Neuropsychological
Outcome
of Pediatric
Liver
Transplantation
Sunita
M. Stewart,
PhD*f;
Cheryl
Hiltebeitel,
PhD*;
Janice
Nici,
PhD*;
David
A. Wailer,
MD*t;
Ricardo
Uauy,
MD, PhD*t;
and
Walter
S. Andrews,
MD**
From the *universfty of Texas Southwestern Medical Center at Dallas and Chlldren’s Medical Center of Dallas
ABSTRACT. Children with end-stage liver disease who undergo liver transplantation may have unrecognized
neuropsychological and academic deficits, for which re-mediation programs may be available. Intellectual,
aca-demic, and neuropsychological measures of 28 pediatric
patients who had received successful liver transplanta-tion at least 1 year previously were compared with those
of 18 patients with cystic fibrosis (to control for effects of growth retardation and chronic illness) matched for
age, age at diagnosis, physical growth, and parents’ socio-economic status. Liver transplant patients had signifi-cantly lower scores on nonverbal intelligence tests (mean
± SD for liver transplant vs cystic fibrosis patients: 89.1
± 19.1 vs 105.8 ± 17.6), lower academic achievement, and
lower zscores for age in the areas of learning and memory (-0.68 ± 1.09 vs 0.19 ± 1.24), abstraction and concept formation (-1.73 ± 1.58 vs -0.79 ± 1.37), visual-spatial function (-0.66 ± 1.09 vs 0.10 ± 0.69), and motor function (-0.13 ± 0.85 vs 0.36 ± 0.57). No differences were found
on tests of verbal intelligence, or in alertness and
concen-tration, perceptual-motor, and sensory-perceptual areas.
Cyclosporine levels were found to correlate positively
with motor speed (r = .41, P < .05). Thorough psycho-educational and neuropsychological evaluations should
be considered for pediatric patients who receive liver
transplantation to allow these children to maximize their potential. Pediatrics 1991;87:367-376; pediatric liver transplantation, neuropsychological function, intelligence
quotient, academic achievement.
Assessment of neuropsychological function in children with medical illnesses affecting the brain may provide insights into learning disabilities and
other problem areas of psychological function that
Received for publication Dec 15, 1989; accepted Mar 1, 1990. Reprint requests to (S.M.S.) Dept of Psychiatry, 5th Floor, Children’s Medical Center, 1935 Motor St, Dallas, TX 75235.
PEDIATRICS (ISSN 0031 4005). Copyright © 1991 by the
American Academy of Pediatrics.
are not apparent from standard general measures
of intelligence. The definition of cognitive problems
in specific neuropsychological terms may permit
more appropriate expectations of achievement from
the child, as well as the development of remediation programs specifically targeted for the child’s
diffi-culties. This is especially important as pediatricians focus on “quality of life” issues for children who are
now able to survive diseases that previously resulted in death.
With the advent of liver transplantation, life
expectancy for children with end-stage liver disease
has dramatically improved. While there have been
recent studies comparing the intellectual function of children with liver disease before and after trans-plantation,’4 there are no follow-up investigations
of neuropsychological status. Information on
neu-ropsychological function in this patient population
is important for at least three reasons. First, hepatic
encephalopathy is a frequent sequela of chronic
liver disease, and it may leave subtle deficits in
brain function that are more likely to be detected
by neuropsychological assessment. Second,
neuro-psychological deficits have been documented in
adults with liver disease even when encephalopathy
is not clinically evident.5” Finally, cyclosporine,
an essential antirejection treatment following
transplantation, has well-documented adverse
den-tral nervous system effects when levels are high.12
The impact of normal maintenance doses on
neu-ropsychological function in children has not yet
been examined.
Gross intellectual deficits have been found in children with end-stage liver disease before and
after transplantation,3’ “ but not all children with
liver disease show such deficits. Two important
chil-dren who display mental delays is that they are more likely to have had early onset of liver disease and significantly greater growth retardation.1’ 2
Fol-lowing transplantation, the relationship between
early onset of liver disease and delayed intellectual
function persists.’ The negative impact of early
onset of chronic illness and growth retardation on
mental function has been shown in other
popula-tions.’4 In studies evaluating mental function in
children with liver disease, chronic illness and
growth retardation are confounding factors, and the contribution of these variables to the deficits meas-ured has not been previously parceled out.
When neuropsychological test results of adults
with chronic liver disease are compared with those of control subjects, those with liver disease more often have deficits in visual-spatial and perceptual-motor skills, but their verbal test results and global measures of intellectual function are equivalent to
those of control subjects.” Previous studies have
used only global measures of intellectual function
when evaluating children with liver disease. A
ma-jor limitation of these measures is that patterns of
strengths and weaknesses are missed, and patients
who have subtle deficits of the kind found in adults with liver disease appear to be intact.
This study aims to examine intellectual,
ada-demic, and comprehensive neuropsychological
function in children following liver transplantation,
compared with a control group of chronically ill
children with an equivalent age of onset of disease
and degree of growth retardation. If the control
group shows deficits equivalent to those of children with a history of liver disease, then it is possible
that a substantial proportion of the deficits
re-ported in past studies of children with liver disease
may be secondary to chronic illness and growth
retardation. Conversely, if the control group does not show the same degree of intellectual delay, then the deficits are more likely to be specific to liver
disease. If children with a history of liver disease
show the same pattern of neuropsychological
dys-function as occurs in adults with liver disease, this
finding could have important implications. Deficits
in the areas of perceptual-motor and visual-spatial
function could make these children vulnerable to
the development of learning disabilities. Early de-tection of this vulnerability could result in early intervention with the possibility of remediation for future patients.
METHODS
Subjects
The experimental group consisted of 14 male and
14 female pediatric patients who had received liver
transplantation between 1 and 4 years prior to
participation in the study (mean ± SD = 26.2 ±
12.3 months, range = 12 through 48 months). The
liver transplant population spans a broad age
grouping. Although it would be methodologically ideal for all subjects to receive the same tests, the numbers of these patients available within a narrow age grouping are quite limited at any single
trans-plant center. Thus a methodological decision had
to be made that would minimize the heterogeneity
of tests while allowing a large enough group to be
included to achieve meaningful statistical analyses. Our lower age restriction was instated to allow the administration of more specific neuropsychological
measures than is possible at very young ages. The
upper age restriction corresponds to the age limit
of the Halstead-Reitan Neuropsychological Test
Battery for Children. Patients were recruited into the study if they were at least 4 and no more than
14 years old at the time of participation (mean ±
SD = 7.6 ± 2.5 years, range = 4.6 to 13.4 years). While this group is still more heterogeneous in age (and therefore in test taken) than ideal, we made an effort to match this group closely to the control group in relation to age, so that proportionately equal numbers of children within the experimental and control groups would be taking the same tests.
All patients had received liver transplantation at
Children’s Medical Center of Dallas and were re-cruited into the study at the time of annual
reeval-uation following transplantation. Subjects were
considered for inclusion in the study if they had
received liver transplantation at least 12 months prior to the study. Exclusion criteria were factors independent of liver disease that could affect
cog-nitive function: gestational age of less than 34
weeks or birth weight of less than 2000 g; history of perinatal asphyxia; and coexisting diseases such
as cystic fibrosis, renal disease unrelated to hepatic
disease, significant heart disease, or endocrine
dis-ease. Four patients who had seizures during the
posttransplant period underwent magnetic
reso-nance imaging scans. Three of these patients had
evidence of structural neurological damage, ie, brain
infarcts evident on magnetic resonance imaging,
and were not included in our study. The rationale
for exclusion of these patients is that they were
likely to have numerous neuropsychological diffi-culties secondary to factors independent of liver disease per se (eg, the location and size of the
infarct), and it would have been difficult to separate
these problems from those that are of central inter-est in this study.
Diagnoses of patients prior to transplantation
were as follows: extrahepatic biliary atresia (12
intrahepatic biliary hypoplasia (syndromatic type) (4 patients), chronic active hepatitis (2 patients), progressive cirrhosis of unknown cause (1 patient),
hepatoma (1 patient), and hamartoma of the liver
(1 patient). All children were medically stable at
the time of study participation and were tested as
outpatients. Table 1 presents liver and renal
func-tion test values at the time of study participation.
All patients received immunosuppressive therapy consisting of both prednisone and cyclosporine,
administered at the lowest possible levels consistent
with avoiding rejection as well as minimizing side
effects. The prednisone dosage (mean ± SD) was
0.13 ± 0.05 mg/kg per day (range 0.04 to 0.29 mgI
kg per day). The cyclosporine dosage (mean ± SD)
was 11.1 ± 2.9 mg/kg per day (range 5.4 to 17.3 mgI
kg per day). Cyclosporine levels, which were
meas-ured by high-pressure liquid chromatography, were
228 ± 95.9 g/L (range 96 to 521 g/L). One patient
had a cyclosporine level in the toxic range (higher than 400 g/L), measured the day following
partic-ipation in the study. However, on physicians’s
ex-amination for routine purposes on the day of
test-ing, she was not found to have a clinically toxic
cyclosporine level, and therefore her scores were
included in our analyses. Her cyclosporine level 3
days later was in the normal range (265 g/L). The
three excluded subjects had cyclosporine levels
ranging from 86 tg/L to 138 g/L.
The control group consisted of 11 male and 7
female patients who had a diagnosis of cystic
fibro-sis. These patients were recruited from the Chil-then’s Medical Center cystic fibrosis clinic.
Inclu-sion criteria were (1) diagnosis of cystic fibrosis, and (2) present age between 4 and 14 years.
Exclu-sion criteria were perinatal factors that could in-dependently affect cognitive function, coexisting
disease, and impairment in liver function secondary
to cystic fibrosis. As the control data were gathered
after the experimental group had been tested, the
mean age of diagnosis of the liver transplant
pa-tients was known, and an attempt was made for the
control patients to approximate the experimental group on this variable. All children were medically
TABLE 1. Liver and Renal Function Test Results of 28 Liver Transplant Patients at Time of Study Participation
Measure/Normal Value Mean ± SD Range
Total bilirubin, mg/dL/0-1.5 0.81 ± 0.71 0.3-3.4
Creatinine, mg/dL/0.3-1.3 0.81 ± 0.26 0.4-1.5
‘y-Glutamyl transpeptidase, 75 ± 127 11-614
IU/6-44
Aspartate aminotransferase, 72 ± 88 14-429
IU/8-37
Alanine aminotransferase, 60 ± 82 8-397
IU/8-36
stable at the time of study participation. All cystic
fibrosis patients received a physical evaluation
within 7 days of study participation and were
graded on the modified Schwachman Scale for
cys-tic fibrosis.’5 This scale provides rating criteria on
a scale from 0 through 25, with scores 0 through 5
corresponding to “very poor,” 6 through 10 to
“poor,” 1 1 through 15 to “fair,” 16 through 20 to
“good,” and 21 through 25 to “very good,” for four variables related to progression of cystic fibrosis: (1) activity based on case history, (2) pulmonary
findings and cough, (3) growth and nutrition, and
(4) roentgenogram. Mean ± SD and range for the
group on each variable were as follows: activity,
21.3 ± 3.3 (16 through 25); pulmonary findings, 18.4
± 4.6 (9 through 25); growth and nutrition, 18.8 ±
4.8 (11 through 25); and roentgenogram, 18.4 ± 4.2
(11 through 25).
Children with cystic fibrosis were chosen as
con-trol subjects because they have a history of
life-threatening chronic illness, growth retardation, and varying ages of onset of symptoms leading to
diag-nosis. Comparison of age at evaluation, age at di-agnosis, and growth measures of the two groups is
presented in Table 2. Statistical analyses indicate that the experimental group was not different from
the control group in gender distribution (Fisher
Exact Test, P > .33), age at evaluation, and age at diagnosis. On measures of growth, control patients
were either equally growth retarded or else more
discrepant than were the experimental group from
the healthy-children medians for age and gender.
An additional potentially confounding variable is
socioeconomic status. Members of high
socioeco-nomic groups perform better on the average on tests
of intelligence than do members of low
socioeco-nomic groups.’6 Socioeconomic scale positions
based on the Hollingshead Indices of Social
Position’7 were determined for each family. Both
groups had families that fell into each of the seven scale positions. Four mothers in the liver transplant group and mothers of three children in the cystic fibrosis group did not have a high school education.
This difference was not significant for the two
groups (Fisher Exact Test, P > .74).
Measures
The following measures were obtained from all
subjects.
Psychological Testing. Psychological tests
in-cluded intellectual, academic, and specific
neuro-psychological tests. Testing was conducted by
trained technicians who were “blind” to the
hy-potheses of the study but not to group membership
se-TABLE 2. Age at Evaluation, Age at Diagnosis, and Growth Measures in 28 Liver Transplant vs 18 Cystic Fibrosis Subjects*
* Values are given as mean ± SD (range).
t Analysis of variance.
::Percent of median for age and gender.
§Significant at the .05 level.
Measure
Age at evaluation, y Age at diagnosis, mo
Weights
Heights
Weight for heights Head circumferences
Triceps skin-fold thicknes4 Mean arm circumferences Arm muscle circumferences
Liver Transplant Group
7.7 ± 2.5 (4.6-13.4)
9.9 ± 20.9 (1-101)
95.2 ± 17.9 (53-145)
95.5 ± 5.4 (86-111) 105.8 ± 12.1 (80-138)
99.5 ± 3.0 (92-105) 82.5 ± 36.6 (50-223)
101.4 ± 11.5 (79-127) 104.9 ± 9.7 (84-122)
Cystic Fibrosis Group
7.8 ± 2.4 (4.8-13.3)
9.6 ± 24.7 (0-108)
88.0 ± 15.9 (57-133)
96.4 ± 3.5 (89-104)
94.9 ± 11.0 (76-119) 98.5 ± 4.1 (92-109)
63.6 ± 32.7 (32-175) 92.3 ± 11.8 (64-118) 98.7 ± 8.9 (79-111)
Pt
.85
.96
.18 .50
.004
.37 .09 .02 .04
lected to measure several different categories of
abilities. There are several different possible models that conceptualize and categorize brain functions. All include a combination of the following skills: verbal, nonverbal (or visual-spatial), motor, cogni-tive processes such as abstraction, concept forma-tion, and problem solving, attention or
concentra-tion, and sensory-perceptual skills important in
input of information (vision, hearing, and touch).
The groupings used in this study are based on
Reitan and Wolfson’s approach.’8 In addition, the
perceptual-motor category has been added to better
compare the findings in our subjects with those
from adults with liver disease. The perceptual-mo-tor category overlaps with the visual-spatial date-gory both theoretically and in tests used to assess function. The rationale used for assignment of tests into one of these two categories was that tests with
a perceptual but not visual component were
in-cluded only in the perceptual-motor category, and
tests with a minimal motor component were
in-cluded only in the visual-spatial category.
The tests administered and the categories of
abil-ity they measure are presented in Table 4. Selected tests will be described in more detail inasmuch as
their distribution profiles will be presented under
“Results.” The Target Test requires the subject to
reproduce designs connecting dots, presented
visu-ally, following a 3-second delay. The score is based on the number of designs that are correctly
repro-duced. The Test of Visual-Motor Integration
con-sists of geometric designs to be reproduced by the
subject using paper and pencil. The test is scored
for accuracy of reproduction. The Marching Test
requires that the subject track a connected series of circles. Part of the test requires alternating
movements of both hands, following the sequence
on the left side of the page with the left hand, and right side of the page with the right hand. This part of the test is scored for number of circles correctly
contacted. The Tactual Performance Test consists
of a formboard into which the subject places blocks
while blindfolded. The memory component of this
test measures incidental learning as at the end of
the test the subject is asked to draw the board and
the blocks from memory. This part of the test is
scored for number of blocks correctly remembered. The intelligence and academic tests yield scores
that are standardized for age, with the normal
population having a mean of 100 and SD of 15. The
neuropsychological tests yield raw scores, which
were converted to z scores for age for each subject,
based on available norms for healthy children.192#{176}
z Scores are standardized scores, obtained by the
following formula: z = (raw score - mean for age)/ SD for age. This conversion allows the individual
child’s score to be compared with those of his or
her age-group of normal children, who have a mean
score of 0 and an SD of 1. For the
sensory-percep-tual category and for the Grip and Name Writing
Tests, no age-normed means and SDs are available. For these tests, the z score was based on the corn-bined group mean and SD of the liver transplant and cystic fibrosis patients. On all tests where lower
scores signify better performance (such as where
the score represents the number oferrors on a task),
the scales were reversed when the z scores were
obtained.
The academic tests have norms only for children
older than 5 years of age, and so they were not
administered to younger children. The analyses
represent the scores from 24 liver transplant
pa-tients and 14 cystic fibrosis patients. Some of the
specific neuropsychological measures were also
used only for the “younger” (ages up to and
includ-ing 8 years) or “older” (9 years and older) children as required by the Halstead-Reitan battery
stand-ardization. These tests are indicated in Table 4.
Twenty-one children in the liver transplant group
younger than age 9, and 7 in the liver transplant
group and 3 in the cystic fibrosis group were older.
One test proved particularly difficult for some of
the younger children, even though norms do exist
for the age population. This was the Tactual
Per-formance Test, which requires that the child be
blindfolded during the task. Six children in the liver
transplant group and 4 children in the control group
could not tolerate the blindfold. The test was
dis-continued in these cases and their scores were not
included in the analyses. Thus the Tactual
Per-formance Test results are based on scores from 22
patients in the liver transplant group and 14
pa-tients in the cystic fibrosis group. Two children in
the liver transplant group and 1 in the cystic
fibro-sis group had no previous preschool experience, and they were not administered the name-writing tests.
One child in the liver transplant group was not
administered the sensory-perceptual examination
as her parents did not return her to the test session
following a lunch break.
The intelligence and academic tests are widely
used measures with well-documented reliability and
validity. The neuropsychological tests were
ob-tamed primarily from the Halstead-Reitan and
Rei-tan-Indiana Neuropsychological Test Batteries for
Children.’9 The Halstead-Reitan batteries have
been validated by extensive research in a number
of clinical conditions.’8
Growth Measures. Growth measures were all ob-tamed within 1 week of the neuropsychological evaluation. Weight, length or height, head
circum-ference, triceps skin-fold thickness, and midarm
circumference were measured. Arm-muscle
dircum-ference (measured in centimeters) was calculated
as follows: midarm circumference (cm) - [3.14 x
triceps skin-fold (cm)] = arm-muscle
circumfer-ence. All measures were performed by trained
die-titians using the standardized methods previously
described.’’ Growth measures were expressed as
percentage of the ideal relative to the median of the
National Center for Health Statistics standards for
sex and age.223 Growth measures are missing for
two liver transplant patients.
Age of Onset. Age of onset of disease was indicated
by date of diagnosis, obtained from medical records
maintained by the liver transplant office or cystic
fibrosis clinic.
Clinical Parameters. Cyclosporine dosage and
clinical measures of liver and renal function for
each liver transplant patient at time of evaluation
were obtained from the National Institutes of
Health-sponsored liver transplant registry
main-tamed by the liver transplant program at our
insti-tution. Cyclosporine dosages were adjusted by one
of the authors (W.S.A.), the liver transplant
sur-geon, and levels were obtained in all cases within 36 hours of study participation.
Educational Parameters. Information about classroom placement was obtained from all parents.
Parents were asked to indicate whether the child
was receiving special education and whether he or she had ever repeated a grade.
Data Analyses
For the initial analyses, summary z scores were
calculated for each subject to represent his or her performance in each of the eight categories of
abil-ity. z Scores for all tests within each category of abilities were summed and averaged to obtain the
summary score for that category. Summary scores
have been used by other investigators of neuropsy-chological data,24 and are appropriate for the
follow-ing reasons. Neuropsychological skills are assessed using a variety of different tests for each category
of ability, and individual scores are not directly
comparable across tests. A summary score for each
category of ability allows performance in skill areas, rather than results of specific tests, to be compared.
This is particularly important where the tests are
modified in any fashion for different ages within each experimental group, as it allows the data to be pooled within each group. Further, initial analysis
of the summary scores significantly reduces the
high likelihood of the type I error that would exist if scores were compared separately for each of these tests.
The initial analyses compared summary z scores
in the eight categories of ability for the two groups
by means of analysis of variance. Following the
initial analyses, where significant differences (de-fined in our study as P < .05) were found, the scores
on the individual tests were also compared by using
analyses of variance.
Information about school placement was
com-pared for the two groups by means of the Fisher
Exact Test.
A secondary aim of this study was to investigate
the relationship between cyclosporine levels and
neuropsychological function. Again, initial analyses
were conducted by computing the Pearson
correla-tion coefficient for summary z scores to cyclospor-me levels. Where the correlation reached signifi-cance (P < .05), further analyses were conducted to determine the relationship between cyclosporine levels and performance on specific tests.
RESULTS
Table 3 presents the summary z scores of the
liver transplant and cystic fibrosis patients. Initial
TABLE
From 28
4. Scores on Intellectual, Liver Transplant and 18
Academic, and Neuropsychological tests, Obtained Cystic Fibrosis Patients*
Category of Ability/Testt Transplant
Patients
Control Subjects
P
-1.21 ± 1.38 -0.10 ± 0.93
-0.09 ± 1.57 0.41 ± 1.39 -0.50 ± 1.29 0.70 ± 2.01
Intellectual/academic
WISC-R Verbal IQ
WISC-R Performance IQ
WRAT-R Reading (S)
WRAT-R Spelling (5)
WRAT-R Arithmetic (5)
Learning and memory Target Test (Y)
Tactual Performance Test Memory
Localization
Abstraction-concept formation Category Test
Matching Pictures Test
Color Form Test (Y) Progressive Figures Test (Y)
Alertness and concentration
Tactual Performance Test (correct)
Memory Localization Perceptual-motor
Test of Visual Motor Integration
Marching Test (Y) Dominant hand Nondominant hand Both hands
Tactual Performance Test Dominant hand
Nondominant hand
Both hands
Trailmaking Test (0)
Part A
Part B
Visual-spatial
Test of Visual Motor Integration Individual Performance Tests (Y)
Matching Figures
Matching Vs Star
Concentric Squares
.130
TABLE 3. Means and Standard Deviations for Summary z Scores for Ability for 28 Liver Transplant and 18 Cystic Fibrosis Patients*
Categories of
Category of Ability Transplant Patients
Control Subjects
Pt
Intellectual/academic -0.88 ± 1.23 -0.02 ± 0.82 .013
Learning and memory -0.68 ± 1.09 0.19 ± 1.24 .016
Abstraction and concept formation -1.73 ± 1.58 -0.79 ± 1.37 .044
Alertness and concentration -0.29 ± 1.28 0.55 ± 1.68 .096
Perceptual-motor -0.76 ± 1.78 0.03 ± 1.54 .130
Visual-spatial -0.66 ± 1.09 0.10 ± 0.69 .012
Motor -0.13 ± 0.85 0.36 ± 0.57 .036
Sensory-perceptual -0.13 ± 0.81 0.20 ± 0.75 .172
* z Scores were obtained for each category of ability by combining and averaging z scores
for age for all tests within the category. A population of normal children will have z scores
with a mean of 0 and SD of 1. Values are given as mean ± SD. t Analysis of variance.
:1:Significant at the .05 level.
92.0 ± 16.4 89.1 ± 19.1 82.8 ± 21.5 80.5 ± 22.8 80.9 ± 22.1
-0.90 ± 1.36 -1.82 ± 1.03 -2.57 ± 3.60 -1.91 ± 2.83
-0.09 ± 1.57
-0.50 ± 1.29
-1.55 ± 1.02
-0.34 ± 1.52 -0.14 ± 2.07 -1.62 ± 1.15
0.15 ± 2.21 -0.29 ± 2.46 -0.83 ± 6.17
-0.02 ± 2.14 -1.50 ± 3.71
-1.55 ± 1.02
-0.02 ± 1.20 0.27 ± 1.17
-0.04 ± 0.76 0.26 ± 0.87
98.8 ± 18.9 105.8 ± 17.6 99.8 ± 7.7 97.1 ± 9.5 94.2 ± 14.3
-0.14 ± 1.45 -2.12 ± 0.75 -0.41 ± 2.85 -0.82 ± 2.72
0.41 ± 1.39
0.70 ± 2.01
-0.54 ± 0.79
0.40 ± 1.07
0.67 ± 1.71
-1.26 ± 1.41
0.94 ± 2.14 0.32 ± 2.66 0.09 ± 4.58
0.93 ± 0.87
-0.94 ± 2.88
-0.54 ± 0.79
0.33 ± 0.69 0.91 ± 0.65 0.04 ± 0.92 0.37 ± 0.94
TABLE 4. Continued
Category of Ability/Testt Transplant Patients
Control Subjects
P
Trailmaking Test (0)
Part A -0.02 ± 2.14 0.93 ± 0.87 .490
Part B -1.50 ± 3.71 -0.94 ± 2.88 .823
Target Test (Y) -1.21 ± 1.38 -0.10 ± 0.93 .011
Motor .036
Finger Oscillation
Dominant hand -0.14 ± 1.94 0.52 ± 1.06 .192
Nondominant hand 0.09 ± 2.06 0.58 ± 1.46 .385
Grip Strength
Dominant hand -0.27 ± 0.96 0.43 ± 0.93 .019
Nondominant hand -0.28 ± 0.97 0.44 ± 0.91 .015 Name Writing
Dominant hand -0.04 ± 1.07 0.07 ± 0.91 .728
Nondominant hand -0.08 ± 1.07 0.12 ± 0.89 .544
Sensory-perceptual .172
Auditory, visual, and tactile Imper- 0.04 ± 0.98 -0.06 ± 1.06 ceptions
Finger Recognition -0.20 ± 1.02 0.30 ± 0.92
Fingertip Symbol/Number Writing -0.23 ± 0.95 0.33 ± 1.00
Tactile Form Recognition -0.16 ± 1.17 0.23 ± 0.64
* Intelligence and achievement test scores are standard scores (mean = 100, SD = 15 in normal population). All others are z scores (mean = 0, SD = 1 in normal population).
Values are given as mean ± SD.
t WISC-R, Wechsler Intelligence Scale for Children Revised; WRAT-R, Wide Range
Achievement Test Revised; (5) indicates test administered to school-age children; (Y)
indicates test administered only in the “younger” children’s battery; (0) indicates test
administered only in the “older” children’s battery.
: Analysis of variance.
§Significant at the .05 level.
liver transplant group scored significantly lower
than the cystic fibrosis control group in the areas of intellectual/academic function (P < .013), learn-ing and memory (P < .016), abstraction and concept
formation (P < .044), visual-spatial skills (P <
.012), and motor skills (P < .036). No difference was found between the two groups on tests that measure alertness and concentration,
perceptual-motor function, and sensory-perceptual function. Table 4 presents scores on individual tests.
Anal-yses of variance were performed only when the test
contributed to a summary score in an area where
there were significant differences between the liver
transplant and control groups; thus only these P
values are reported. All neuropsychological tests
data are presented as z scores and can be compared with those obtained by normal, healthy children
who have a mean of 0 and SD of 1 on the tests.
Intelligence and academic test scores are presented
as IQ or standard scores (mean = 100, SD = 15 in
the normal population).
On tests of intellectual and academic function, the liver transplant patients scored significantly lower (P < .005 to .051) on all variables except
Verbal IQ, on which the two groups did not differ (P > .20). The differences in school placement for
the two groups were not statistically significant (P values > .10): 6 of the 24 school-age children in the liver transplant group and 1 of the 14 in the cystic
fibrosis group were receiving special education
serv-ices. Seven of the liver transplant children had
repeated a grade compared with 2 of the cystic
fibrosis patients. To assess whether the liver
trans-plant patients were receiving the special services
they need given their academic deficiencies, the
number of children with a 15-point or greater
dis-crepancy between IQ and academic standard score
(the criterion used by many school systems to
qual-ify a child for special education services for the
learning disabled) was also determined. Eighteen of
the liver transplant patients had a 15-point or
greater discrepancy between their IQ and academic achievement, and 5 of these patients were receiving special education.
On tests of neuropsychological function, the liver transplant group showed significantly poorer func-tion on the Target Test (P < .011), Localization
component of the Tactual Performance Test (P <
70--6 -4
Z-SCORE
.3
70--6 -4 -3 -2 -1 0 .1 .2 #{149}3
Z-SCORE
MARCHING TEST TARGET TEST
70-So
-%
60-
40-
30-. 20-
10--6 -4 -3 -2 -1 0 .1 ‘2 .3
Z-SCORE
-6 -4 -3 -2 -1 0 .1 .2 .3
z-Sc0RE
TACTUAL PERFORMANCE MEMORY VISUAL MOTOR INTEGRATION
70-#{149}0 - LIVER TRANSPLANT
% so - CYSTIC FIBROSIS
40- NORMALS
30-C
20-
1
0-so
-%
60-e
40-I
30-S
20-
10-
so-% 60-
40-:
C
L20-
‘10-Figure. Distributions of z scores for age on selected neuropsychological tests in liver transplant patients and cystic fibrosis control subjects, with scores for the normal
popu-lation indicated by the curve. Significance levels for the difference between the liver
transplant and the cystic fibrosis groups for each of the tests are as follows: Tactual
Performance Test, Memory Component, P > .05; Test of Visual Motor Integration, P <
.001; Marching Test, Both Hands Component, P > .05; and Target Test, P < .011.
TABLE 5. Relationship Between Cyclosporine Levels and zScores for Age on Neuropsychological Tests
Meas-uring Motor Function in 28 Children After Liver Trans-plantation
Category of Ability/Test r df Pt
Motor summary score .41 26 <.05
Finger Oscillation Test
Dominant hand .41 26 <.05
Nondominant hand .32 26 <.10 Grip Strength
Dominant hand .09 26 >10 Nondominant hand .08 26 >10 Name Writing
Dominant hand .22 24 >10 Nondominant hand .14 24 >.10
* for significance for the Pearson correlation
coeffi-cient.
Form Test (P < .063), and the Matching Vs Test
(P < .064).
Distribution profiles for a selected group of tests are provided in the Figure. These tests were selected
to highlight the differences between the groups.
Descriptions of these tests are given under
“Meth-ods.” The four profiles show scores for the two
groups relative to each other as well as to the
normal population. On the Tactual Performance
Test (Memory component), the scores of all three
groups were very similar. On the Marching Test
(both hands component), the two experimental groups had similar scores, but both had abnormal results compared with normal children. On the test of Visual Motor Integration and the Target Test, the cystic fibrosis control subjects had scores simi-lar to those of normal subjects, but the liver
trans-plant group had a curve displaced below the mean
for the control subjects.
Initial analyses revealed a significant positive relationship between motor function and cyclospor-me levels (r = .41, P < .05). Correlation coefficients
for the relationship between cyclosporine levels and
individual test scores in the motor category are
presented in Table 5. These results indicate that
children with higher levels of cyclosporine had
bet-ter function primarily on the Finger Oscillation Test, a test of tapping speed.
DISCUSSION
Children who had received liver transplantation
at least 1 year prior to participation in the study,
when compared with age- and socioeconomic
equivalent age of diagnosis and degree of growth
delay, showed deficits on measures of intellectual
and academic function and on specific
neuropsy-chological measures. These findings suggest that
the physiological and psychosocial impact of
chronic, life-threatening disease, and the growth retardation secondary to the nutritional problems
of liver transplantation, are not sufficient to
ad-count for the cognitive deficits observed in the liver
transplant sample. Our control group consisted of chronically ill children. The deficits observed in the
liver transplant population were even greater when
their performance was compared with that of nor-mal age-peers. Our findings are based on a relatively small group of children who were heterogeneous for
age, and they should be replicated with larger, more
homogeneous groups observed prospectively after diagnosis of end-stage liver disease.
In contrast to patients with adult-onset liver
disease, children show deficits on global intellectual measures as well as specific neuropsychological measures, and these deficits persist following
trans-plantation. The reason for the difference between
adults and children with liver disease may relate to the fact that the developing brain is more
vulnera-ble to the cerebrotoxic effects of liver disease, and
the deficits in function may therefore be more per-sistent, even after normal liver function is restored.
The greater vulnerability of the developing brain
has been documented,25 and the more serious con-sequences of earlier-onset brain dysfunction have been shown in other conditions.26
Perceptual-motor skills have been found to be diminished in adults with liver disease. While the
perceptual-motor grouping score was not
signifi-cantly different in our two groups, it is notable that
there was a dramatic difference on the Test of
Visual-Motor Integration (see Figure) between the
liver transplant group and both the cystic fibrosis
control subjects and normal children. This test has
an important perceptual and fine-motor
compo-nent, and performance on this test has been found
to correlate with the acquisition of academic skills in the early school years.27
The neuropsychological deficits observed are likely to have negative consequences for daily
ad-aptation and to be chronic (inasmuch as the
pa-tients were at least 1 year postoperative). It is
notable that on tests that have to do with input of
information (alertness and concentration and
sen-sory-perceptual groupings), the two groups were not
significantly different. Rather, the liver transplant
children had difficulty in “higher level” abilities,
such as abstract thinking, logical analysis, and flex-ibility of thought and memory. Such deficits are
likely to have significant impact on the manner in
which these children adapt and learn. There is a
growing interest in rehabilitation programs for
neu-ropsychological dysfunctions in the pediatric
pop-ulation,28 and remediation is available but not
uti-lized for these patients.
The reasons for the academic difficulties noted on the standardized tests taken by children who
have received liver transplantation are not
corn-pletely clear. Their neuropsychological difficulties
may relate to their academic difficulties, as visual-spatial deficits have been associated with learning disabilities.29’ 30 In addition, these children had
likely missed instruction because of frequent school absences, in contrast to the cystic fibrosis subjects,
who were a relatively well group of children and
had missed very little school. Despite their
ada-demic deficits, fewer than a third of the liver trans-plant patients were receiving the special education
services they needed. Parents and teachers may be hesitant to recognize and/or draw attention to the
special needs of these children because expectation of their performance is low. While understandable,
this approach may not be in these children’s best
interest, as it deprives them of the remedial services
they need.
Does the cystic fibrosis group provide an
ade-quate control to the liver transplant group? They were not as acutely ill as the liver transplant
pa-tients had been prior to transplant, and most cystic fibrosis patients had their disease well under
con-trol. However, this group of patients did have a
life-threatening chronic illness, with early diagnosis
and growth retardation that was even greater than
that of the liver transplant patients. In addition, the liver transplant patients at the time of
evalua-tion were all at least 1 year postoperative and in
comparatively good health. Other attributes of the
liver transplant population for which no control
was provided, ie, acute medical crises prior to
trans-plantation and major surgery followed by a period
of relatively good health and normal function, are
specific to the transplant population. The use of a
different transplant population such as children
who have received kidney transplant, while
con-trolling for some of these factors, could introduce
other confounding variables inasmuch as kidney
disease is known to be associated with central nerv-ous system effects.3’
The significant relationship between a test of
to clarify the relationship between cyclosporine
1ev-els and neuropsychological function.
CONCLUSIONS
Liver transplant patients have intellectual, ada-demic, and neuropsychological deficits.
Remedia-tion in all three areas is often available but not
utilized for these children. Our findings suggest that careful neuropsychological and psychoeducational evaluations should be undertaken for this
popula-tion, inasmuch as in many cases these children
have both underlying neuropsychological deficits
and frequently unrecognized difficulty in achieving
in school commensurate with their potential.
ACKNOWLEDGMENTS
This study was supported by the Clinical Research Center, Pediatric Subunit, US Public Health Service
grant MO1-RR00633.
We thank Betsy Fyock, RN, and Elizabeth Foster, RN,
for their help with subject recruitment and clinical data gathering, and Margareta Benser, RD/LD, and Alice Cunningham, MS, RD/LD, for obtaining the anthropo-metrics.
REFERENCES
1. Stewart SM, Uauy R, Waller DA, Kennard BD, Andrews WS. Mental and motor development correlates in patients
with end-stage biliary atresia awaiting liver transplantation.
Pediatrics. 1987;79:882-888
2. Stewart SM, Uauy R, Kennard BD, Waller DA, Benser M, Andrews WS. Mental development and growth in children with chronic liver disease of early and late onset. Pediatrics.
1988;82:167-172
3. Stewart SM, Uauy R, Waller DA, Kennard BD, Benser M, Andrews WS. Mental and motor development, social corn-petence and growth one year following successful pediatric liver transplantation. J Pediatr. 1989;1 14:574-581
4. Zitelli BJ, Miller JW, Gartner JC Jr, et al. Changes in life-style after liver transplantation. Pediatrics. 1988;82:173-180 5. Rikkers L, Jenko P, Rudman D, Freides D. Subclinical
hepatic encephalopathy: detection, prevalence, and relation-ship to nitrogen metabolism. Gastroenterology. 1978;75:462-469
6. Trzepacz PT, Brenner RP, Coffman G, van Thiel DH. Delirium in liver transplantation candidates: discriminant analysis of multiple test variables. Biol Psychiatry.
1988;24:3-14
7. Bernthal P, Hays A, Tarter RE, van Thiel D, Lecky J, Hegedus A. Cerebral CT scan abnormalities in cholestatic and hepatocellular disease and their relationship to neuro-psychological test performance. Hepatology. 1987;7:107-114 8. Gilberstadt 5,1, Gilberstadt H, Zieve L, Buegel B, Collier
RO, McClain CJ. Psychomotor performance defects in cir-rhotic patients without overt encephalopathy. Arch Intern
Med. 1980;140:519-521
9. Tarter RE, Hegedus AM, van Thiel DH, Schade RR, Gavaler JS, Starzl TE. Nonalcoholic cirrhosis associated with neu-ropsychological dysfunction in the absence of overt evidence
of hepatic encephalopathy. Gastroenterology. 1984;86:1421-1427
10. Tarter RE, Hegedus AM, van Thiel DH, Edwards N, Schade RR. Neurobehavioral correlates of cholestatic and hepato-cellular disease: differentiation according to disease specific characteristics and severity of the identical cerebral dys-function. Int J Neurosci. 1987;32:901-910
11. Tarter RE, Hegedus AM, van Thiel DH. Neuropsychiatric sequelae of portal-systemic encephalopathy: a review. Int J Neurosci. 1984;24:203-216
12. de Groen PC, Aksamit AJ, Rakela J, Forbes GS, Krom RAF. Central nervous system toxicity after liver transplantation: the role of cyclosporine and cholesterol. N EngI J Med.
1987;317:861-866
13. Jervis GA. Encephalopathy in infantile hepatic cirrhosis.
Acta Neuropathol. 1979;48:73-75
14. Galler JR. The behavioral consequences of malnutrition in early life. In: Galler JR, ed. Nutrition and Behavior. New York, NY: Plenum Press; 1984:63-188
15. Doershuk CF, Matthews LW, Tucker AS, et al. A 5 year clinical evaluation of a therapeutic program for patients with cystic fibrosis. J Pediatr. 1964;65:677-693
16. Tyler LE. The Psychology of Human Differences. New York, NY: Appleton-Century-Crofts; 1965
17. Hollingshead AB. Four Factor Index of Social Status. New Haven, CT: Yale University; 1975
18. Reitan RM, Wolfson D. The Halstead-Reitan Neuropsycho-logical Test Battery: Theory and Clinical Interpretation. Tuc-son, AZ: Neuropsychology Press; 1985
19. Reitan RM. NeuropsychologicalEvaluation of Children.
Tuc-son, AZ: Neuropsychology Press; 1987
20. Klonoff H, Low M. Disordered brain function in children and adolescents: neuropsychological and
electroencephalo-graphic correlates. In: Reitan RM, Davison LA, eds. Clinical
Neuropsychology: Current Status and Applications. New York, NY: Hemisphere Publishing Corp; 1974
21. Frisancho AR. New norms for upper limb fat and muscle areas for assessment of nutritional status. Am J Clin Nutr.
1981;34:2540-2545
22. Hamill PVV, Drizd TA, Johnson CL, et al. Physical growth: National Center for Health Statistics percentiles. Am J Clin Nutr. 1979;32:607-629
23. Nellhaus G. Head circumference from birth to eighteen years: practical composite international and interracial
graphs. Pediatrics. 1968;41:106-114
24. Nici J, Reitan RM. Patterns of neuropsychological ability in brain-disordered vs normal children. J Consult Clin Psy-chol. 1986;54:542-545
25. Reitan RM. Cerebral lesions in young children. In: Reitan RM, Davison LA, eds. Clinical Neuropsychology: Current Status and Applications. Washington, DC: Winston and Sons; 1974
26. Ryan CM. Neurobehavioral disturbances associated with disorders of the pancreas. In: Tartar RE, Van Thiel DH, Edwards KL, eds. Medical Neuropsychology. New York, NY: Plenum Press; 1988;121-158
27. Beery KE. Revised Administration, Scoring, and Teaching Manual for the Developmental Test of Visual-Motor Integra-tion. Cleveland, OH: Modern Curriculum Press; 1982 28. Rourke BP, Fisk JL, Strang JD.
NeuropsychologicalAssess-ment of Children: A Treatment-Oriented Approach. New York, NY: The Guilford Press; 1986
29. Ingram TTS, Mason AW, Blackburn I. A retrospective study
of82 children with reading disability. Dev Med Child Neurol.
1970;12:271-281
30. Mattis 5, French JH, Rapin I. Dyslexia in children and young adults: three independent neuropsychological syn-dromes. Dev Med Child Neurol. 1975;150-163
1991;87;367
Pediatrics
and Janice Nici
Sunita M. Stewart, David A. Waller, Ricardo Uauy, Walter S. Andrews, Cheryl Hiltebeitel
Neuropsychological Outcome of Pediatric Liver Transplantation
Services
Updated Information &
http://pediatrics.aappublications.org/content/87/3/367
including high resolution figures, can be found at:
Permissions & Licensing
http://www.aappublications.org/site/misc/Permissions.xhtml
entirety can be found online at:
Information about reproducing this article in parts (figures, tables) or in its
Reprints
http://www.aappublications.org/site/misc/reprints.xhtml
1991;87;367
Pediatrics
and Janice Nici
Sunita M. Stewart, David A. Waller, Ricardo Uauy, Walter S. Andrews, Cheryl Hiltebeitel
Neuropsychological Outcome of Pediatric Liver Transplantation
http://pediatrics.aappublications.org/content/87/3/367
the World Wide Web at:
The online version of this article, along with updated information and services, is located on
American Academy of Pediatrics. All rights reserved. Print ISSN: 1073-0397.
