466606

2012
MSJ19710.1177/1352458512466606Multiple Sclerosis JournalHarder et al.

MULTIPLE
SCLEROSISMSJ
Research Paper JOURNAL

Multiple Sclerosis Journal

Cognitive functioning in 19(7) 947952

The Author(s) 2012
Reprints and permissions:
pediatric transverse myelitis sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/1352458512466606
msj.sagepub.com

Lana L Harder1,2,3, Alice Ann Holland1,2, Elliot Frohman1,3,

Donna Graves1,3 and Benjamin M Greenberg1,3

Abstract
Background: Transverse myelitis (TM) is an inflammatory disease of the spinal cord. In pediatric TM patients, cognitive
and psychological problems have been described only anecdotally.
Objectives: Study aims include describing cognitive dysfunction among a cohort of pediatric TM patients as well as
qualitatively exploring the impact of depression, medication, and fatigue on cognitive functioning.
Methods: Twenty-four consecutive TM patients referred to a pediatric demyelinating diseases clinic completed
neuropsychological screening. Means, standard deviations (SD), and percentages of patients performing at or below 1.0,
1.5, and 2.0 SD from the mean on tests administered are presented.
Results: Means were generally average across domains; however, scores ranged widely across subjects within each
domain. The highest rate of deficits was observed in fine-motor speed/dexterity. Slightly higher frequencies of impairment
were observed in attention and memory as compared to processing speed and verbal fluency. Results did not suggest a
clear association between cognitive problems and depression or medication use but did suggest that fatigue may impact
cognitive functioning.
Conclusions: This study is the first to document cognitive deficits in pediatric TM and raises questions regarding our
understanding of the central nervous system (CNS) injury associated with TM. Findings warrant further exploration of
neuropsychological outcomes in TM to inform appropriate intervention.

Keywords
Neuropsychology, transverse myelitis, pediatric demyelinating disease
Date received: 30th May 2012; revised: 12th September 2012; accepted: 7th October 2012

Introduction
Idiopathic transverse myelitis (TM) is diagnosed based on
evidence of inflammation within the spinal cord and the
absence of cerebral involvement, generally differentiating
it from acute disseminated encephalomyelitis (ADEM) and
multiple sclerosis (MS). Estimates suggest that up to 20%
of TM cases emerge prior to age 18 years.1 Previous
research has not systematically characterized this highly
disabling disorder among pediatric patients; however,
research has begun to elucidate the unique experiences of
pediatric patients with TM.2,3
Cognitive impairment is commonly associated with
brain-based demyelinating diseases including MS and
ADEM, conditions that target both the brain and spinal
investigation of cognitive functioning likely derives from
the perceived restricted anatomic localization of TM.6 In
our multi-disciplinary TM center, we routinely assess all
central nervous system (CNS) demyelinating disease
patients with neuropsychological screening. The objec-
tive of the present study was to describe cognitive dys-
function among pediatric TM patients.
1Childrens Medical Center Dallas, USA.
2University of Texas Southwestern Medical Center, Department of
Psychiatry, USA.
3University of Texas Southwestern Medical Center, Department of

cord.4 Further, pediatric demyelinating disorders are
commonly associated with a constellation of highly com-
plex psychosocial manifestations.5 Problems with cogni-
tive and psychological functioning in pediatric TM have
been described anecdotally.3 The paucity of systematic
Neurology and Neurotherapeutics, USA.
Corresponding author:
Lana L Harder, Childrens Medical Center Dallas, 6300 Harry Hines
Blvd, Suite 900, Dallas, Texas 75235, USA.
Email: lana.harder@childrens.com
948 Multiple Sclerosis Journal 19(7)
Methods
Participants
Twenty-four consecutive pediatric idiopathic TM patients,
aged 5 to 18 years, were evaluated during a visit at the
Childrens Medical Center Dallas Pediatric Demyelinating
Diseases Clinic over a period of 29 months. One patient had
experienced recurrent TM. Patients who met criteria for
ADEM, MS, or neuromyelitis optica (NMO) were excluded.
Age of onset ranged from 1 to 17 years, with a mean of 9.67
years. All were at least 30 days from acute symptoms and/or
any acute therapeutic intervention (i.e. steroids); however,
approximately 46% of participants were prescribed medica-
tion (i.e. selective serotonin reuptake inhibitors (SSRIs),
GABAergic medications, tricyclic antidepressants, anticho-
linergic agents). Time since symptom onset ranged from one
to 135 months with a mean of 22 months.
All subjects for whom imaging data were available (n =
20) had normal brain magnetic resonance images (MRIs)
except one patient, who had one non-specific T2 hyperin-
tense lesion not characteristic of ADEM or MS. Four
patients were remote from their diagnosis years earlier at
outside institutions and thus brain imaging was unavailable
for review. MRI scans of the cervical and thoracic spine
were available for all but one patient. Fifty-four percent of
patients had longitudinally extensive lesions defined as a
lesion spanning greater than or equal to three vertebral seg-
ments; however, NMO-immunoglobulin G (IgG) testing
was negative in these patients. Cervical lesions were pre-
sent in 54% of TM patients. Forty-six percent of patients
had normal ambulation, 17% had an abnormal gait but
ambulated independently, 29% required bilateral support
(i.e. crutches), and 8.3% were wheelchair bound.
Procedure
This project was a retrospective study approved by the
Institutional Review Board (IRB) at the University of Texas
Southwestern Medical Center. Each participant completed
a brief 60-minute neuropsychological screening evaluation
administered by a trained examiner as part of his or her
clinical evaluation. Additionally, parents completed ques-
tionnaires related to their childs functioning in the areas of
psychological and school functioning. Caregivers as well
as patients completed questionnaires related to the patients
experience of fatigue. Clinical information was recorded
including details regarding diagnosis (i.e. age at onset),
medication use, and ambulation.
Measures
Neuropsychological evaluation. A brief neuropsychological
battery was administered to assess multiple domains. Multi-
ple qualified examiners administered assessments. There
were slight variations in tests administered based on the age
of the participant and available age-based normative data for
each measure. Ages of participants for whom normative data
were available are listed by test below. Of note, two patients
had no use of their hands and thus were not administered
tasks requiring fine-motor skills; in those cases, no hand
preference was indicated. Verbal learning and memory was
assessed with the California Verbal Learning Test, Childrens
Version (CVLT-C; 516 years)7 and CVLT, Second Edition
(CVLT-II; 1718 years).8 Auditory attention and working
memory were assessed with the Digit Span subtest from the
Wechsler Intelligence Scale for Children, Fourth Edition
(WISC-IV; 616 years)9 and Wechsler Adult Intelligence
Scale, Third Edition (WAIS-III; 1718 years).10 Motor-based
processing speed was assessed with the Symbol Search sub-
test from the WISC-IV/WAIS-IV. The Symbol-Digit Modali-
ties Test (SDMT; 818 years), Oral Version,11 provided a
measure of motor-free processing speed. Visual-motor inte-
gration and visual perception were assessed with the Beery
Developmental Test of Visual-Motor Integration, Fifth Edi-
tion (VMI-5; 518 years).12 The Grooved Pegboard13 (518
years) was used to assess bilateral fine-motor speed and
coordination. Measures of simple and complex attention and
sequencing were assessed with the Trail-Making Test (TMT;
918 years).14 Verbal fluency was measured with the Letter
Fluency subtest from the Delis-Kaplan Executive Function
System (DKEFS; 818 years).15
Assessment of emotional functioning, fatigue, and school perfor-
mance. At the time of the evaluation, patients and caregiv-
ers completed the Pediatric Quality of Life (PedsQL; 518
years) Multidimensional Fatigue Scale.16 Caregivers com-
pleted the Behavior Assessment System for Children, Sec-
ond Edition (BASC-2; 518 years)17 and a non-standardized
school history questionnaire, which included a Likert-type
rating of academic functioning that allowed the parent to
rate their childs performance in reading/language arts, his-
tory/social studies, math, and science on the following
scale: failing, below average, average, above average.
Mobility assessment. The Hauser Ambulation Index is an
established rating scale that assesses mobility based on
time to walk 25 feet and need for assistance. Ratings range
from 0, indicating normal ambulation, to 9, indicating that
the patient is restricted to a wheelchair and unable to trans-
fer independently.
Statistical analyses
Descriptive data was generated using the Statistical Package
for the Social Sciences, version 18.0.
Results
Demographic information and clinical data are presented in
Table 1. Age of participants at the time of the evaluation
Harder et al. 949

Table 1. Medical and demographic information. moderate, and severe deficits were revealed in 40.9%,
Variable Mean SD Range 22.7%, and 13.6% of participants, respectively. In the area
of graphomotor (i.e. paper/pencil) skills, mild, moderate,
Age (years) at onset 9.67 4.84 117 and severe deficits in visual-motor integration were
Age (years) at evaluation 11.46 3.36 518 observed in 28.6%, 19%, and 4.8%, respectively. Across
Time since symptom onset (months) 22.13 39.37 1135 areas assessed, the lowest rate of impairment was observed
Hauser Ambulation Index 2.58 2.90 09 in the area of visual perception, with only 4.5% of the
Variable n %
cohort falling in the mildly impaired range and none falling
On medicationa 11 45.80
into moderate or severe ranges.
Cervical 13 54.20
Longitudinally extensive lesionb 13 54.20
Female 15 62.50 Attention and executive function
Right handed 21 87.5
Race/ethnicity Approximately 18.2% of participants showed moderate
White 15 62.50 impairment in auditory attention and working memory (i.e.
Black or African-American 1 4.20 Digit Span), while 40.9% had at least a mild deficit. Parents
Hispanic or Latino 6 25.00 reported at-risk or subclinical attention problems in
American Indian or Alaska Native 1 4.20 approximately 30% of participants. Interestingly, no
Asian 1 4.20 patients demonstrated severe problems in this area, nor did
aIncluded
participants parents report clinically significant attention
medication known to have potential cognitive impairment
(i.e. selective serotonin reuptake inhibitors (SSRIs), GABAergic problems. While 5.6% of participants showed severe defi-
medications, tricyclic antidepressants, and anticholinergic agents); cits in simple attention (i.e. Trail-Making Test A), 11.1%
bdefined as lesion spanning greater than or equal to three cord segments.
showed severe deficits in complex attention and sequenc-
ing (i.e. Trail-Making Test B). Mild and moderate deficits
were noted in the area of verbal fluency for 25% and 20%
ranged from 5 to 18 years, with a mean of 11 years. Sixty- of the sample, respectively.
three percent were female and approximately 88% were
right handed. Most participants were White (62.5%) fol-
Verbal memory
lowed by 25% who were Hispanic or Latino.
Results of the neuropsychological screening evaluation In the area of verbal memory, deficits in initial free recall
show that all mean scores for the group were within the (i.e. CVLT-C/II Trial 1) were mildly and moderately
average range with the exception of fine-motor speed and impaired at rates of 33.3% and 20.8%, respectively. Rates
dexterity with the non-dominant hand, which was slightly of deficits in free recall generally decreased with opportuni-
below average. Since scores ranged considerably among ties for rehearsal of the information (i.e. CVLT-C/II Trial 5),
this cohort, mean scores as well as ranges and percentages with 12.5% falling in the mildly impaired range and 4.2%
of the sample falling at or below 1.0, 1.5, and 2.0 standard falling in the severely impaired range. Long delay free
deviations (SD) from the mean are presented in Table 2. recall was mildly, moderately, and severely impaired in
Scores are presented as standard scores (mean = 100; SD = 25%, 8.3% and 4.2% of participants, respectively.
15), scaled scores (mean = 10; SD = 3), or T-scores (mean
= 50; SD =10).
Processing speed
For the purposes of describing performance on standard-
ized performance-based measures, scores at or below 1.0 Timed tasks were administered to measure motor-based
SD from the mean, or the 16th percentile, are described as processing speed (i.e. Symbol Search) and non-motor pro-
mildly impaired. Scores at or below 1.5 SD, or at approxi- cessing speed (i.e. SDMT). Mild and moderate deficits in
mately the fifth percentile, are moderately impaired. Scores motor-based processing speed were observed at rates of
falling at or below 2 SD from the mean (second percentile) 20% and 10%, respectively. When the motor component
are considered severely impaired. was removed, moderate and severe impairment were
observed at rates of 10% and 5%, respectively.
Fine-motor and visual-motor skills
Additional clinical and psychosocial factors
Within this cohort, the highest frequency of scores falling
below the mean was in the area of bilateral fine-motor Data regarding parent-report of symptoms of depression
speed and dexterity, with a lower mean score observed for and attention problems, parent and self-report of fatigue,
the non-dominant hand. For the non-dominant hand, mild school performance, and need for further testing are pre-
deficits were observed in 45.5% of subjects, with severe sented in Table 3. According to parent report on the BASC-
deficits noted in 36.4%. For the dominant hand, mild, 2, TM subjects experienced subclinical (21.7%) or clinical
950 Multiple Sclerosis Journal 19(7)

Table 2. Neuropsychological characteristics.

Variable n Mean SD Range 1 SD 1.5 SD 2 SD
VMI SS 21 88.76 10.19 68106 28.60 19.00 4.80
VMI Visual Perception SS 22 98.73 9.87 84117 4.50 0 0
Grooved Pegboard
Dominant SS 22 90.27 18.72 55116 40.90 22.70 13.60
Non-dominant SS 22 82.41 21.18 55116 45.50 36.40 36.40
Digit Span ScS 22 8.55 2.20 513 40.90 18.20 0
Symbol Search ScS 20 9.75 2.79 517 20.00 10.00 0
Letter Fluency ScS 20 9.35 2.74 514 25.00 20.00 0
Trail-Making Test A SS 18 100.39 13.81 55116 5.60 5.60 5.60
Trail-Making Test B SS 18 101.00 17.48 55122 11.10 11.10 11.10
SDMT SS 20 102.05 17.08 61142 10.00 10.00 5.00
CVLT-C/II
Total 15 TS 24 49.17 11.58 2165 29.20 12.50 4.20
Trial 1 SS 24 95.92 13.68 78115 33.30 20.80 0
Trial 5 SS 24 103.71 16.24 55130 12.50 4.20 4.20
Long Delay Free SS 24 99.25 14.44 55115 25.00 8.30 4.20
Recognition SS 24 97.71 14.34 70115 25.00 16.70 12.50

Scores are presented as standard scores (SS) with mean = 100 and SD = 15; scaled scores (ScS) with mean = 10 and SD = 3; or T-scores (TS) with
mean = 50 and SD =10 1, 1.5 and 2 standard deviation (SD) columns indicate percentage of sample at or below those levels;VMI:Visual-Motor Inte-
gration; SDMT: Symbol-Digit Modalities Test; CVLT-C:/II California Verbal Learning Test, Childrens Version, second edition.

Table 3. Clinical and psychosocial characteristics.

Variable n % Mean SD Range
School problems 8 33.30
Referral for further testing 7 29.20
Depression 53.78 11.42 3782
At-risk 5 21.70
Clinically significant 2 8.70
Attention problems 50.13 10.04 3665
At-risk 7 30.40
Clinically significant 0 0
Sleep/rest fatigue Parent 66.86 15.19 4292
Mild 9 42.90
Severe 6 28.60
Self 66.76 17.29 38100
Mild 8 38.10
Severe 0 0
General fatigue Parent 57.81 20.47 488
Mild 7 33.30
Severe 11 52.40
Self 62.29 19.60 2196
Mild 3 14.30
Severe 9 42.90

Depression and attention problems are based on Behavior Assessment System for Children, Second Edition (BASC-2) Parent report; fatigue scores
are derived from the Pediatric Quality of Life (PedsQL) Multidimensional Fatigue Scale.

(8.7%) levels of depression in a total of 29% of participants.
Participants and their parents completed rating forms on the
patients experience of fatigue during the screening evalua-
tion. Parent report indicated mild and severe general fatigue
(33.3% and 52.4%, respectively) and mild and severe sleep-
related fatigue (42.9% and 28.6%, respectively). Self-report
of general fatigue indicated rates of 14.3% and 42.9% mild
and severe fatigue, respectively. Thirty-eight percent of par-
ticipants indicated mild sleep-related fatigue and none
reported severe fatigue in this area. Across the sample,
Harder et al. 951
parent report indicated school problems, as defined by
below average or failing performance in at least one of four
academic domains, in 33% of participants.
As a marker of cognitive dysfunction, data were col-
lected indicating which participants were referred for addi-
tional testing (i.e. full neuropsychological evaluation)
based on poor performance on the neuropsychological
screening evaluation. Of the 24 participants, seven (29%)
patients were referred for a full neuropsychological evalua-
tion. School problems were reported in all but one of the
participants referred for further testing (86%).
To examine the possible role of depression, medica-
tion, and fatigue in cognitive functioning, qualitative
review of the data was conducted. Of the seven partici-
pants who showed elevated symptoms of depression
based on parent report, two received referrals for addi-
tional cognitive testing. Likewise, the role of medication
was examined. Of the 11 participants who were prescribed
medication (i.e. SSRIs, GABAergic medications, tricyclic
antidepressants, and anticholinergic agents), only two
were referred for further cognitive testing. In this cohort,
results do not suggest a clear association between cogni-
tive dysfunction and mood-related factors or adverse
medication side effects. In contrast, mild to severe symp-
toms of fatigue were reported in six out of seven partici-
pants (86%) referred for further cognitive testing. These
results suggest that fatigue may play a meaningful role in
performance on cognitive testing.
Discussion
The current study describes neuropsychological function-
ing in a cohort of 24 pediatric patients with TM. Although
mean scores were largely within the average range, partici-
pant performance varied widely within specific domains.
Varying rates of impairment were described across areas of
fine-motor and visual-motor skills, verbal memory, atten-
tion and executive function, and processing speed. The
highest frequency of impairment was noted in the area of
bilateral fine-motor coordination, which is not surprising
given motor impairment associated with TM. Beyond
motor-based functioning, results showed that attention and
memory problems are present in up to 40.9% and 33% of
TM patients, respectively. It should be noted that these
problems did not typically reach the severe range of impair-
ment. Consistent with this finding, 30.4% of parents
endorsed only subclinical attention problems. Difficulties
occurring at a slightly lower rate included those related to
processing speed and verbal fluency. Findings are generally
consistent with research on pediatric CNS demyelinating
diseases (i.e., MS, ADEM), which has identified cognitive
problems in the areas of attention, memory, and processing
speed.4,18 Evidence of these cognitive deficits is detected in
this cohorts rate of reported school problems. Thirty-three
percent of patients experienced school problems. Notably,
all but one of the patients referred for a full neuropsycho-
logical evaluation experienced school difficulties.
This study also begins to qualitatively explore clinical
factors that may contribute to performance on cognitive
testing such as depression, medication, and fatigue. When
using a referral for further testing based on poor perfor-
mance on the screening battery as an indicator of cognitive
dysfunction, there was no clear relation between cognitive
problems and depression or medication use. In contrast,
parent and self-reported fatigue appeared to be present in
most of the TM participants who were referred for further
evaluation, suggesting that fatigue may play an important
role in cognitive dysfunction. This would not be surprising
given previous research suggesting that patient report of
fatigue is associated with report of cognitive problems.19
Further exploration of factors such as mood, medication
use, and fatigue is warranted in this population to better
understand the role they play in cognitive functioning,
which will inform targeted interventions.
Previous research has identified unique profiles of CSF
inflammation in patients with TM.20 Given the absence of
obvious cerebral pathology, cognitive deficits were not
expected in TM patients. Yet, in our cohort, varying levels of
cognitive impairment were observed. Theoretically, circulat-
ing components of the inflammatory cascade could produce
cerebral damage without obvious lesions on MRI. Previous
research has identified cognitive dysfunction in patients with
isolated optic neuritis, 24 to 31 years from the demyelinating
event, even in patients with no obvious cerebral demyelina-
tion.21 Studies of MS patients have previously recognized
the inability of conventional MRI to identify cortical pathol-
ogy.22 Thus, all TM patients should be screened for cerebral
pathology, but even in the absence of MRI changes, neu-
ropsychological testing should be completed.
Findings of the current study highlight the need for regu-
lar multi-disciplinary surveillance and treatment of pediat-
ric patients with any CNS demyelinating disease.
Historically, cognitive testing has been restricted to patients
with obvious cerebral involvement, but such a paradigm
would result in missing those TM patients exhibiting
impairment who would be likely to benefit from academic
and/or therapeutic intervention services. This study is the
first to describe cognitive functioning in pediatric patients
with idiopathic TM.
Limitations of this study include the sample size. Given
that pediatric TM is an extremely rare disorder, specialized
clinics provide an opportunity to study patient outcomes.
Future studies should involve multiple sites to allow for
larger cohorts to address this limitation. Another limitation
of the study surrounds the understanding of patients pre-
morbid functioning. Given the retrospective nature of the
current study, information regarding premorbid status was
not available. Of note, capturing this type of information
would have been virtually impossible for approximately
one quarter of the participants in this study given that TM
952 Multiple Sclerosis Journal 19(7)
onset occurred prior to traditional school age and problems,
such as those related to attention and/or learning, may have
been difficult to detect at such an early age. Future prospec-
tive studies of TM and cognitive functioning should attempt
to address this limitation by collecting data to determine
whether cognitive dysfunction pre-dated the demyelinating
event. Such information would help clarify the role of CNS
inflammation on cognitive outcomes.
Funding
This research received no specific grant from any funding agency
in the public, commercial, or not-for-profit sectors.
Conflict of interest
The authors declare that there are no conflicts of interest.
References
1. Kerr DA, Krishnan C and Pidcock F. Acute transverse myeli-
tis. In: Singer HS, Kossoff EH, Harman AL, et al. (eds) Treat-
ment of pediatric neurologic disorders. Boca Raton: Taylor
and Francis, 2005: pp.445451.
2. Pidcock FS, Krishnan C, Crawford TO, et al. Acute transverse
myelitis in childhood: Center-based analysis of 47 cases. Neu-
rology 2007; 68: 14741480.
3. Trecker CC, Kozubal DE, Quigg M, et al. Quality care in
transverse myelitis: A responsive protocol. J Child Neurol
2009; 24: 577583.
4. Deery B, Anderson V, Jacobs R, et al. Childhood MS and
ADEM: Investigation and comparison of neurocognitive fea-
tures in children. Dev Neuropsychol 2010; 35: 506521.
5. MacAllister WS, Boyd JR, Holland NJ, et al. The psychoso-
cial consequences of pediatric multiple sclerosis. Neurology
2007; 68: S66S69.
6. Frohman EM and Wingerchuk D. Transverse myelitis. N Engl
J Med 2010; 363: 564572.
7. Delis DC, Kramer JH, Kaplan E, et al. California verbal
learning test, childrens version. San Antonio, TX: The Psy-
chological Corporation, 1994.
8. Delis DC, Kramer JH, Kaplan E, et al. California verbal
learning test, second edition. San Antonio, TX: The Psycho-
logical Corporation, 2000.
9. Wechsler D. Wechsler intelligence scale for childrenfourth
edition. San Antonio, TX: Pearson, 2003.
10. Wechsler D. 2008. Wechsler adult intelligence scalethird
edition. San Antonio, TX: Pearson, 1997.
11. Smith A. Symbol digit modalities test. Los Angeles: Western
Psychological Services, 1982.
12. Beery KE and Beery NA. The Beery-Buktenica developmen-
tal test of visual motor integration, fifth edition. San Antonio,
TX: Pearson, 2006.
13. Lafayette Instrument. Grooved pegboard test. Lafayette, IN:
Lafayette Instrument, 2002.
14. [United States] Army Individual Test Battery: Manual of
directions and scoring. Washington, DC: War Department,
Adjutant Generals Office, 1944.
15. Delis DC, Kaplan E and Kramer JH. Delis-Kaplan executive
function system. San Antonio, TX: The Psychological Corpo-
ration, 2001.
16. Varni JW, Seid M and Rode CA. The PedsQL: Measurement
model for the pediatric quality of life inventory. Med Care
1999; 37: 126139.
17. Reynolds CR and Kamphaus RW. BASC-2: Behavior assess-
ment system for children, second edition. Circle Pines, MN:
American Guidance Service, 2004.
18. Amato MP, Goretti B, Ghezzi A, et al. Cognitive and psy-
chosocial features of childhood and juvenile MS. Neurology
2008; 70: 18911897.
19. MacAllister WS, Christodoulou C, Troxwell R, et al. Fatigue
and quality of life in pediatric multiple sclerosis. Mult Scler
2009; 15: 15021508.
20. Kaplin A, Deshpande D, Scott E, et al. IL-6 induces regionally
selective spinal cord injury in patients with the neuroinflam-
matory disorder transverse myelitis. J Clin Invest 2005; 115:
27312741.
21. Nilson P, Rorsman I, Larsson EM, et al. Cognitive dysfunc-
tion 2431 years after isolated optic neuritis. Mult Scler 2008;
14: 913918.
22. Kidd D, Barkhof F, McConnell R, et al. Cortical lesions in
multiple sclerosis. Brain 1999; 122: 1726.
Reproduced with permission of the copyright owner. Further reproduction prohibited without
permission.