Neurodevelopment
of Preterm
Infants:
Neonatal
Neurosonographic
and
Serum
Bilirubin
Studies
Leonard
J.
Graziani, MD; Donald C. Mitchell, MD;Michael Kornhauser, MD; Frank S. Pidcock, MD; Daniel A. Merton;
Christian Stanley, RN, MSN; and Linda McKee, MHS
ABSTRACT. In this study of 249 preterm infants of less
than 34 weeks’ gestation, the relationships between
maximal serum total bilirubin concentrations during the
neonatal period, neonatal cranial ultrasonographic
abnormalities, and severe neurodevelopmental sequelae
are described. The subjects, who were selected on the
basis of serial cranial ultrasonographic findings, had
repeated neurologic and developmental examinations
during late infancy and early childhood that established
the presence (n 45) or absence (n 204) of spastic forms
of cerebral palsy. Of the 204 subjects without cerebral
palsy, 23 scored abnormally low on standardized
devel-opmental testing during early childhood. All but seven
of the subjects with cerebral palsy had grade ffl/!V
intracranial hemorrhage or moderate to severe
periven-tricular echogenicity or both, ultrasonographic
abnor-malities that probably reflect a disruption in the
blood-brain barrier as well as extravasation of blood into brain
tissue; however,
analysis
of the data did not suggest thatthese cranial ultrasonographic abnormalities increased
either the maximum serum bilirubin concentration
dur-ing the neonatal period or the susceptibility of the
subjects to neurologic damage from hyperbilirubinemia.
Also, there was no evidence to suggest that bilirubinemia
in the range studied (2.3 to 22.5 mg/100 mL total serum
biirubin) was causally related to cerebral palsy, early
developmental delay, or the development of
periventric-ular cysts in this population of preterm infants.
Uni-variate
analyses
revealed that Apgar scores weresignif-icantly lower, while grade ffl/IV intracranial
hemor-rhage, large periventricular cysts, moderate to severe
periventricular echodensity, ventriculomegaly, assisted
ventilation, and bronchopulmonary dysplasia occurred
significantly more often in infants with cerebral palsy
than in those without cerebral palsy; however, a stepwise
logistic regression revealed that of these clinical
van-ables only large periventricular cysts, moderate to severe
periventricular echodensity, and the need for assisted
ventilation were associated with the occurrence of
cere-bral palsy at a statistically significant level. Apgar scores
and average birth weight were significantly lower, and
duration of assisted ventilation was significantly longer,
but neurosonographic findings did not differ in the 23
developmentally delayed infants compared with the
nor-mal group. Pediatrics 1992;89:229-234; preterm infants,
bilirubine,nia, cerebral palsy, cranial ultrasonographic
abnormalities, periventricular leukomalacia, intracran#{252}d
hemorrhage.
From the Departments of Pediatrics, Neurology, and Radiology, Thomas Jefferson University Hospital, Philadelphia, Pennsylvania.
Received for publication Aug 8, 1990; accepted Apr 8, 1991.
Reprint requests to (LJG.) Dept of Pediatrics, Jefferson Medical College of ThomasJefferson University, Philadelphia, PA 19107.
PEDIATRICS (ISSN 0031 4005). Copyright © 1992 by the American Acad-emy of Pediatrics,
A causal relationship between maximal serum total
bilirubin concentrations and early
neurodevelopmen-tal outcome in a large population of preterm infants
was recently reported; cerebral palsy was the
predom-inant major neurologic sequelae of
hyperbilirubine-mia,
although neonatal intracranial hemorrhage,cer-ebral ventriculomegaly, and bronchopulmonary
dys-plasia also contributed significantly to an adverse
outcome.1 A later study of the same subjects at age 5
years, however, found the association between the
risk of handicap and maximal serum total bilirubin
concentration only in those children who had
neo-natal intracranial hemorrhage.2 In another clinical and
neurosonographic study of preterm infants, the
pos-sible role of bilirubin toxicity in the development of
periventricular leukomalacia and neurologic deficits
is described.3
We reported in a previous study that neonatal
neurosonographic examinations disclosed intracranial
hemorrhage in the majority and penventricular
echodensity in all of the infants in whom cerebral
palsy eventually developed.4 Intracranial hemorrhage
results in little or no increase in serum bilirubin
levels5’6 but theoretically may increase the
suscepti-bility of the immature brain to the potentially
damag-ing effects of bilirubin-as may the lesion underlying
periventricular echodensity-by disrupting the
blood-brain barrier.7 However, whether acute brain
abnormalities detected in preterm infants by neonatal
cranial ultrasonography increase the risk or extent of
damage from serum bilirubin has not been reported.
Furthermore, others have noted that reported
associ-ations between neonatal hyperbiirubinemia and
neu-rologic sequelae have not been consistent.8
In the current study of preterm infants of less than
34 weeks’ gestation, we describe the
interrelation-ships of maximal total serum bilirubm concentrations,
neonatal cranial ultrasonographic abnormalities, and
the occurrence of cerebral palsy and developmental
delay.
Subjects
METHODS
Between August 1, 1981, and May 31, 1989, all 497 surviving infants with a birth weight of less than 1501 g or gestational age of less than 34 weeks who were admitted to Thomas Jefferson
University Hospital had, during their nursery course, serial cranial
ultrasonography utilizing high-resolution sector scanners. The 249 subjects of this report include the following: (1) 134 surviving infants who had neurosonographic evidence of moderate to severe
penventricular echodensity, grade III/IV intracranial hemorrhage,
at Viet Nam:AAP Sponsored on September 1, 2020
www.aappublications.org/news
or periventricular cysts, alone or in combination, at any time during their nursery course and whose parents consented to follow-up
evaluations; an additional 6 infants in this ultrasound category
were not followed because the parents declined; (2) a sample of I 15 of the remaining 357 infants whose neonatal ultrasonographic
studies revealed no abnormalities or only grade I/I! hemorrhage or
mild periventricular echodensity and whose parents consented to
follow-uj evaluations; a sample was selected because our previous
studies4- -‘#{176}disclosed that spastic forms of cerebral palsy do not
occur in this group of relatively low risk survivors. The 249 subjects were followed up until repeated examinations established the pres-ence (n = 45) or absence (n = 204) of cerebral palsy.
The following clinical features were recorded for each infant:
gestational age, birth weight, Apgar scores at 1 and 5 minutes, days
of phototherapy, maximal total and direct reacting serum bilirubin
concentrations, mechanical ventilation, and bronchopulmonary dysplasia (defined as supplementary oxygen dependency for more than 28 days).
Cranial Ultrasonographic Studies
Cranial ultrasonographic examinations of all infants were ob-tamed within 1 week of birth, at least weekly until 1 month of age, and then every 2 to 4 weeks until discharge. Subjects were studied
with either the AlL Mark 100 Unit or the Ultramark 4 Unit
(Advanced Technologies Lab, Bellevue, WA), equipped with 5- and 7.5-MHz sector scanner. Scans were performed through the
ante-nor fontanelle in the coronal and parasagittal planes. Coronal
images were obtained from as far forward as the frontal lobes to
the plane posterior and superior to the atria or trigones. Parasagittal views were obtained at increments from the midline to the area lateral to the lateral ventricles.
All images were reviewed by two of the authors (D.G.M. and
DAM.) who did not have knowledge of the subjects’ clinical
course; their interpretations were compared and when in accord were classified as previously reported.4 Periventricular echodensity was graded as none, mild, moderate, or severe. Cysts were recorded
as none, small (2 mm in diameter or less), or large (3 mm in
diameter or larger); large areas of multiple and contiguous small cysts were classified as large. Intracranial hemorrhage was graded
as none, grade I/I!, grade III, or grade IV. Ventricular size was measured and ventricular ratios were calculated for each subject at
approximately 36 weeks postconceptional age as previously
de-scribed.5#{176} The maximum grade of each ultrasonographic
abnor-mality noted during an infant’s nursery course was used in the
data analysis. Images that were interpreted differently by the two examiners were re-reviewed; all disagreements involved images that were ultimately classified as normal or as mild periventricular
echogenicity.
Bilirubin Measures
Serum total bilirubin concentrations were obtained in all infants
within 72 hours of birth and were obtained again as indicated by
the nursery course and laboratory results. Direct reacting bilirubin concentrations were obtained at least once, near or at the time of the maximal total bilirubin concentrations. Total and direct reacting bilirubin measurements were performed with the Advanced
Bill-rubin Stat Analyzer(Advanced Instruments Inc. Needham Heights,
MA). Phototherapy was used and exchange transfusions were
performed according to published guidelines” which were occa-sionally modified for clinical reasons. All but nine infants required phototherapy (mean number of days, 7.0 ± 4.8), and nine infants
required at least one exchange transfusion.
Twenty-four infants had direct reacting bilirubin levels above 4 mg/100 mL; results of statistical analyses were not changed by elimination of these subjects and all subjects are included in the analyses presented.
Neurologic and Developmental Measures
The presence or absence of cerebral palsy was determined in all subjects through repeated outpatient neurologic examinations
be-tween 6 months and 8 years of age by three of the authors (L.J.G.,
CS., and F.S.P.) who in some instances were aware of some
neonatal clinical and ultrasonographic findings. Cerebral palsy was
defined as spasticity of at least one lower extremity associated with
increased deep tendon reflexes, ankle clonus, and plantar extensor
responses in the affected lower limb and with markedly delayed
motor development or spastic gait or both. Cerebral palsy was
diagnosed when the characteristic abnormal neurologic findings
were noted independently by at least two of the examiners at a
minimum of three sequential examinations. Subjects who had nonspecific developmental delay without pathologic reflexes or
lower extremity spastidty, or who had transient dystonia of
prematurity’2 and eventually ambulated without a spastic gait, were not considered to have cerebral palsy. The diagnosis of spastic diparesis, hemiparesis, or quadriparesis was based on the
predom-inant pattern of extremity involvement.’3 If there was disagreement
among the examiners regarding the diagnosis, the subject was
reevaluated at I-to 3-month intervals until agreement was reached.
In the present study, the latest examination of the children with cerebral palsy was at an average age of 58.9 ± 24. 1 months (range,
21 months to 8 years).
Standard scores on the Mental and Psychomotor Indexes of the Bayley Scales of Infant Development’4 (in subjects younger than age 31 months) or the Mullen Scales of Early Learning’5 (in subjects older than 30 months) were obtained in 182 of the 204 subjects
without cerebral palsy at least once between the ages of 15 and 60 months by examiners who were unaware of the neonatal clinical and ultrasonographic findings. (Twenty-two children without
evi-dence of cerebral palsy when last evaluated neurologically did not have a standardized developmental test.) Standard scores were corrected for prematurity until age 24 months; subjects who scored
2 or more standard deviations below the mean for age on either of the Bayley indexes or any of the Mullen scales were classified as
developmentally delayed. Only the most recent standard scores
(average age 37.7 ± 13.6 months, range 15 to 72 months) were used to classify each subject as normal or developmentally delayed.
All 249 infants had ophthalmologic and hearing evaluations at the time of discharge and were reevaluated during the first year of
life.
Data Analysis
The relationship of maximal serum biirubin levels to penven-tricular cysts, cerebral palsy, and developmental delay was ana-lyzed by contingency tables, analysis of variance, Mantel-Haenszel
x2 test for trend, and logistic regression. Depending on whether
the data are continuous, ordinal, or categorical, univariate
corn-parisons of clinical neonatal variables in subjects with and without
cerebral palsy or developmental delay were analyzed by two-group ttest, Wilcoxon Rank Sum Test, and Fisher’s Exact Test. Stepwise
logistic regression was used to determine variables that significantly
predicted cerebral palsy. To evaluate the independent effect of elevated bilirubin levels on the risk of developing cerebral palsy, the significantly predictive variables were included in a logistic
model along with bilirubin level. A second model, to evaluate an increasing risk of cerebral palsy with increasing level of biirubin, was also developed using categorized levels of maximal serum
bilirubin. Finally, the data were analyzed by two-way analysis of
variance to determine whether the maximal serum total bilirubin
concentrations were increased by intracranial hemorrhage or
pen-ventricular echodensity.
All 249 subjects were included in the initial analyses of the
interrelationships between clinical neonatal features, ultrasono-graphic studies, and the occurrence of cerebral palsy (Table 1). All
subsequent analyses (Tables 2, 3, and 4) were restricted to the 134 higher risk subjects who had moderate to severe ultrasonographic
abnormalities, including grade III/IV intracranial hemorrhage,
moderate to severe penventricular echogenicity, and penventricular
cysts, alone or in combination.
RESULTS
A spastic form of cerebral palsy was diagnosed in
45 of the 249 subjects-32 with spastic diparesis, 9
with spastic quadriparesis, and 4 with spastic
hemi-paresis. Twelve of the children with cerebral palsy
were younger than age 3 years (range, 21 to 35
months) when last evaluated, at which time all had
persisting, marked spasticity with pathologic reflexes
in at least both lower extremities. The remaining 33
children with cerebral palsy, who were older than 36
absent ambulation with persisting spasticity and
path-ologic reflexes in at least two extremities. In all 45
children, cerebral palsy was initially diagnosed by age
22 months at the latest, and none were later
reclas-sified as normal, although 2 died. Seven had evidence
of at least unilateral mild sensorineural hearing loss,
but none had bilateral moderate to severe hearing
loss and none were blind. None of the 22 children
who
did
not have standardized developmental testsdisclosed neurologic signs of cerebral palsy when last
evaluated, at an average age of 27 ± 5 (SD) months
(range, 1 8 to 39 months).
In 44 of the 45 children with cerebral palsy,
neo-natal neurosonographic studies disclosed moderate to
severe periventricular echodensity, periventricular
cysts, and grade 1111W intracranial hemorrhage, alone
or in combination. Of the 182 subjects without
cere-bral palsy who had standardized developmental tests,
23 had abnormally low Bayley or Mullen scores when
last evaluated (age 30 ± 13 months; range, 15 to 58
months) and were classified as developmentally
de-layed.
The 45 subjects with cerebral palsy had slightly but
significantly lower Apgar scores and were
signifi-cantly more likely to have severe ultrasonographic
abnormalities and to have required mechanical
yen-tilation than those in whom cerebral palsy
did
notdevelop; other neonatal factors, including gestational
age, birth weight, phototherapy days, and maximum
serum bilirubin concentrations, did not differ
signifi-cantly in the two groups (Table 1).
When the significantly predictive clinical features
in Table 1 were simultaneously controlled by logistic
regression (in which only those 95% confidence limits
that do not include 1 are statistically significant),
maximal serum bilirubin concentration did not
in-crease the risk for cerebral palsy in the 134 infants
with the more severe cranial ultrasonographic
abnor-malities (Table 2). There was also no evidence of a
significant association between increasing
concentra-tions of maximal serum bilirubin and either cerebral
palsy (Table 3) or periventricular cysts (Table 4).
When maximal serum biirubin concentrations were
categorized using the incremental ranges of bilirubin
in Tables 3 and 4, logistic regressions adjusted for all
of the other neonatal clinical variables in Table 1
disclosed no increased risk for occurrence of cerebral
palsy, developmental delay, or penventricular cysts
at the higher biirubin ranges.
Excluding infants with 4 mg/100 mL or more of
direct reacting bilirubin, maximal serum total bilirubin
concentrations occurred at an average age of 6.4 ±
3.1 days.
x2
Analysisdid
not disclose a significantrelationship between maximal serum bilirubin
con-centrations and the preceding or subsequent
ultra-sonographic abnormalities; in addition, no effect of
intracranial hemorrhage or periventricular
echoden-sity or both on maximal serum bilirubin
concentra-tions was noted on two-way analysis of variance.
Of the 182 children without cerebral palsy who
had standardized developmental tests, 64 had
mod-erate to severe periventricular echodensities or grade
hilly intracranial hemorrhage or both which,
theo-retically, increased the risk of neurotoxic damage from
hyperbilirubinemia
by
disrupting
the blood-brainbarrier. Within this group of 64 subjects, the
two-group t test disclosed that the average maximal serum
total bilirubin concentration of 10.2 ± 1.9 mg/100 mL
in the 55 subjects classified as developmentally
nor-mal was not significantly different from the 9.6 ± 1.4
mg/100 mL (P = .37) in the 9 without cerebral palsy
but classified as developmentally delayed.
In a separate univanate analysis of the clinical
features in Table 1, the 23 subjects without cerebral
palsy who were developmentally delayed (at age 32
± 1 1 months) were compared with the 159 subjects
without cerebral palsy who scored within the normal
range (at age 34 ± 13 months); except for lower Apgar
scores at 1 and 5 minutes, lower average birth weight,
and longer duration of assisted ventilation in the
TABLE 1. Neonatal Clinical Features in Pre term Infants With and Without Cerebra I Palsy
Clinical Features Cerebral Palsy P
Yes (n = 45) No (n = 204) Gestational age, wk (mean ± SD)
Birth weight, g (mean ± SD)
Apgar score, median (range) 1 Minute
5Minute
Phototherapy, d (mean ± SD)
Ventilated, no. (%) If ventilated, d (mean ± SD)
BPD, no. (%)
PVE,t no. (%)
ICH4 no. (%)
Cysts, no. (%)
Ventricular ratiofi (mean ± SD)
Peak bilirubin,1 mg/dL (mean ± SD)
29.4 ± 2.2 30.0 ± 2.2 1271 ± 353 1318 ± 357
3 (0-8) 5 (0-9) 7 (4-9) 8 (2-10)
8.0 ± 4.6 6.8 ± 4.8
42 (93) 151 (74)
38.8 ± 50.0 34.2 ± 44.1
24 (53) 81 (40)
38 (84) 53 (26)
13 (29) 21 (10)
31 (69) 14 (7)
38.5 ± 9.0 33.0 ± 4.2 10.6 ± 2.9 10.1 ± 2.1
NS#
NS
CC CC
NS
CC
NS
CC CC CC CC CC NS CBronchopulmonary dysplasia; supplemental oxygen dependency for 28 days or more.
t Periventricular echodensity: moderate/severe.
IIntracranial hemorrhage: grade Ill/IV.
§Periventricular cysts: 3 mm.
II
Average of the right and left ventricular ratios.I Maximal serum total bilirubin concentrations.
# NS, not significant (P> .05).
TABLE 2. Neonatal Clinical Features and Adjusted Odds Ratios for Occurrence of Cerebral Palsy
(n = 44) in 134 Preterm Infants With Moderate to Severe Neurosonographic Abnormalities
Clinical Features Odds RatioC 95% Confidence
Limit
PVEt (none/mild, moderate/severe) 5.7 2.1-15.7
Cyst (none/small, large) 5.4 2.0-15.0
Ventilation (no, yes) 4.4 1.1-17.1
Peak bilirubin (continuous)% 1 .1 0.9-1.3
CAdjusted for all other variables in table and for gestational age.
tPeriventricular echodensity.
Periventricular cysts: small, 3 mm; large, 3 mm.
Maximal serum total bilirubin concentration as a continuous variable.
TABLE 3. Cerebral Palsy (n = 44) in Relationship to Maximal
Serum Total Bilirubin Concentrations in 134 Preterm Infants With Moderate to Severe Neurosonographic Abnormalities
Bilirubin, mg/dL Cerebral Palsfl No Cerebral Palsy
(n = 44), (n = 90), No. (%) No. (%)
8 6(13.6) 10(11.1)
8.0-10.9 20 (45.4) 49 (54.4)
11.0-12.9 12 (27.3) 20 (22.2)
13 6(13.6) 11 (12.2)
CGrade llI/IV intracranial hemorrhage, moderate to severe pen-ventricular echogenicity, or periventricular cysts.
t
Not significant; Mantel-Haenszel x2test for trend = .05; P= .81.TABLE 4. Neonatal Periventricular Cysts in Relationship to
Maximal Serum Total Bilirubin Concentration in 134 Preterm
In-fants With Moderate to Severe Neurosonographic Abnormalities
Bilirubin, mg/dL Periventricular Cystst
No Cysts Small Cysts Large Cysts
(n = 42), (n = 47), (n = 45), No. (%) No. (%) No. (%)
8 2(4.8) 7(14.9) 7(15.6)
8.0-10.9 22 (52.4) 25 (53.2) 22 (48.9) 11.0-12.9 12 (28.6) 11 (23.4) 9 (20.0)
13 6(14.3) 4(8.5) 7(15.6)
CGrade III/IV intracranial hemorrhage, moderate to severe peri-ventricular echogenicity, or periventricular cysts.
t
Not significant; x2 4.4; P= .62.delayed group (P < .05), no significant differences
between the two groups were found, although these
results should be considered preliminary. None of the
subjects without cerebral palsy had bilateral severe
retinopathy or hearing loss.
In summary, our data analysis disclosed that large
penventricular cysts, moderate to severe
periventric-ular echogenicity, and the need for but not the
dura-tion of mechanical ventilation significantly increased
the risk for the occurrence of cerebral palsy. Maximal
serum total biirubin levels were not associated with
a statistically significant increased risk for
periventric-ular cysts, cerebral palsy, or early developmental
de-lay.
DISCUSSION
Hyperbilirubinemia resulting in kernicterus
pro-duces widespread neuronal damage and neurologic
deficits, with the most common sites of staining and
destruction involving the hippocampus, basal ganglia,
and the brainstem nuclei in the floor of the fourth
ventricle, including the dorsal cochlear nucleus.’6
Au-topsy studies have noted histologic evidence of
ker-nicterus despite relatively low concentrations of
serum bilirubin in preterm asphyxic infants.17 The
neurologic deficits associated with kemicterus, at least
in full-term infants, include dyskinetic forms of
cer-ebral palsy with high-frequency hearing loss and
impaired vertical gaze.’8 In contrast, the
neuropath-ologic correlates of nonkernicteric brain injury in
pre-term infants predominantly involve penventricular
white matter; as a result, spasticity without
dys-kinesia, especially of the lower extremities, is
charact-eristic of prematurely born children with cerebral
palsy.’9 Whether hyperbilirubinemia is causally
re-lated to less severe neurologic and neuropathologic
abnormalities than observed in kernicterus is
uncer-tain,7’2#{176}although we previously reported that
hyper-bilirubinemia had an adverse effect on peripheral but
not central auditory pathways in studies of brainstem
auditory evoked potentials in high-risk preterm and
full-term neonates.21
In a study of 831 preterm infants, van de Bor et al1
suggested a causal relationship between maximal
bil-irubin concentrations and unspecified type of cerebral
palsy diagnosed at the corrected age of 2 years,
al-though Newman and Maisels8 noted that the strength
of the relationship may be weak; also, by age 5 years
the risk of handicap in relationship to
hyperbilirubi-nemia was found only in those children who had
neonatal intracranial hemorrhage.2 In another clinical
and cranial ultrasonographic study, Ikonen et al3
re-ported that neonatal bilirubin neurotoxicity in 5
pre-term infants was probably responsible for large
peri-ventricular cysts with ventriculomegaly and severe
neurologic sequelae including mental retardation and
spastic forms of cerebral palsy.
In the present study, we were unable to
demon-strate a significant association between maximal
serum bilirubin levels and the subsequent occurrence
of neonatal neurosonographic abnormalities
(includ-ing
large periventricular cysts and ventriculomegaly),cerebral palsy, or developmental delay. Van de Bor et
aP reported a much larger group (n = 831), including
50 infants (6%) with maximal total bilirubin levels
above 14.6 mg/100 mL, 5 of whom had ‘major
hand-icaps.’ By contrast, only 3 infants (1.2%) in our study
had levels above 14.6 mg/100 mL, 1 of whom
devel-oped cerebral palsy; the latter infant had a maximal
serum total bilirubin level of 22.5 mg/100 mL, 11
mg/dL of which, however, was direct reacting.
De-spite the differences in distribution of bilirubin
infants identified as having cerebral palsy are similar in the two studies.
We4 and others2224 previously reported that the
risk of developing cerebral palsy was increased in
preterm infants with the more severe grades of
intra-cranial hemorrhage or with moderate to severe
pen-ventricular echodensity and periventricular cysts.
Therefore, the relatively high incidence of cerebral
palsy (1 8. 1
%)
in the present study of 249 subjectswas due to the selection of virtually all infants with
severe neurosonographic abnormalities while
includ-ing only a representative sample (n = 1 15) from the
remaining 357 lower risk infants. Assuming complete
ascertainment, cerebral palsy occurred in 9. 1 % of the
497 surviving preterm infants, a sequelae rate
con-sistent with our previous report of 9% in a smaller
population4 but higher than the incidence of 5.4% of
“major handicap’ reported by van de Bor et al.’ The
incidence of neurologic sequelae may differ among
studies of preterm infants because of such factors as
age at diagnosis, gestational or pennatal variables
associated with etiologic mechanisms, and size of the
study population. Additionally, because of limitations
related to the upper range of bilirubin concentrations
in the present study, we were unable to determine
the risk of cerebral palsy associated with maximal
serum bilirubin levels above 14 mg/100 mL.
Intracranial hemorrhage may be difficult or
impos-sible to distinguish from severe periventricular
echo-density without serial ultrasonographic studies, and
stippling or bright specks within the echodensity may
represent petechial hemorrhages. However, we found
that neither periventricular echodensity nor
in-tracranal hemorrhage, regardless of severity, was
re-lated to increased maximal serum bilirubin
concentra-tions. In addition, the most severe grade of
intracra-nial hemorrhage occurred an average of 2 days after
the mean maximal total bilirubin concentration.
The neuropathology of nonkernicteric brain injury
in preterm infants was recently reviewed by Volpe,’9
who discussed the (theoretical) pathogenesis of
pen-ventricular leukomalacia and hemorrhagic infarction.
Periventricular echodensity and periventricular cysts,
readily detected by high-resolution sector scanners,
are the neurosonographic correlates of periventricular
leukomalacia and, in addition to grade IV intracranial
hemorrhage, are characteristically noted during the
nursery course of preterm infants in whom cerebral
palsy develops.4’9’2224 In the present study, we were
able to confirm the highly predictive value of these
relatively well-defined neonatal ultrasonographic
ab-normalities. We also noted, as did van de Bor et al,’
that the cerebral ventricles were significantly larger
and the incidence of both bronchopulmonary
dyspla-sia and intracranial hemorrhage was significantly
greater in subjects in whom cerebral palsy was later
diagnosed. Logistic regression analysis, however,
dis-closed that only moderate to severe periventricular
echodensity, large periventricular cysts, and the need
for mechanical ventilation were significantly
associ-ated with the occurrence of cerebral palsy. We found
no statistically significant evidence that either
intra-cranial hemorrhage or periventricular echodensity
in-creased the susceptibility of our subjects to the
puta-tive neurotoxic effects of serum bilirubin in
relation-ship to the subsequent development of penventricular
cysts, cerebral palsy, or early developmental delay. It
is possible, however, that some children in our study,
including
the 1 2 with a diagnosis of cerebral palsywho were younger than age 36 months when last
evaluated, have been misclassified. Also, the absence
of a significant relationship between maximal total
serum bilirubin levels and abnormally low Bayley or
Mullen standard scores in 23 children without
cere-bral palsy must be considered a preliminary finding
because of the relatively small number of affected
children and the wide age range of those tested to
date.
The significant association between the need for
mechanical ventilation and increased risk for cerebral
palsy in preterm infants requires further study. In
addition, to determine whether neonatal
hyperbili-rubinemia and neurosonographic abnormalities are
related to the academic deficiencies reported to occur
at school age in 25% to 45% of prematurely born
children,’9 longer term studies are required.
ACKNOWLEDGMENTS
This research was supported by the National Institutes of Health,
grants HD2 1453 and NS27463.
We thank G. McCann for editing, P. Micou for data filing and compilation, and J. Lawrence, PhD, for review of the manuscript. We also thank the nursing staff of the Intensive Care Nursery of Thomas Jefferson University Hospital and A. R. Spitzer, MD, for
their help and support.
REFERENCES
I. van de Bor M, van Zeben-van der Aa TM, Verloove-Vanhorick SP,
Brand R, Ruys JH. Hyperbilirubinemia in preterm infants and
neurode-velopmental outcome at 2 years of age: results of a national collaborative survey. Pediatrics. 1989:83:915-920
2. van de Bor M, Veen S. Ens-Dokkum M, Schreuder AM, Brand R, Verloove-Vanhorick SP. Hyperbilirubinemia in preterm infants and
neurodevelopmental outcome at 5 years of age. Pediatr Res. 1990;27:259A
3. Ikonen RS, Kuusinen EJ, Janas MO, Koivikko MJ, Sorto AE. Possible etiological factors in extensive periventricular leukomalacia of preterm infants. Acta Paediatr Scand. 1988;77:489-495
4. Pidcock FS, Graziani U, Stanley C, Mitchell DC, Merton D.
Neuroson-ographic features of periventricular echodensities associated with cere-bral palsy in preterm infants. IPediatr. 1990116:417-422
5. Epstein MF, Leviton A, Kuban KCK, et al. Bilirubin, intraventricular
hemorrhage, and phenobarbital in very low birth weight babies.
Pedi-atrics. 1988;82:350-354
6. Amato M, Fauchere JC, von Muralt G. Relationship between periven-tricular hemorrhage and neonatal hyperbilirubinemia in very low birth
weight infants. Am JPerinatol. 1987;4:275-278
7. Hansen TWR, Bratlid D. Bilirubin and brain toxicity. Ada Paediatr Scand. 1986;75:513-522
8. Newman TB, Maisels MJ. Bilirubin and brain damage: what do we do
now? Pediatrics. 1989;83:1062-1065
9. Graziani U, Pasto M, Stanley C, et al. Neonatal neurosonographic correlates of cerebral palsy in preterm infants. Pediatrics. 1986;78:88-95
10. Graziani U, Pasto ME, Stanley C, et al. Cranial ultrasound and clinical
studies in preterm infants. I Pediatr. 1985;106:269-276
I I. Maisels MJ. Neonatal jaundice. In: Avery GB, ed. Neonatology. Philadel-phia, PA: JB Lippincott; 1975
12. Drillen CM. Abnormal neurologic signs in the first year of life in low birthweight infants: possible prognostic significance. Dcv Med Child
Neurol. 1972;14:575
13. Taft LT. Cerebral palsy. Pediatr Rev. 1984;6:35-45
14. Bayley N. Bayley Scales of Infant Development. New York, NY: Psycho-logical Corporation; 1979
at Viet Nam:AAP Sponsored on September 1, 2020
www.aappublications.org/news
15. Mullen E. Mullen Scales of Early Learning. East Providence, RI: TOTAL
Child mc; 1984
16. Friede RL. Developmental Neuropathology. New York, NY:
Springer-Verlag 1975
I 7. Ahdab-Barmada M, Moossy 1. The neuropathology of kernicterus in the premature neonate: diagnostic problems. I Neuropathol Exp Neurol. 1984;43:45-56
18. Hagberg B, Hagberg C, Olow I. The changing panorama of cerebral
palsy in Sweden 1954-1970: analysis of various syndromes. Ada
Paed-iatr Scand. 1975;64:193
19. Volpe JJ. Brain injury in the premature infant: is it preventable? Pediatr Res. 1990;27:S25-S33
20. Lucey JF. Bilirubin and brain damage: a real mess. Pediatrics.
1982;69:381-382
2 1. Streletz U, Graziani U, Branca PA, Desai HJ, Travis SF, Mikaelian DO. Brainstem auditory evoked potentials in fullterm and preterm newborns
with hyperbilirubinemia and hypoxemia. Neuropediatrics.
1986;17:66-71
22. Guzetta F, Shackelford GD, Volpe 5, Perlman JM, Volpe JJ. Periventric-ular intraparenchymal echodensities in the premature newborn: critical determinant of neurologic outcome. Pediatrics. 1986;78:995-1006 23. Graham M, Trounce JQ, Levene MI, et al. Prediction of cerebral palsy
in very low birthweight infants: prospective ultrasound study. Lancet.
1987;2:593-596
24. Bozynski MEA, Nelson MN, Genaze D, et al. Cranial ultrasonography
and the prediction of cerebral palsy in infants weighing less than 1200
grams at birth. Dcv Med Child Neurol. 1988;30:342-348
ANNOUNCEMENT
THE 1992 GENERAL PEDIATRICS CERTIFYING EXAMINATION OF THE
AMERICAN BOARD OF PEDIATRICS
The 1992 General Pediatrics Certifying Examination will be administered on
Tuesday and Wednesday, October 27 & 28 in various cities throughout the United
States and Canada.
NEW APPLICANTS
For new applicants, regular registration begins December 2, 1991 and extends
through March 2, 1992. The registration fee is $1,050. The registration fee must
accompany the completed application material and be postmarked by March 2, 1992.
On December 2, 1991 application material will be sent to qualified residents at their
training program address. However, any resident who does not receive the
material by mid-December should request the material directly from the
Amer-ican Board of Pediatrics (ABP). There will be a three-part partial payment plan for
new applicants only. The deadline for participation in that plan is January 13, 1992;
details will be included in the application material. Late registration begins March
3, 1992. The late registration fee is $1,250 (which includes a $200 nonrefundable
late fee). The final deadline for receipt of applications is May 1, 1992.
RE-REGISTRATION
In mid-February re-registration material will be mailed to those active candidates
who meet the examination requirements. Those who do not meet the examination
requirement of having a copy of their permanent, unrestricted license on file will
not be sent the re-registration material. Regular re-registration begins on February
17, 1992 and extends through April 17, 1992. Qualified individuals who do not
receive the material by March 2 should request the material directly from the ABP.
The re-registration fee is $950. The re-registration fee must accompany the
corn-pleted re-registration form and be postmarked by April 17, 1992. Late re-registration
begins April 18, 1992. The late re-registration fee is $1,150 (which includes a $200
nonrefundable late fee). The final deadline for receipt of re-registration material
is May 1, 1992.
SITE ASSIGNMENT
Site assignment is based on the date of receipt of the application or the
re-registration material. There may be limited seating at some sites so early registration
is suggested.
The American Board of Pediatrics
111 Silver Cedar Court
Chapel Hill, NC 27514-1651
1992;89;229
Pediatrics
Merton, Christian Stanley and Linda McKee
Leonard J. Graziani, Donald G. Mitchell, Michael Kornhauser, Frank S. Pidcock, Daniel A.
Bilirubin Studies
Neurodevelopment of Preterm Infants: Neonatal Neurosonographic and Serum
Services
Updated Information &
http://pediatrics.aappublications.org/content/89/2/229
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
1992;89;229
Pediatrics
Merton, Christian Stanley and Linda McKee
Leonard J. Graziani, Donald G. Mitchell, Michael Kornhauser, Frank S. Pidcock, Daniel A.
Bilirubin Studies
Neurodevelopment of Preterm Infants: Neonatal Neurosonographic and Serum
http://pediatrics.aappublications.org/content/89/2/229
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.