavioral regression.53

The concept of “dysmetria or ataxia of thought or behavior” has been introduced in newer studies of the role of the cerebellum and its relationship to thecerebrum, particularly the frontal cortex.54 Dysfunc-tion in cerebellar-cerebral circuits may underlie theaffective dysregulation and behavioral disturbancesseen in opsoclonus myoclonus syndrome, as dys-function of other cerebellar circuits underlie themovement symptoms characteristic of the disorder.

Additional investigation into the treatment of op-soclonus-ataxia is clearly warranted. Given the dis-parate findings of this and other series of patientswith neuroblastoma and opsoclonus-ataxia regard-ing possible treatment effects (ACTH versus oralsteroids, IVIg, and the effects of chemotherapy), arandomized study comparing various treatment mo-dalities is needed.

ACKNOWLEDGMENTS

This study was made possible by a grant from Aventis-Behring(formerly Centeon). We thank the parents and children who par-ticipated, and Sandra Greenberg, organizer of the opsoclonus-ataxia parent group web site, for help in contacting interestedfamilies.

We thank Jane Taveras for statistical assistance.

REFERENCES

1. Dropcho EJ. Neurologic paraneoplastic syndromes. J Neurol Sci. 1998;

153:264–278

2. Liblau R, Benyahia B, Delattre JY. The pathophysiology of paraneoplas-

tic neurological syndromes. Ann Med Intern (Paris). 1998;149:512–520

3. Posner JB, Dalmau JO. Paraneoplastic syndromes affecting the central

nervous system. Annu Rev Med. 1997;48:157–166

4. Dalmau J, Gultekin HS, Posner JB. Paraneoplastic neurologic

syndromes: pathogenesis and physiopathology. Brain Pathol. 1999;9:

275–284

5. Posner JB, Dalmau JO. Paraneoplastic syndromes of the nervous sys-

tem. Clin Chem Lab Med. 2000;38:117–122

6. Boltshauser E, Deonna T, Hirt HR. Myoclonic encephalopathy of infants

or “dancing eyes syndrome.” Report of 7 cases with long-term fol-

low-up and review of the literature (cases with and without neuroblas-

toma). Helv Paediatr Acta. 1979;34:119 –133

7. Bray PF, Ziter FA, Lahey ME, Myers GG. The coincidence of neuroblas-

toma and acute cerebellar encephalopathy. Trans Am Neurol Assoc.

1969;94:106–109

8. Brissaud HE, Beauvais P. Opsoclonus and neuroblastoma. N Engl J Med.

1969;280:1242

9. Harel S, Yurgenson U, Rechavi G, Burstein Y, Spirer Z. Cerebellar ataxia

and opsoclonus as the initial manifestations of myoclonic encephalop-

athy associated with neuroblastoma. Child Nerv Syst. 1987;3:245–247

10. Moe PG, Nellhaus G. Infantile polymyoclonia-opsoclonus syndrome

and neural crest tumors. Neurology. 1970;20:756–764

11. Sandok BA, Kranz H. Opsoclonus as the initial manifestation of occult

neuroblastoma. Arch Ophthalmol. 1971;86:235–236

12. Pranzatelli MR. The neurobiology of the opsoclonus-myoclonus syn-

drome. Clin Neuropharmacol. 1992;15:186 –228

13. Rudnick E, Khakoo Y, Antunes NL, et al. Opsoclonus-myoclonus-ataxia

syndrome in neuroblastoma: clinical outcome and antineuronal anti-

bodies—a report from the Children’s Cancer Group Study. Med Pediatr

Oncol. 2001;36:612– 622

14. Koh PS, Raffensperger JG, Berry S, et al. Long-term outcome in children

with opsoclonus-myoclonus and ataxia and coincident neuroblastoma.

J Pediatr. 1994;125:712–716

15. Russo C, Cohn SL, Petruzzi MJ, de Alarcon PA. Long-term neurologic

outcome in children with opsoclonus-myoclonus associated withneuroblastoma: a report from the Pediatric Oncology Group. Med Pedi-

atr Oncol. 1997;28:284 –288

16. Altman AJ, Baehner RL. Favorable prognosis for survival in children

with coincident opso-myoclonus and neuroblastoma. Cancer. 1976;37:

846 – 852

17. Telander RL, Smithson WA, Groover RV. Clinical outcome in children

with acute cerebellar encephalopathy and neuroblastoma. J Pediatr Surg.

1989;24:11–14

18. Janss A, Sladky J, Chatten J, Johnson J. Opsoclonus/myoclonus: para-

neoplastic syndrome of neuroblastoma. Med Pediatr Oncol. 1996;26:

272–279

19. Mitchell WG, Snodgrass RS. Opsoclonus/ataxia in childhood neural

crest tumors: a chronic neurologic syndrome. J Child Neurol. 1990;5:

153–158

20. Hiyama E, Yokoyama T, Ichikawa T, et al. Poor outcome in patients

with advanced stage neuroblastoma and coincident opsomyoclonussyndrome. Cancer. 1994;74:1821–1826

21. Okano HJ, Park WY, Corradi JP, Darnell RB. The cytoplasmic Pur-

kinje onconeural antigen cdr2 down-regulates c-Myc function: im-

plications for neuronal and tumor cell survival. Genes Dev. 1999;13:

2087–2097

22. Cao Y, Abbas J, Wu X, Dooley J, van Amburg AL. Anti-Yo positive

paraneoplastic cerebellar degeneration associated with ovarian

carcinoma: case report and review of the literature. Gynecol Oncol.

1999;75:178–183

23. Mason WP, Graus F, Lang B, et al. Small-cell lung cancer, paraneoplastic

cerebellar degeneration and the Lambert-Eaton myasthenic syndrome.

Brain. 1997;120:1279–1300

24. Albert ML, Austin LM, Darnell RB. Detection and treatment of activated

T cells in the cerebrospinal fluid of patients with paraneoplastic cere-

bellar degeneration. Ann Neurol. 2000;47:9 –17

25. Connolly AM, Pestronk A, Mehta S, Pranzatelli MR 3rd, Noetzel MJ.Serum autoantibodies in childhood opsoclonus-myoclonus syndrome:

an analysis of antigenic targets in neural tissues. J Pediatr. 1997;130:

878 – 884

26. Connolly AM, Mitchell WG, Pestronk A, et al. Serum IgM autoantibod-

ies in patients with paraneoplastic and idiopathic opsoclonus myoclo-

nus syndrome (OMS) recognize neurofilament and histone in neural

preparations. Ann Neurol. 1999;46:536 –537

27. Antunes NL, Khakoo Y, Matthay KK, et al. Antineuronal antibodies in

patients with neuroblastoma and paraneoplastic opsoclonus-myoclo-

nus. J Pediatr Hematol Oncol. 2000;22:315–320

28. Fisher PG, Wechsler DS, Singer HS. Anti-Hu antibody in a neuroblas-

toma-associated paraneoplastic syndrome. Pediatr Neurol. 1994;10:

309 –312

29. Salmaggi A, Nemni R, Pozzi A, et al. Antineuronal antibody in a patient

with neuroblastoma and opsoclonus-myoclonus-ataxia: a case report.

Tumori. 1997;83:709 –711

30. Scaravilli F, An SF, Groves M, Thom M. The neuropathology of para-

neoplastic syndromes. Brain Pathol. 1999;9:251–260

31. Hammer MS, Larsen MB, Stack CV. Outcome of children with opsoc-

lonus-myoclonus regardless of etiology. Pediatr Neurol. 1995;13:21–24

32. Pohl KR, Pritchard J, Wilson J. Neurological sequelae of the dancing eye

syndrome. Eur J Pediatr. 1996;155:237–244

33. Borgna-Pignatti C, Balter R, Marradi P, Colamaria V. Treatment with

intravenously administered immunoglobulins of the neuroblastoma-

associated opsoclonus-myoclonus. J Pediatr. 1996;129:179 –180

34. Eiris J, del Rio M, Castro-Gago M. Immune globulin G for treatment of

opsoclonus-polymyoclonus syndrome. J Pediatr. 1996;129:175

35. Savino PJ, Glaser JS. Opsoclonus. Pattern of regression in a child with

neuroblastoma. Br J Ophthalmol. 1975;59:696 – 698

36. Schuler D, Koos R, Krauze I, Peter A. Paraneoplastic syndrome in

childhood. Acta Paediatr Acad Sci Hung. 1977;18:31– 40

37. Zafeiriou DI, Kontopoulos EE, Katzos GS, Gombakis NP, Kanakoudi






FG. Single dose immunoglobulin therapy for childhood Guillain-Barre

syndrome. Brain Dev. 1997;19:323–325

38. Blaes F, Strittmatter M, Merkelbach S, et al. Intravenous immunoglobu-

lins in the therapy of paraneoplastic neurological disorders. J Neurol.

1999;246:299–303

39. Uchuya M, Graus F, Vega F, Rene R, Delattre JY. Intravenous immu-

noglobulin treatment in paraneoplastic neurological syndromes with

antineuronal autoantibodies. J Neurol Neurosurg Psychiatry. 1996;60:

388 –392

40. Finsterer J, Grass R, Stollberger C, Mamoli B. Immunoglobulins in

acute, parainfectious, disseminated encephalo-myelitis. Clin Neurophar-

macol. 1998;21:258 –261

41. Dalakas MC. Intravenous immune globulin therapy for neurologic dis-

eases. Ann Intern Med. 1997;126:721–73042. Stangel M, Hartung HP, Marx P, Gold R. Intravenous immunoglobulin

treatment of neurological autoimmune diseases. J Neurol Sci. 1998;153:

203–214

43. Frederico P, Zochodne DW, Hahn AF, Brown WF, Feasby TE. Multifo-

cal motor neuropathy improved by IVIg. Randomized, double-blind,

placebo-controlled study. Neurology. 2000;55:1256–1262

44. Petruzzi MJ, de Alarcon PA. Neuroblastoma-associated opsoclonus-

myoclonus treated with intravenously administered immune globulin

G. J Pediatr. 1995;127:328 –329

45. Pless M, Ronthal M. Treatment of opsoclonus-myoclonus with high-

dose intravenous immunoglobulin. Neurology. 1996;46:583–584

46. Sugie H, Sugie Y, Akimoto H, Endo K, Shirai M, Ito M. High-dose i.v.

human immunoglobulin in a case with infantile opsoclonus polymyo-

clonia syndrome. Acta Paediatr. 1992;81:371–372

47. Veneselli E, Conte M, Biancheri R, Acquaviva A, De Bernardi B. Effect

of steroid and high-dose immunoglobulin therapy on opsoclonus-

myoclonus syndrome occurring in neuroblastoma. Med Pediatr Oncol.

1998;30:15–17

48. Sattler JM. Assessment of Children. San Diego, CA: Jerome M. Sattler

Publisher; 1992:21–22

49. Achenbach TM, Edelbrock CS. Child Behavior Checklist. Burlington, VT:

University of Vermont, Department of Psychiatry; 1986

50. Folio MR, Felwell RR. Peabody Developmental Motor Scale. Allen, TX:DLM Teaching Resources; 1983

51. Bruininks RH. Bruininks-Oseretsky Test of Motor Proficiency. Circle Pines,

MN: American Guidance Service, Inc; 1978

52. Miller JF. Assessing Language Production in Children. Baltimore, MD:

University Park Press; 1981

53. Papero PH, Pranzatelli MR, Margolis LJ, Tate E, Wilson LA, Glass P.

Neurobehavioral and psychosocial functioning of children with opsoc-

lonus-myoclonus syndrome. Dev Med Child Neurol. 1995;37:915–932

54. Schmahmann JD, Sherman JC. The cerebellar cognitive affective syn-

drome. Brain. 1998;121:561–579

THE LETHAL MYTH OF MASCULINITY

“The Marlboro Man emanated in 1954 from the minds of Chicago admen LeoBurnett and John Benson, who were trying to devise a more macho pitch for PhillipMorris’ filter tip cigarette and agreed that the ‘most masculine figure in America’was the cowboy. In the next 40 years the smoking cowboy traveled the world (and2 actors who played him died of lung cancer).”

Nugent W. Into the West. New York, NY: Knopf; 1999

Submitted by Student







2002;109;86-98Pediatrics

Elvira Burger, Susan B. Turkel, Mark S. Borchert, Susan Hollar and Sonia PadillaWendy G. Mitchell, Yolanda Davalos-Gonzalez, Virdette L. Brumm, Sonia K. Aller,

Neurologic SequelaeOpsoclonus-Ataxia Caused by Childhood Neuroblastoma: Developmental and

& ServicesUpdated Information

http://www.pediatrics.org/cgi/content/full/109/1/86including high-resolution figures, can be found at:

Citations http://www.pediatrics.org/cgi/content/full/109/1/86#otherarticles

This article has been cited by 13 HighWire-hosted articles:

Subspecialty Collections

http://www.pediatrics.org/cgi/collection/tumors Tumors

following collection(s):This article, along with others on similar topics, appears in the

Permissions & Licensing

http://www.pediatrics.org/misc/Permissions.shtmltables) or in its entirety can be found online at:Information about reproducing this article in parts (figures,

Reprints http://www.pediatrics.org/misc/reprints.shtmlInformation about ordering reprints can be found online:




http://www.pediatrics.org/cgi/content/full/109/1/86#otherarticles



http://www.pediatrics.org/cgi/collection/tumors




http://www.pediatrics.org/misc/Permissions.shtml



http://www.pediatrics.org/misc/reprints.shtml








neuroblastoma and atxia

Documents