Opsoclonus-Ataxia Caused by Childhood Neuroblastoma: Developmental and Neurologic Sequelae Wendy G. Mitchell, Yolanda Davalos-Gonzalez, Virdette L.

Brumm, Sonia K. Aller, Elvira Burger, Susan B. Turkel, Mark S. Borchert, Susan Hollar and Sonia Padilla Pediatrics 2002;109;86-98

The online version of this article, along with updated information and services, is located on the World Wide Web at: http://www.pediatrics.org/cgi/content/full/109/1/86

PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly publication, it has been published continuously since 1948. PEDIATRICS is owned, published, and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village, Illinois, 60007. Copyright 2002 by the American Academy of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.

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

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, MD; Mark S. Borchert, MD#; Susan Hollar, MS, CCC; and Sonia Padilla, MA*
ABSTRACT. Objective. Opsoclonus-ataxia, also called dancing eye syndrome, is a serious neurologic condition that is often a paraneoplastic manifestation of occult neuroblastoma in early childhood. Despite resection of tumor and immunosuppressive therapy, outcome generally includes significant developmental and behavioral sequelae. There is controversy about how treatment alters outcome. The goals of this study were to understand the ongoing neurologic and developmental deficits of children who are treated for opsoclonus-ataxia with associated neuroblastoma; to relate treatment history to outcome; and to quantify objectively the acute changes in motor function, speech, mood, and behavior related to intravenous immunoglobulin (IVIg) treatment. Methods. Patients were children with opsoclonusataxia caused by neuroblastoma, regardless of interval since diagnosis. Records were reviewed, and children underwent comprehensive evaluations, including neurologic examination and tests of cognitive and adaptive function, speech and language, and fine and gross motor abilities. Psychiatric interview and questionnaires were used to assess current and previous behavior. In 6 children, a videotaped standardized examination of eye movements was performed. Additional examinations were 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 neuroblastoma resected 3 months to 11 years previously. None received any other treatment for the tumor. All but 1 had received at least 1 year of either oral corticosteroids or corticotropin (ACTH); 12 had received 1 or more courses of IVIg, 2 g/kg. Three had received other immunosuppressive treatment, including cyclophosphamide. Cognitive development and adaptive behavior were delayed or abnormal in nearly all children. Expressive language was more impaired than receptive language. Speech was impaired, including both intelligibility and overall output. Fine and gross motor abilities were impaired. Increased age was strikingly associated with lower scores in all areas. Behavioral problems early in the course included severe
From the *Neurology Division, University Affiliated Program, Division of Physical and Occupational Therapy, Psychiatry Program, and Ophthalmology Division, Childrens Hospital Los Angeles, Los Angeles, California; Departments of #Neurology, **Pediatrics, Psychiatry, and Ophthalmology, Keck School of Medicine, University of Southern California, Los Angeles, 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: wmitchell@chla.usc.edu PEDIATRICS (ISSN 0031 4005). Copyright 2002 by the American Academy of Pediatrics.

irritability and inconsolability in all; later, oppositional behavior and sleep disorders were reported. Opsoclonus abated in all, but abnormalities in pursuit eye movements were found in all 6 children cooperative with standardized examination. Outcome did not differ in children who were treated with ACTH versus oral steroids. Three children who had received cyclophosphamide fared poorly. Immediate versus delayed treatment was not associated with better outcome. IVIg improved both gross and fine motor and speech function acutely, but we could not confirm long-term benefit of IVIg. Total number of courses of IVIg was not associated with outcome. Conclusions. Opsoclonus-ataxia caused by neuroblastoma causes substantial developmental sequelae that are not adequately prevented by current treatment. The increased deficits in older children raise concern that this represents a progressive encephalopathy rather than a time-limited single insult. Although the study is crosssectional and neither randomized nor blinded, we were unable to confirm a purported advantage of either ACTH over corticosteroids or of cyclophosphamide. A randomized study is needed but is difficult for this rare condition. Pediatrics 2002;109:86 98; opsoclonus, ataxia, neuroblastoma, paraneoplastic disorders, developmental sequelae.
ABBREVIATIONS. IVIg, intravenous immunoglobulin; CHLA, Childrens Hospital Los Angeles; OKN, optokinetic; SD, standard deviation; MLU-m, mean length of utterance in morphemes.

araneoplastic neurologic syndromes are neuronal degenerative disorders that occur as an indirect result of malignancies.15 In adults, paraneoplastic cerebellar degeneration, with or without 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 paraneoplastic neurologic syndrome related to occult, low-grade neuroblastomas.6 11 The syndrome occurs almost exclusively in young children (6 months to 3 years of age) and is generally thought to occur in approximately 2% to 3% of children with neuroblastoma.12,13 The syndrome occurs primarily in children with localized, small neuroblastomas (stage I or II), with favorable histologic features, and without amplification of the N-myc gene.14,15 Survival in children with neuroblastoma presenting with opsoclonus-ataxia is substantially higher than in children with neuroblastoma without neurologic presentation.7,9,16 19 A recent review of children registered in

86

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

the Childrens Cancer Group Neuroblastoma Registry found 100% survival in children with stage I or II neuroblastoma presenting with opsoclonus-ataxia.13 Children who have higher stage (metastatic or recurrent) neuroblastoma with N-myc amplification and who have opsoclonus-ataxia, either as a presenting feature or later in their course, do not seem to have this good prognosis, although the number of cases reported is very small.20 It is commonly assumed that paraneoplastic neurologic syndromes occur when there is antigenic similarity between tumor cells and neurons. Although antibodies may provide some biological control of the tumor, they promote progressive damage to target neurons.21 In several adult syndromes of paraneoplastic cerebellar degeneration, identifiable antibodies are associated with the syndrome (anti-Yo in gynecologic malignancies, anti-Hu in small cell carcinoma of the lung).22,23 Recently, Albert et al24 reported that cytotoxic T-lymphocytes were present in blood and cerebrospinal fluid of adults with paraneoplastic cerebellar degeneration. For one syndrome in which the tumor antigen has been cloned, the cytotoxic T-cells were directed against this antigen. In children with opsoclonus-ataxia, antineuronal antibodies are less consistently found and their associated antigen is still uncertain, although Connolly et al25,26 recently reported finding antibodies to various components of cerebellar neurons in at least some children with opsoclonus-ataxia. In a study of banked serum samples from children with neuroblastoma with and without opsoclonus-ataxia, antineuronal antibodies were found more frequently in the children with opsoclonus-ataxia but were found in some children with neuroblastoma alone and were highly variable.27 Anti-Hu and Anti-Yo antibodies have been found only rarely.28,29 T-cell activation in childhood opsoclonus-ataxia has not yet been demonstrated but is suspected to be involved in the mechanism of both tumor suppression and neuronal damage. Adult paraneoplastic cerebellar degeneration is neuropathologically characterized by severe loss of Purkinje cells.30 In children with opsoclonus-ataxia and neuroblastoma, tumors are small and indolent and rarely metastasize or recur after resection, so neuropathological examinations are not available. However, it is generally assumed that similar mechanisms apply to pediatric and adult paraneoplastic syndromes. Despite treatment with corticosteroids, corticotropin (ACTH), intravenous globulin (intravenous immunoglobulin [IVIg]), and a variety of other immunosuppressant medications, most children who have had paraneoplastic opsoclonus-ataxia have significant residual neurologic and developmental deficits.1315,19,31 Ongoing treatment, usually with corticosteroids, continuously or intermittently, is often required to maintain function such as ambulation, speech, and continued developmental gains.19,31 Children with nonparaneoplastic (presumably postinfectious) opsoclonus-ataxia are also noted to

have severe neurologic sequelae despite prolonged treatment with corticosteroids and/or ACTH.32 Other immune-modulating treatments including intravenous globulin (IVIg),33,34 cyclophosphamide, azathioprine, and plasmapheresis have been used, with variable reported success. As recently as the 1980s, it was assumed that the neurologic syndrome often remitted spontaneously after surgical removal of the tumor.35 However, several reports have concluded that neurologic symptoms commonly continue after tumor resection and that development rarely returns fully to normal.14,19,31 Currently, most pediatric neurologists and oncologists who are experienced 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 all neurologic symptoms to more modest goals such as reestablishing independent ambulation. The role of chemotherapeutic agents is uncertain. One study reported that children who were treated with chemotherapy had better neurologic outcomes.15 Treatment was presumably for the tumor but with secondary effects of significant immunosuppression. A single patient, reported in 1977, who developed ataxia without opsoclonus after detection of neuroblastoma received extensive chemotherapy, with outcome reported to be good with only mild mental retardation.36 One patient, published in a previous series at this institution, who had apparent spontaneous resolution of opsoclonus-ataxia immediately after resection of a stage III neuroblastoma received multiagent chemotherapy and reportedly had a good neurologic outcome.19 IVIg has been used as an immunosuppressant therapy for a variety of neurologic conditions with presumed autoimmune causes, including chronic immune demyelinating peripheral neuropathy, Guillain-Barre syndrome,37 adult paraneoplastic cerebellar degeneration,38,39 and acute demyelinating encephalomyelopathy.40 Only in Guillain-Barre, chronic immune demyelinating peripheral neuropathy, and multifocal motor 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 least some individuals. No manufacturer of IVIg has yet obtained a labeled indication for any chronic neurologic condition. IVIg has been used for opsoclonusataxia with varying dosage schedules, in various combinations with corticosteroids, ACTH, or other medications.44 47 Although parents often report near-immediate changes in neurologic function after an IVIg treatment, these changes have been difficult to quantify or confirm. This study was undertaken with several goals: 1) to understand the ongoing neurologic and developmental deficits of children who are treated for opsoclonus-ataxia associated with neuroblastomas; 2) to attempt to relate treatment history to outcome; and 3) to attempt objectively to quantify improvements in motor function, speech, mood, and behavior related to IVIg treatment.
87

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

METHODS Patient Recruitment

Patients who were followed in the Neurology Clinic for opsoclonus-ataxia caused by neuroblastomas at Childrens Hospital Los Angeles (CHLA) were invited to participate in the study. In addition, an announcement of the study was placed on the child neurology Internet bulletin board and on the web site of a parentrun support group for opsoclonus-ataxia. Patients who had not previously been treated at CHLA contacted the researchers either by electronic mail or telephone and were prescreened for inclusion. All parents signed an informed consent approved by the institutional 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 younger than 4 years at diagnosis. Children with opsoclonus-ataxia unrelated to a neuroblastoma were not included, nor were children in whom neurologic symptoms developed after treatment of a tumor.

completed in most cases as the majority of the children exhibited low tolerance for structured tasks and exhibited difficulties in naming or expressive vocabulary. Intelligibility assessments were made by the same 2 speech/language pathologists and by parental report.

Psychiatric Examination
The child psychiatrist interviewed the parents with their child and other family members who accompanied them. The parents were asked to describe their childs behavior before and coincident with the onset of neurologic symptoms, past or current psychiatric symptoms, social skills and relationships, and academic function. The parents spontaneously answered open-ended questions based on their memories. The child psychiatrist observed the childs play and interactions and recorded a mental status examination for each child.

Neurologic Examination Record Review

Medical records were reviewed and parents and physicians were interviewed to obtain the following diagnostic information: age at first presentation of neurologic symptoms, presence of any preexisting developmental delays before opsoclonus-ataxia, duration of opsoclonus-ataxia before resection of neuroblastoma, location 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 as well as medications to control symptoms, such as antipsychotics, antidepressants, anxiolytics, clonidine, or sedatives. Imaging studies of the original tumor were reviewed. A standard neurologic examination was performed by the child neurologist. Specific attention was paid to eye movements, gait patterns, reflexes, muscle tone, and behavior during the visit.

Examination of Eye Movements

Patients were asked to observe the movement of red lights from a distance of 1 m on a light bar (Tracoustics Inc, Austin, TX). Saccadic stimuli were flashing lights that alternated from center to 10 degrees off center on the vertical and horizontal meridians at a frequency of 1 Hz, 12 Hz, and 13 Hz. Pendular pursuit stimuli were arrays of light-emitting diodes subtending 0.5 degrees and moving at a sinusoidal velocity either horizontally or vertically over a distance subtending 10 degrees at a frequency of 0.65 Hz, 0.3 Hz, 0.17 Hz, and 0.14 Hz. Optokinetic (OKN) pursuit stimuli were similar arrays of light-emitting diodes separated by a distance of 5 degrees and moving at a constant velocity in one horizontal or vertical direction over a distance subtending 10 degrees. The speed 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 the patients disease later reviewed this tape. Saccadic eye movements were scored as good if they were accurate and maintained the frequency of the stimuli, fair if there was mild to moderate dysmetria or delay, and poor if accuracy and latency were unreliable. Pendular and OKN pursuit eye movements were scored as good if they smoothly maintained the frequency of the stimulus, fair if there was mild to moderate jerkiness; lesser quality pursuit was considered poor.

Standardized Testing
The specific tests used and age ranges are listed in Table 1. For 4 patients who could not obtain valid basal scores on an ageappropriate test as a result of low level of function, a cognitive function test for a younger age group was used. Scores were converted by calculating mental age divided by chronological age, then converted to standard scores.48

Overall Cognitive, Adaptive, and Behavioral Assessment

All children were evaluated by a pediatric neuropsychologist to ascertain their current developmental status and cognitive function, using standardized, age-appropriate measures, in the childs preferred language. Parent questionnaires were administered by the research coordinator in the parents preferred language.

Motor Examination
Fine motor and gross motor skills were evaluated by an occupational therapist. All of the patients had initial evaluations to determine their level of gross and fine motor skills. The standardized tests that were chosen had comparable gross motor scores and examined balance, bilateral coordination, and ambulation. The fine motor categories were also comparable. Both tests included evaluation of eye-hand coordination, upper limb speed, and dexterity.

Reexamination Before and After IVIg

For each child currently receiving IVIg, a brief examination was done before and 2 to 3 days after at least 1 of the treatments. Test selection was determined by results of the detailed baseline examination. Pre- and posttreatment examinations emphasized motor skills, speech, and behavior, as these were the areas that parents most frequently reported as sensitive to the treatment.

Motor Examination
Sample activities were chosen from the initial test for the preand post-IVIg testing sessions. The activities that were chosen for the pre- and post-IVIg testing were ones that the patient could do partially but did not fully meet the standards during the initial evaluation. Because only certain portions of the test were used to score the patients pre- and post-IVIg performance, a standardized score was not possible. Despite this, the items used were still scored in a standardized way on a 3-point scale as described by the test developers. Each childs pre-IVIg performance was measured against his or her post-IVIg performance. The pre- and post-IVIg sessions were also videotaped for later review and comparison. The videotaped material made it possible to detect slight differences in performance that were too subtle to register on the standardized tests, such as changes in quality of performance, tremors, and time needed to complete tasks.

Speech and Language Assessment

Each child was evaluated by pediatric speech and language pathologists to determine developmental levels in receptive and expressive vocabulary and language, articulation/speech intelligibility, and oral motor function. A variety of different tests were used, depending on age, level of functioning, and primary language. Test scores were converted to standard scores and combined into 3 domains: receptive vocabulary, expressive vocabulary, and receptive language. Specific measures used for each domain are listed in Table 1. Two additional measures of expressive language (mean length of utterance in morphemes and Browns 14 grammatical morphemes) were used in all but 3 patients, who were out of the age range. Articulation patterns were obtained during all assessment tasks. A formal test of articulation was attempted but could not be

88

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

TABLE 1.

Standardized Tests Used Scores Generated Standardized Age Range of Test Ages of Patients Tested Out of Standard Range, With Calculated Scoring* 3.79 y, 3.94 y (2 patients) 7.89 y, 8.65 y (2 patients) None None None

Cognitive function Bayley Scales of Infant Development II Wechsler Preschool and Primary Intelligence Scales, Revised Wechsler Intelligence Scales for Children, Third Edition Test of Nonverbal Intelligence 2 Adaptive functioning Vineland Adaptive Behavior Scales Behavior problems Achenbach Child Behavior Checklist Speech and language Peabody Picture Vocabulary Test III Peabody Picture Vocabulary TestR Expressive Vocabulary Test McArthur Communication Development Inventory Expressive One Word Picture Vocabulary TestR Preschool Language Scale3 Clinical Evaluation of Language FundamentalsPreschool Clinical Evaluation of Language Fundamentals3 Test of Auditory Comprehension of LanguageR The Rossetti Infant-Toddler Language Scale Mean Length of Utterance in Morphemes

Mental Development Index Full Scale IQ, Verbal IQ, Performance IQ Full Scale IQ, Verbal IQ, Performance IQ Nonverbal IQ Total, Communication, Daily Living, Social Skills Total, Internalizing, Externalizing Receptive Vocabulary Receptive Vocabulary Expressive Vocabulary Expressive Vocabulary Expressive Vocabulary Receptive Language Receptive Language Receptive Language Receptive Language Receptive Language Standard score for age

3 mo through 3 y 2 y 11 mo through 6 y 6 through 16 y 11 mo Used for single patient age 10.27 y Birth through 18 y 11 mo 2 through 18 y 2.5 y through adult 2.5 y through adult (Spanish translation) 2.5 y through adult 8 through 30 mo 2.5 to 12 y (Spanish translation) Birth through 83 mo 3 through 6.5 y 6 y through adult 3 to 10 y Used for a single patient, age 1.9 y 21 mo to 6 y

None None None Single patient age 3.95 y None None None Single 10.3-y-old patient with very limited speech 6.28 through 8.65 y (3 patients whose age equivalent on vocabulary were under 6 y) 6.28 through 12.62 y (6 patients) None

Browns 14 Morphemes Motor functioning Peabody Developmental Motor Scales: Fine Motor Scores

Number of morphemes used Fine Motor Total, Grasping, Hand Use, Eye-Hand Coordination, Manual Dexterity Total, Balance, NonLocomotor, Locomotor, Receipt/Propulsion Fine Motor Total, EyeHand Coordination, Manual Dexterity, Upper Limb Coordination, Response Speed Gross Motor Total, Balance, Running speed/agility, Bilateral Coordination, Strength

21 mo to 6 y

Birth through 5 y

Gross Motor Scores Bruininks-Oseretsky Test of Motor Proficiency: Fine Motor Scores

Same

None

5 y through adult

None

Gross Motor Scores

Same

None

* Score mental age/chronological age 100.

Speech and Language

Post-IVIg assessments of a subgroup of the available study sample consisted primarily of observations of expressive language based on a brief language sample and parental report and of changes in the childs speech intelligibility per examiner judgment. Because of the short time intervals between pre- and posttreatment conditions, as well as the variable levels of participation

by the very youngest of the children, attempts to readminister formal assessment instruments were limited to the vocabulary tests.

Statistical Analysis
To facilitate comparison across age ranges and testing instruments, we converted all speech and language and motor test

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

89

IVIg # Courses

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

Onset to IVIg (Month) First IVIg (Month) Resection to RX (Month) Onset to RX (Month) Other Chemo (Y/N) ACTH (Y/N) Initial Steroid (Month) Age NB Diagnosed (Month)

RESULTS

Seventeen children were examined: 7 boys and 10 girls. Twelve were primarily followed for treatment of opsoclonus-ataxia at CHLA. This represented all but 2 of children known to have opsoclonus-ataxia caused by neuroblastoma seen at CHLA during the previous 17 years. No family that could be contacted declined to participate. During the same time period, only a single child who was younger than 4 years and seen at onset of opsoclonus-ataxia at CHLA was not found to have a neuroblastoma. She was not included in the study group. Five additional children whose source of neurologic care for opsoclonusataxia was elsewhere participated. In 1 of the 5 patients, the original diagnosis and tumor resection had been at CHLA, but all subsequent neurologic treatment was provided in another state. Age at examination ranged from 1 to 13 years. The interval between onset of opsoclonus-ataxia and examination ranged from 1 week to 12 years.
Medical Review

Medical diagnostic and treatment information is summarized in Table 2. Fifteen of the 17 children had opsoclonus-ataxia; 2 had ataxia alone, without disorder of eye movements. Neuroblastoma was diagnosed by a variety of imaging modalities, most frequently contrast-enhanced thin-cut computerized tomography. Thirteen children had abdominal tumors, 3 had thoracic tumors, and 1 had a tumor in the 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 medication, in various dosages, schedules, and combinations. All received oral corticosteroids, at least as the initial treatment. In 6 children, ACTH (Acthar gel, Rhone-Poulenc Pharmaceuticals, Collegeville, PA) was substituted when initial response to oral medication was suboptimal or when oral dosing was not tolerated. IVIg was used in 12 patients at a dose of 2 g/kg over 1 or 2 days, ranging from a single course to 40 treatments. Two children received cyclophosphamide for symptoms of opsoclonus-ataxia, 1 immediately after diagnosis and 1 a year later. One child received bolus methylprednisolone, cyclophosphamide, methotrexate, and chlorambucil when he developed steroid-resistant nephrotic syndrome several years after beginning treatment for opsoclonusataxia. Although the treatment was not directed at his neurologic condition, chronic ataxia improved after intensive immunosuppression. The interval between the onset of opsoclonus-ataxia and diagnosis of neuroblastoma ranged from 2 days to 20 months in the group. The interval between the onset of symptoms and initiation of immunosuppressant treatment ranged from 1 week to 29 months, with delays of 1 year in 4 children. One additional pa90

18 21 21 18 25 21 14 23 18 36 28 16 13 16 17 38 18 arp indicates abdominal, retroperitoneal; aa, abdominal, adrenal; t, thoracic; n, neck; nd, not done. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 M M F F M F F F F F F M F F M M M arp arp aa t aa aa aa arp aa t aa aa n arp aa arp aa 2 2 2 1 2 2 2 1 2 1 2 2 1 2 1 1 2 1 1 1 1 1 1 1 1 1 1 nd 1 nd nd nd nd nd 1.75 1.90 2.74 2.27 3.05 3.30 3.68 3.79 3.94 4.26 5.95 6.28 7.89 8.65 10.27 10.33 12.62 CHLA CHLA Parent CHLA CHLA CHLA CHLA CHLA Parent Parent Parent CHLA Parent CHLA CHLA CHLA CHLA 18 18 20 17 18 21 14 19 17 31 19 15 11 13 13 18 15

Medical Information: Tumor Location, Stage, and Treatment

TABLE 2.

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

Patient

Gender

Location

Stage

N-Myc

Age at Examination Years

Recruited From CHLA Versus Parent Group

First Symptoms (Month)

18 22 37 18 28 21 17 24 47 31 19 35 13 33 18 40 19

N N N N N Y N Y N N Y Y Y N Y N N

N N N N N N N Y N N N N Y N N N Y

0 4 17 1 10 0 3 5 30 0 0 20 2 20 5 22 4

0 1 16 0 3 0 3 1 29 5 9 19 0 17 1 2 1

18 nd 39 18 nd 21 15 25 nd 33 38 nd 50 35 19 nd 19

1 0 8 4 0 3 18 41 0 15 11 0 2 8 12 0 1

0 nd 19 0 nd 0 1 2 nd 2 19 nd 39 22 6 nd 4

tient was treated briefly with steroids, followed by IVIg during the year after diagnosis, then restarted corticosteroid and IVIg treatment 9 years later. Three children began corticosteroids, ACTH, and/or IVIg before the resection of the neuroblastoma; 1 had a delay of 5 months between recognition of opsoclonus-ataxia and beginning therapy and detection of the tumor, during which she received ACTH and oral steroids. All children received physical and occupational therapy for periods ranging from a few months to 5 years after diagnosis and with frequency of treatment sessions ranging from monthly to 4 times a week. All but 2 of the children received intervention for speech/language delay. The intensity of the intervention ranged from 1 to 4 sessions a week. Formal hearing test results were not available, but there were no reports or history of hearing impairment. All children who were older than 3 years at the time of evaluation were enrolled in school programs with variable levels of special intervention, ranging from self-contained classes for the multiply handicapped to a few hours a week of resource specialist. None were fully mainstreamed educationally.
Neurologic Findings

He was ambulatory only with a walker, had severe drooling, and had essentially no speech. Subsequent to the study visit, he restarted oral corticosteroids with moderate improvement in gait, cessation of drooling, and minor improvements in speech. Patient 1 was not yet fully ambulatory (crawling, pulling to stand, cruising) at the time of the study visit, 3 months after diagnosis, while on oral corticosteroids and IVIg. Six patients had mild to moderate lower extremity hyperreflexia (nonsustained clonus, Babinski responses, and/or crossed adductor reflexes), but none had significant clinical spasticity or toe-walking.
Eye Movements

One child had a sixth nerve palsy, which developed shortly after onset of opsoclonus-ataxia and had not cleared as of the examination date, 3 months after onset. One had epilepsy (patient 12). At the time of the examination, 2 children were ambulatory only with a walker (patients 9 and 16). Patient 9 had begun treatment with prednisone for the first time shortly before the study visit, 30 months after onset of opsoclonus-ataxia and resection of her neuroblastoma. Subsequent to the study visit, while receiving oral prednisone, she continued to have rapid improvement in motor function and became fully ambulatory without assistance several months later. Patient 16 was treated for approximately 1 year after diagnosis but had been off all immunosuppressive medications for 7 years at the time of the study visit.
TABLE 3. Patient Cognitive and Adaptive Function Full Scale IQ/ Mental Development Index 56 54 86 50 61 94 91 58* 53* 70 60 65 50* 44* 68 48 46 Verbal IQ N/A N/A N/A N/A N/A 92 93 N/A N/A 71 61 55 53* 35* N/A 55 54 Performance IQ N/A N/A N/A N/A N/A 97 89 N/A N/A 75 66 80 46* 52* N/A 49 47

Six patients (patients 1115 and 17) were cooperative for the videotaped examination of eye movements. All 6 patients had greater impairment of pursuit eye movement than saccadic eye movements. Five patients had good horizontal saccades at all target speeds. None had good vertical saccades at 1 Hz, and only 2 had good vertical saccades at slower target speeds. All patients had poor pursuit of the fastest pendular (0.65 Hz) and OKN (22 degrees/sec) targets. Vertical pursuit eye movement was usually worse and never better than horizontal eye movement. Deficiencies in eye movement usually improved with slower stimuli. Often an eye movement was better in one direction than in the opposite direction, and this directional asymmetry was consistent for any given patient. There seemed to be a relationship between eye movement scores and IQ. However, the small number of patients with eye movement examinations made statistical testing impractical.
Cognitive Functioning

Scores on cognitive and adaptive measures are summarized in Table 3. The neuropsychologic data indicate significant deficits in overall cognitive development. Overall cognitive functioning was normal in only 3 children, whose full scale IQ or Mental

Test Used

Vineland Total 74 75 76 74 75 89 71 56 69 70 63 58 56 41 31 53 52

Vineland Communication 80 79 91 76 79 98 84 61 74 79 70 61 51 41 33 44 57

Vineland Daily Living 77 79 73 87 81 98 83 60 90 78 73 64 60 41 20 60 49

Vineland Social Skills 80 82 84 80 76 89 77 62 78 79 73 65 72 50 48 58 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Bayley-II Bayley-II Bayley-II Bayley-II Bayley-II WPPSI-R WPPSI-R Bayley-II* Bayley-II* WPPSI-R WPPSI-R WISC-III WPPSI-R* WPPSI-R* TONI-2 WISC-III WISC-III

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

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

91

NonLocomotor

Development Index was 1 SD below the norm. All other children were functioning in the impaired range, with IQ or MDI 2 SD below the mean (see Table 3). Adaptive functioning was similarly impaired; only 1 childs score was within 1 SD of the mean, and 6 were in the borderline range (1 to 2 SD below normal). Increasing age was associated with increasing deficits in adaptive functioning. The parents reported more behavioral difficulties on the adaptive functioning measures than on a behavioral measure, the Achenbach Child Behavior Checklist,49 but both of these underestimated behavioral disturbances observed during the clinical interviews. In general, parent responses on the Child Behavior Checklist did not indicate significant emotional/behavioral problems. Only 4 patients had a Total Score in the clinical range. Only 1 patient had a clinically significant level of Externalizing symptoms, and no patient had Internalizing symptoms. On the Syndrome Scales, clinically significant elevations were Attention Problems (5), Social Problems (4), and Thought Problems (3).
Motor Skills

Strength

Bilateral Coordination

Receipt/ Propulsion

Running Speed/ Agility

Locomotor

Speech and Language

Detailed scores on speech and language measures are in Table 5. The majority of the children participated in the assessment tasks with moderate to maximum support to maintain attention and interest.
92

TABLE 4.

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

Data are expressed as z scores, with mean 0, SD 1. PDMS indicates Peabody Developmental Motor Scales; BO, Bruininks-Oseretsky Test of Motor Proficiency.

Detailed Test Scores, Motor Testing

Fine Motor Total

Detailed motor testing results are listed in Table 4. In general, younger children were more likely to score within the age-acceptable range than the older children. Ten children were tested using the Peabody Developmental Motor Scale.50 Seven were tested with the Bruininks-Oseretsky Test of Motor Proficiency.51 The younger children, tested on the Peabody Developmental Motor Scale, scored lower on the fine motor tasks than on the gross motor tasks. Five of 10 children performed poorly on eye-hand coordination. In contrast, hand usage and grasping were within or above the acceptable ranges in 9 of 10 patients. The gross motor performance of the children was slightly better than the fine motor scores. The locomotor, balance, and receipt and propulsion categories were the most impaired, although only 3 of 10 had z scores lower than 1.5. Nine scored within the acceptable ranges for all of the gross motor categories except for the nonlocomotor category, in which 10 children scored within the acceptable ranges. In contrast to the mostly acceptable scores for the younger group, the older children had z scores that were mostly below the acceptable range for both gross and fine motor abilities. Two children scored above 1.5 on the balance and strength categories, but all of the other gross motor scores were below the acceptable range and even lower than 2. The fine motor performance of the older group was also below the acceptable age range, and only 1 child could score above 1.5 on response speed and visual motor control.

Response Gross Balance Speed motor Total

Upper Limb Coordination

Manual Dexterity

Eye-Hand Coordination

Grasping

Patient

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

PDMS PDMS PDMS PDMS PDMS PDMS PDMS PDMS PDMS PDMS BO BO BO BO BO BO BO

Test Used

0.05 0.28 0 2.33 2.05 0.02 1.13 1.56 2.02 1.75 4.48 3.32 3.54 2.46 3.56 3.44 3.32

1.56 0.00 0.00 0.68 2.33 0.71 1.25 2.33 2.05 2.33

0.61 2.33 2.33 0.02 0.92 2.33 2.33 2.33 1.04 2.33

Hand Use

0.02 0.01 0.01 2.05 0.92 0.23 1.00 1.88 1.88 2.05 2.80 2.40 2.80 2.80 2.80 1.20 2.80

0.36 0.33 0.18 1.08 0.39 0.92 0.10 1.13 1.75 1.40 2.80 2.80 2.80 2.80 2.80 2.80 2.80

2.80 2.80 2.80 3.00 3.00 2.80 2.60

1.80 1.60 2.40 2.40 2.80 2.60 0.60

0.81 2.33 0.05 0.31 2.33 0.18 2.05 1.56 2.05 0.15 2.60 3.35 2.55 3.56 5.24 3.46 3.45

1.04 0.71 0.28 0.08 0.13 0.55 1.75 0.23 2.33 0.58 2.80 1.40 2.80 2.80 2.80 1.40 2.60

1.18 1.18 1.13 0.33 0.50 0.50 0.41 0.33 0.33 0.05

0.95 2.05 0.05 0.25 0.10 0.36 1.34 1.34 2.33 0.10

2.80 2.80 2.80 2.80 2.80 2.80 2.80

0.05 0.44 0.05 0.68 1.08 0.33 0.15 1.56 2.05 0.71

2.40 1.80 2.80 2.80 2.80 2.60 2.20

2.40 1.00 1.80 2.80 2.80 2.40 1.40

TABLE 5. Patient

Language and Speech Receptive Vocabulary Score Test Used Expressive Vocabulary Score N/A N/R 1.13 1.64 2.13 1.47 0.60 2.47 1.40 N/R 0.80 2.94 3.60 3.33 N/V 2.20 2.40 Test Used Receptive Language Score N/A 1.53 1.00 N/R 0.10 N/R N/R 1.60 1.13 N/R 1.70 2.07 3.00 N/R 2.33 2.26 3.90 Test Used Rossetti PLS-3 PLS-3 PLS-3 PLS-3 PLS-3 PLS-3 CELF-P CELF-3 TACL-R CELF-3 CELF-3 1.61 1.77 1.00 1.64 2.72 0.13 0.86 2.91 3.50 0.70 0.20 2.83 3.50 3.50 N/V N/A N/A MLU-m Browns 14 Morphemes 0 0 3 3 2 3 0 0 1 8 6 0 2 1 N/V 14 8

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

N/A N/R 0 N/R 0.53 0 0.20 2.33 0.60 N/R 1.60 2.40 1.90 4.0 1.60 2.20 2.13

PPVT-III PPVT-III PPVT-III PPVT-III PPVT-III PPVT-III PPVT-III PPVT-III PPVT-III PPVT-R PPVT-III PPVT-III PPVT-III

EVT MacArthur EVT EVT EVT EVT EVT EVT EVT EVT EOWPVT-R EVT EVT

Scores for language are expressed as standardized z scores, with mean 0, SD 1. Browns 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 Vocabulary TestIII; PPVT-R, Peabody Picture Vocabulary TestR; MacArthur, MacArthur Communicative Development Inventory: Words and Sentences; EVT, Expressive Vocabulary Test; EOWPVT-R, Expressive One-Word Picture Vocabulary Test-R; Rossetti, The Rossetti Infant-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.

Inconsistency in attending and a tendency for agitation and impulsivity were observed and were judged to be at levels greater than expected in a typical population of children of the same chronological age. In a number of cases, nonparticipation resulted in missing scores. All test scores were converted to z scores for comparison across domains, with the exception of Browns 14 morphemes, which is presented as raw score. The language sampling measure, mean length of utterance in morphemes (MLUm), was converted to z scores by using the formula provided by Miller.52 Language sample sizes obtained were of variable length and in most cases were smaller than the standard 50 utterances on which the normative data are based. In comparison to typically developing children, the low quantity of output noted for most of the study children represented an area of significant clinical deficit. Abnormalities in vocabulary and language development were pervasive; several children had such severe deficits that they were below the scorable range. Only two z scores were 1 SD above the mean, obtained by 2 different children on the expressive vocabulary measure. Comparing receptive and expressive vocabulary performance in the 12 patients for whom scores on both tests were available, receptive vocabulary z score was better than expressive vocabulary in 7, expressive vocabulary was better in 4, and scores were equal in 1. The language sample analysis revealed the following results. The range of observed spontaneous utterance length in words for the ages for which this is a 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 syntactic complexity referenced by MLU-m, only 5 of 12 children (45%) in the age range had MLU-m within 1 SD of the mean for age. The number of different gram-

matical morphemes (Browns 14) ranged from 0 to 14 morphemes in the sample. Within the standard age range of the measure (27 60 months), the range was 0 to 8. Four children ranging between 30 months and 53 months had 1 or none of the 14 grammatical morphemes in their language samples. In typically developing children, the acquisition of the 14 morphemes corresponds to language stages starting at approximately 27 to 30 months and continuing through 47 to 60 months. Approximately 50% of the morphemes may be acquired by 43 to 46 months, and 100% of the morphemes are generally acquired by 60 months of age. Of 7 patients tested at above 60 months of age, when all 14 morphemes are expected, 1 was nonverbal, 1 had all 14 morphemes, and the remaining patients had from 0 to 8 morphemes. All of the children in the sample exhibited misarticulations involving single-sound errors, a number of phonological patterns associated with young developmental levels, and single-sound and sound combination distortions associated with oral motor coordination and planning difficulties. Difficulties in coordinating respiration and motor patterns necessary for speech were observed. Intelligibility judgments by the 2 evaluators (unfamiliar individuals versus parents) ranged from 5% to 90%, with the highest intelligibility being observed in a 51-monthold child. Parental reports of intelligibility were in the 30% to 100% range, with the mode being 65%. Oral motor examination to assess function of the oral mechanism revealed difficulties in volitional control of movements involving tongue elevation and lateralization. Inefficiency in imitating sounds was observed, often characterized by groping postures and exhalation before articulation. Varying patterns of productions of sounds or words imitated were observed. Hypernasal vocal quality and inconsistent volume modulation were also noted.
93

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

Psychiatric Examination

All children had significant psychiatric and behavioral disturbances early in their course. The most prominent symptom at the onset of illness, coincident with onset of neurologic symptoms, was irritability, noted in 15 of the 17 patients. Most children were remembered as inconsolable, difficult to soothe, or needing to be held all the time. Night terrors and disturbed sleep were prominent, reported in 10 children. Severe tantrums with self-injurious behaviors, such as head banging, were reported in 7 patients and were distressing to the families. Mood swings and poor affective regulation were sometimes recognized at onset (4 patients) but were associated later with steroid or ACTH therapy in 5 additional patients and persisted into the chronic phase in 8 of 10 patients. This pattern represented a marked change from premorbid personality or temperament, and although these psychiatric symptoms were often the most disruptive for the family, they were not the reason that medical attention was sought. When the children were seen for psychiatric evaluation, these symptoms had changed but not disappeared. Irritable, dysphoric mood and mood swings were still common problems (10 of 17), and tantrums and oppositional and aggressive behavior were more prominent than earlier (10 of 17). Older children had a history of difficulty with toilet training. Three of 10 had persistent enuresis, and 1 had encopresis.
Treatment Effects

proximately equal size. Five received no IVIg, 7 received 1 to 8 courses, and 5 received 10 to 41 courses. Age at evaluation of the 3 groups was similar (3.7 years for the group with no IVIg, 3.3 years for the group with 1 to 8 courses, and 4.3 years for the group with 10 or more courses). There were no significant differences in cognitive, adaptive, or motor functioning among treatment groups. However, the group with no IVIg contained a disproportionate number of children with long delays between onset of symptoms and treatment. There was a striking relationship between age at evaluation and level of function, with older children exhibiting significantly more deficits in overall adaptive behavior and gross and fine motor function than younger children. Figure 1 plots age versus IQ or MDI and Vineland total score. Figure 2 plots age versus z-transformed gross and fine motor function. Pearson correlation of age with Vineland score was 0.812 (P .0001), with fine motor function was 0.711 (P .001), and with gross motor function was 0.753 (P .0001). Correlation of age and IQ/MDI was 0.398 (NS). Median delay from onset to treatment was 4 months, with a range of 0 to 30 months. Timing of treatment was unrelated to cognitive, adaptive, or motor 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 outcome.
Acute Effects of IVIg

Six children received ACTH, either as primary treatment or after they did not tolerate or respond to oral prednisone. Median IQ (Wechsler D. Wechsler Intelligence Scales for ChildrenIII. San Antonio, TX: The Psychological Corporation; 1991 or Wechsler D. Wechsler Preschool and Primary Scales of IntelligenceRevised. San Antonio, TX: The Psychological Corporation; 1989) or MDI (Bayley [Bayley N. Bayley Scales of Infant Development, Second Edition. San Antonio, TX: The Psychological Corporation; 1993]) of the group that received ACTH was not different from the group that received only oral steroids (67 without ACTH versus 73 with ACTH; P .366). Similarly, performance on the Vineland Scales (Sparrow SS, Balla DA, Cichetti DV. Survey Form Manual, Vineland Adaptive Behavior Scales. Circle Pines, MN: American Guidance Service, Inc; 1984) or the gross motor measures did not differ in the group that received ACTH versus only oral corticosteroids. Fine motor measures tended to be worse in the group that received ACTH, approaching but not reaching statistical significance (z 3.4 versus 1.75; P .078). Three children who received chemotherapy were compared with 14 who did not. No measure showed any advantage for the small group that received chemotherapeutic agents, with trends toward lower functioning in the small group that received them. Median IQ or MDI for the 3 children who received other chemotherapy or immune modulating agents was 63, compared with 71 for the other 14 children (P .101). The effect of total use of IVIg was examined by dividing the patients into 3 groups of ap94

Five children (ages 16 months to 55 months) were examined immediately before and 1 to 3 days after a treatment with IVIg. Two grams of IVIg per kilogram of body weight were infused over 6 to 8 hours in 4 children. One gram per kilogram daily for 2 doses on 2 successive days was infused in 1 child. Premedication with acetaminophen and diphenhydramine was provided to prevent headache, nausea, or rash.
Speech and Language

An overall increase in the quantity of verbalization was observed and reported after IVIg treatment. This observation by the examiners may have been partially attributable to the posttreatment evaluations being the third visit of the child. However, all parents reported increased posttreatment verbal interaction. There were also reports of increased attentiveness to verbal input, as well as increased length of utterance, albeit sometimes in the form of jargon. The examiners observed increased attention to task and decreased agitation during structured activities. No consistent parental reports regarding changes in speech intelligibility were obtained. There was some perception of increased intelligibility by the examiners, but this may have been attributable to familiarity factors, as well as increased verbalization in context.
Motor Skills

Balance, locomotor, and receipt and propulsion skills were the gross motor categories that were affected by the IVIg treatments. Two children were

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

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

able to climb stairs with an alternate pattern, placing only 1 foot on each stair after the IVIg session, but could not do this task a few days earlier during the evaluation before the IVIg. A younger child who was crawling up the stairs during the pre-IVIg evaluation was able to climb the stairs, placing both feet on each stair while holding onto the rail after her IVIg session. One boy who could not stand on 1 leg before the IVIg session was able to do a single limb stance for 5 seconds. Three of the children tested were able to bounce a tennis ball against a wall after the IVIg session but could not do this task before the IVIg. Eye-hand coordination and hand usage were the fine motor categories that improved after the IVIg treatments. The children were able to copy basic shapes and cut with scissors after the IVIg sessions with more skills and precision than before the treatment.
DISCUSSION

It has been repeatedly observed that paraneoplastic opsoclonus-ataxia caused by neuroblastoma is a chronic neurologic condition associated with significant morbidity attributable to long-term developmental deficits. This study confirms previous reports. There was a striking relationship of age at testing to degree of deficit in this group. The youngest, most recently treated children were functioning at nearly normal levels, whereas deficits were much more striking in the oldest children. Unfortunately, we cannot address a major question regarding this relationship: is the apparent worsening of developmental level with age attributable to plateauing of

development (ie, a static encephalopathy, present fully at onset of symptoms, that becomes more apparent with increasing demands on the child), or is there a progressive process that worsens with each relapse 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 were treated more aggressively received such treatment because of more severe symptoms or more frequent relapses, which might be expected to have a poorer neurologic prognosis. Only 3 children in this group were treated with chemotherapy, at varying times after the onset of symptoms. One child received cyclophosphamide immediately before surgery and for 6 postoperative courses, specifically for opsoclonus-ataxia. A second child received 1 year of cyclophosphamide after she repeatedly relapsed on ACTH and IVIg, with severe psychiatric symptoms and behavioral problems in addition to ataxia. A third child received multiple courses of methotrexate, cyclophosphamide, and chlorambucil after he developed an apparently unrelated autoimmune condition 3 years after diagnosis of opsoclonus myoclonus syndrome. In contrast to the experience of Russo et al,15 these 3 children were functioning at low levels, and the additional chemotherapy did not seem to provide neurologic or developmental protection. We also did not find a selective advantage of
95

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

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 and SD of 1.

ACTH over oral corticosteroids, although dosage, length of course, and schedules varied too widely to be certain. This contrasts with the experience reported by Hammer et al,31 who found oral steroids to be useless and noted responses only to ACTH. In our group, however, several children who had not responded to high-dose oral corticosteroids (generally 2 mg/kg/d of oral prednisone) responded well to intramuscular ACTH gel, giving some indication that it may provide superior results in some instances. We cannot make any firm conclusion about either the possible role of chemotherapy or the benefit of ACTH, given the descriptive nature of the study. This will require randomized prospective trials. Nevertheless, a great deal of attention has been given to findings of Russo et al,15 attributing a longterm advantage in developmental outcome to use of cyclophosphamide. In view of our results and those of 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 examined both before and 2 to 3 days after an infusion. There were clear gains in motor function; both qualitative differences in motor abilities and specific tasks that could not be performed before treatment were completed posttreatment. Speech quality improved, but, more striking, ability to cooperate with the evaluation and overall mood was observed to improve post-IVIg. This study does not provide adequate measures of the long-term efficacy of IVIg in either maintaining improvement or decreasing long96

term deficits. We did not find that patients who were treated with many courses of IVIg benefited in longterm comparison to those who were treated with only a few course or none, but, again, this may be confounded by varying disease severity, which dictated more treatment in severely affected children. Although opsoclonus eventually disappeared in all children, abnormalities of pursuit eye movements and, to a lesser extent, saccadic eye movements persisted in all patients who could cooperate with formal testing. Considering the disordered features observed in the communication of the young children (younger than 7 years), morphologic and syntactic components of language were the most striking. Even in children with relatively longer sentences, the morphologic markers typically expected in the 26- to 48-month age range were frequently absent. The significantly depressed expressive vocabulary levels, both in quantity and variety, may be a function of overall cognitive developmental levels, which were low in nearly all of the patients. Speech characteristics responsible for the extremely low levels of intelligibility were also significant for being directly associated with oral, phonatory, and respiratory incoordination. Although some developmentally recognizable misarticulation/phonological patterns were noted, the very poor intelligibility noted is considered to be motor based. The significant expressive language impairment observed, combined with the severe speech intelligibility problems, results in significantly impaired communicative competence. Even the children who

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

had received early and intensive speech and language intervention exhibited persistent speech and language disorder. The psychiatric and behavioral aspects of opsoclonus-ataxia deserve special mention, as they have not been emphasized in previous reports and contribute substantially to overall morbidity. Irritability has been noted as a presenting symptom of opsoclonusataxia syndrome, regardless of cause.25 Irritability was recognized as a presenting symptom in earlier studies of opsoclonus-ataxia.19,31 Anecdotal reports documented the reduction in the quality of life in patients with opsoclonus-ataxia, and children were recognized to have persistent disruption of socialemotional stability.53 Extensive psychosocial evaluations of these patients noted emotional dysregulation 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 the cerebrum, particularly the frontal cortex.54 Dysfunction in cerebellar-cerebral circuits may underlie the affective dysregulation and behavioral disturbances seen in opsoclonus myoclonus syndrome, as dysfunction of other cerebellar circuits underlie the movement symptoms characteristic of the disorder. Additional investigation into the treatment of opsoclonus-ataxia is clearly warranted. Given the disparate findings of this and other series of patients with neuroblastoma and opsoclonus-ataxia regarding possible treatment effects (ACTH versus oral steroids, IVIg, and the effects of chemotherapy), a randomized study comparing various treatment modalities is needed.
ACKNOWLEDGMENTS
This study was made possible by a grant from Aventis-Behring (formerly Centeon). We thank the parents and children who participated, and Sandra Greenberg, organizer of the opsoclonusataxia parent group web site, for help in contacting interested families. 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 paraneoplastic neurological syndromes. Ann Med Intern (Paris). 1998;149:512520 3. Posner JB, Dalmau JO. Paraneoplastic syndromes affecting the central nervous system. Annu Rev Med. 1997;48:157166 4. Dalmau J, Gultekin HS, Posner JB. Paraneoplastic neurologic syndromes: pathogenesis and physiopathology. Brain Pathol. 1999;9: 275284 5. Posner JB, Dalmau JO. Paraneoplastic syndromes of the nervous system. Clin Chem Lab Med. 2000;38:117122 6. Boltshauser E, Deonna T, Hirt HR. Myoclonic encephalopathy of infants or dancing eyes syndrome. Report of 7 cases with long-term follow-up and review of the literature (cases with and without neuroblastoma). Helv Paediatr Acta. 1979;34:119 133 7. Bray PF, Ziter FA, Lahey ME, Myers GG. The coincidence of neuroblastoma 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 encephalopathy associated with neuroblastoma. Child Nerv Syst. 1987;3:245247 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;86:235236 12. Pranzatelli MR. The neurobiology of the opsoclonus-myoclonus syndrome. 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 antibodiesa report from the Childrens 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;125:712716 15. Russo C, Cohn SL, Petruzzi MJ, de Alarcon PA. Long-term neurologic outcome in children with opsoclonus-myoclonus associated with neuroblastoma: a report from the Pediatric Oncology Group. Med Pediatr 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;24:1114 18. Janss A, Sladky J, Chatten J, Johnson J. Opsoclonus/myoclonus: paraneoplastic syndrome of neuroblastoma. Med Pediatr Oncol. 1996;26: 272279 19. Mitchell WG, Snodgrass RS. Opsoclonus/ataxia in childhood neural crest tumors: a chronic neurologic syndrome. J Child Neurol. 1990;5: 153158 20. Hiyama E, Yokoyama T, Ichikawa T, et al. Poor outcome in patients with advanced stage neuroblastoma and coincident opsomyoclonus syndrome. Cancer. 1994;74:18211826 21. Okano HJ, Park WY, Corradi JP, Darnell RB. The cytoplasmic Purkinje onconeural antigen cdr2 down-regulates c-Myc function: implications for neuronal and tumor cell survival. Genes Dev. 1999;13: 20872097 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 cerebellar 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 autoantibodies in patients with paraneoplastic and idiopathic opsoclonus myoclonus 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-myoclonus. J Pediatr Hematol Oncol. 2000;22:315320 28. Fisher PG, Wechsler DS, Singer HS. Anti-Hu antibody in a neuroblastoma-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 paraneoplastic syndromes. Brain Pathol. 1999;9:251260 31. Hammer MS, Larsen MB, Stack CV. Outcome of children with opsoclonus-myoclonus regardless of etiology. Pediatr Neurol. 1995;13:2124 32. Pohl KR, Pritchard J, Wilson J. Neurological sequelae of the dancing eye syndrome. Eur J Pediatr. 1996;155:237244 33. Borgna-Pignatti C, Balter R, Marradi P, Colamaria V. Treatment with intravenously administered immunoglobulins of the neuroblastomaassociated 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

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

97

38.

39.

40.

41. 42.

43.

44.

45.

FG. Single dose immunoglobulin therapy for childhood Guillain-Barre syndrome. Brain Dev. 1997;19:323325 Blaes F, Strittmatter M, Merkelbach S, et al. Intravenous immunoglobulins in the therapy of paraneoplastic neurological disorders. J Neurol. 1999;246:299 303 Uchuya M, Graus F, Vega F, Rene R, Delattre JY. Intravenous immunoglobulin treatment in paraneoplastic neurological syndromes with antineuronal autoantibodies. J Neurol Neurosurg Psychiatry. 1996;60: 388 392 Finsterer J, Grass R, Stollberger C, Mamoli B. Immunoglobulins in acute, parainfectious, disseminated encephalo-myelitis. Clin Neuropharmacol. 1998;21:258 261 Dalakas MC. Intravenous immune globulin therapy for neurologic diseases. Ann Intern Med. 1997;126:721730 Stangel M, Hartung HP, Marx P, Gold R. Intravenous immunoglobulin treatment of neurological autoimmune diseases. J Neurol Sci. 1998;153: 203214 Frederico P, Zochodne DW, Hahn AF, Brown WF, Feasby TE. Multifocal motor neuropathy improved by IVIg. Randomized, double-blind, placebo-controlled study. Neurology. 2000;55:1256 1262 Petruzzi MJ, de Alarcon PA. Neuroblastoma-associated opsoclonusmyoclonus treated with intravenously administered immune globulin G. J Pediatr. 1995;127:328 329 Pless M, Ronthal M. Treatment of opsoclonus-myoclonus with high-

dose intravenous immunoglobulin. Neurology. 1996;46:583584 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 polymyoclonia syndrome. Acta Paediatr. 1992;81:371372 47. Veneselli E, Conte M, Biancheri R, Acquaviva A, De Bernardi B. Effect of steroid and high-dose immunoglobulin therapy on opsoclonusmyoclonus syndrome occurring in neuroblastoma. Med Pediatr Oncol. 1998;30:1517 48. Sattler JM. Assessment of Children. San Diego, CA: Jerome M. Sattler Publisher; 1992:2122 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 opsoclonus-myoclonus syndrome. Dev Med Child Neurol. 1995;37:915932 54. Schmahmann JD, Sherman JC. The cerebellar cognitive affective syndrome. Brain. 1998;121:561579

THE LETHAL MYTH OF MASCULINITY

The Marlboro Man emanated in 1954 from the minds of Chicago admen Leo Burnett and John Benson, who were trying to devise a more macho pitch for Phillip Morris 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 (and 2 actors who played him died of lung cancer).

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

Submitted by Student

98

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

Opsoclonus-Ataxia Caused by Childhood Neuroblastoma: Developmental and Neurologic Sequelae Wendy G. Mitchell, Yolanda Davalos-Gonzalez, Virdette L. Brumm, Sonia K. Aller, Elvira Burger, Susan B. Turkel, Mark S. Borchert, Susan Hollar and Sonia Padilla Pediatrics 2002;109;86-98
Updated Information & Services Citations Subspecialty Collections including high-resolution figures, can be found at: http://www.pediatrics.org/cgi/content/full/109/1/86 This article has been cited by 13 HighWire-hosted articles: http://www.pediatrics.org/cgi/content/full/109/1/86#otherarticles This article, along with others on similar topics, appears in the following collection(s): Tumors http://www.pediatrics.org/cgi/collection/tumors Information about reproducing this article in parts (figures, tables) or in its entirety can be found online at: http://www.pediatrics.org/misc/Permissions.shtml Information about ordering reprints can be found online: http://www.pediatrics.org/misc/reprints.shtml

Permissions & Licensing

Reprints

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