Neonatal Cerebral Ventricles


Чтобы посмотреть этот PDF файл с форматированием и разметкой, скачайте его и откройте на своем компьютере.
230

radiology.rsna.org
n
Radiology:
Volume 262: Number 1January 2012
PEDIATRIC IMAGING:

Table 3

Prediction of Neonatal Ventricular Size after Birth and Throughout the Neonatal Period
Cranial US 1 (
n
= 625)
Intercept8.03 (0.09) * 1.11 (0.04) * 15.09 (0.22) *
GA, in weeks

0.22 (0.01) * NS0.04 (0.02)

Male sex0.28 (0.07) * 0.12 (0.05)

0.49 (0.19)

Cranial US 1US 2 (
n
= 301)
Intercept
0.90 (0.11) *
Interval birthUS 1, in hours
NS
Interval US 1US 2, in hours
II
0.09 (0.03) *
GA
NS
GA by interval US 1US 2, in hours
II
NS
Male sex
0.16 (0.07)


Vaginal birth

2
0.26 (0.08)

Vaginal birth by interval US 1US 2, in hours
II
0.05 (0.02)

Cranial US 1TEA US (
n
= 79)
Intercept8.30 (0.20) * 0.47 (0.13) * 15.23 (0.53) *
Postmenstrual age, in weeks

0.22 (0.02) * 0.11 (0.01) * 0.20 (0.04) *
,
27 weeksNSNSNS
,
27 weeks by postmenstrual age, in weeks

0.07 (0.03)

NS0.15 (0.07)


Male sexNS0.63 (0.19) * NS
Male sex by postmenstrual age, in weeks

NS
2
0.04 (0.01) * NS
*


P

,
.001.

Centered at 24 weeks.


P

,
.01.


P

,
.05.
ll
Natural logarithm.
to be most pronounced in the occipi-
tal horns, which is in agreement with
other neonatal (
7,1820
21
)
imaging studies in neonates (
22
), chil-
dren, and adolescents (
23
).

Male sex was associated with a larger
ventricular size after birth, but its effect
fants who were followed prospectively.
Sex-related differences in ventricular size
have been reported previously for both
24
) and neonates (
7
) and were
also demonstrated to disappear with fur-
ther growth (
25
).

After birth, a small increase in AHW
nates. This con
rms the concept of ven-
tricular reopening proposed by Nelson
15
), who subjectively assessed the
amount of cerebrospinal 
uid in the lat-
eral ventricles of term infants who were
born vaginally and observed a gradual

Table 4

at Term
R

2
: 0.354)AHW (
R

2
: 0.430)TOD (
R

2
: 0.372)
Intercept11.22 (0.17) * 1.23 (0.07) * 15.84 (0.18) *
,
30 weeks)0.45 (0.18) * 1.20 (0.13) * 3.45 (0.31) *
Postmenstrual age, in weeks

0.19 (0.05) * NSNS

0.17 (0.05) * 0.08 (0.04)

NS
*
P

,
.001.


Centered at 37 weeks.

Centered at 35 cm.


P

,
.05.
intracranial pressure and hence be an
indication for intervention. Measure-
ments of
the AHW and TOD did not
exceed 4 and 24 mm, respectively, in
our cohort of neonates without brain
abnormalities. Therefore, follow-up of
infants
with ventricular measurements
exceeding these values is recommended,
even though it is still unclear at what
degree ventricular enlargement is asso-
ciated with r
aised intracranial pressure.

to play only a minor role. The lateral
ventricles were, in general, slightly larger
Radiology:
Volume 262: Number 1January 2012

n

radiology.rsna.org

229
PEDIATRIC IMAGING:
constant and the occipital horns only
slightly enlarged with GA. These obser-
vations are in line with prior reports
(
4,69
10
),
who demonstrated a marked increase
in AHW and TOD with increasing GA.

The reference intervals for the VI
presented in our study are consistent
with the curve published by Levene (
6
)
30 years ago; h
owever, in
the study
than 26 weeks GA were eligible. Liao
4
) reported slightly higher values
pressure-
driven dilation following a
germinal
matrix-intraventricular hem-
orrhage and ex v
acuo dilation due to
periventricular white matter loss.

In our study, new cross-sectional
reference curves were established for
the ventricular size of newborn infants
as longitudinal curves for the follow-up

Cross-sectional VI measurements
showed a considerable increase with
maturation, whereas the AHW remained

Figure 3


Figure 3:
Cross-sectional reference curves for the
(a)
VI,
(b)
AHW, and

(c)
n
= 625);
presented are the estimated means and 95% reference intervals  tted to
(orange triangles).
clusion of fewer extremely premature
infants in both studies and the lower
transducer resolution in the 1980s may
account for the small differences be-
tween the previous and present refer-
ence values. Reference intervals for the
AHW (


3 mm) are in line with earlier
studies (
4,7,8,10
). Literature regarding
the size of the occipital horns is incon-
clusive; previous reported upper limits
for the TOD range from 7 mm for the
most premature infants up to 24.7 mm
(
7,9,10
). In our cohort, the 97th per-
and 21 mm for term infants.

Although measuring occipital horn
size may be challenging due to obliq-
uity of the transducer and dif
culties
in de
ning the occipital border in some
infants, evaluation of TOD is of clini-
cal value. The occipital horn may show
the earliest and fastest increase in size
in infants with posthemorrhagic ven-
tricular dilation (
16,17
), and even iso-
lated occipital horn dilation may be
accompanied by signs of an increased
226

radiology.rsna.org
n
Radiology:
Volume 262: Number 1January 2012
PEDIATRIC IMAGING:
supervision of neonatologists (L.S.d.V.
or M.J.N.L.B.). Scans were performed
by using an Aplio (Toshiba Medical
or ATL-5000 (Philips Medical Systems,
(until 2010) and a Xario scanner (Toshiba
Medical Systems) (from 2010 onwards)
with a transducer frequency of 58
MHz. Scanning was performed at the
bedside with the infants head in supine
position. With the anterior fontanel
used as an acoustic window, standard
views were obtained in the coronal and
sagittal planes. Ventricular measure-
ments were performed of
ine with Im-
ageJ (version 1.42q;
http://rsb.info.nih
.gov/ij

) or DicomWorks (version 1.3.5;

http://www.dicomworks.com

) by one
of the authors (M.J.B., with 3 years of
experience in neonatal cranial US).



The ventri-
cular index (VI) (de
ned as the
distance
the anterior horn in the coronal plane),
anterior horn width (AHW) (de
ned as
the diagonal width of the anterior horn
measured at its widest point in the
coronal plane), and thalamo-
occipital
distance (TOD) (de
ned as the distance
thalamus at its junction with the cho-
roid plexus and the outermost part of
the occipital horn in the parasagittal
plane) were evaluated (
Fig 2

).

Intra- and interobserver reliability.


The intra- and interobserver reliability
were assessed in 10 infants. To evalu-
ate the intraobserver variability, cranial
US examination and ventricular mea-
surements were repeated on the same
day (with a 14-hour interval) by a single
observer (M.J.B.). The interobserver
reliability was evaluated by repeating
the cranial US procedure and ventric-
ular measurements in each infant by a
second observer (M.J.N.L.B.), who was
unaware of the 
rst observers data.

The intraclass correlation coef
cients
were calculated by using the two-way
random model for absolute agreement.
The intraclass correlation coef
cient
for single measures was considered and
of agreement scale by Brennan and
Silman (
13
).

Cranial US.


The 
rst cranial US
scan (hereafter, US 1) was conducted
within 4 days following birth. To assess
the presence of ventricular reopen-
ing, cranial US scanning was repeated
within the 1st week of life (hereafter,
US 2) in those neonates who underwent
US 1 within 48 hours after birth and
were still admitted. Neonatal cranial
US was performed as part of routine
care at the neonatal medium or inten-
sive care unit, whereas term newborns
in the maternity ward were scanned for
research purposes only. Cranial US at
TEA (hereafter, TEA US) was part of
onates. Subsequent scans (US 2 and/or
TEA US) in infants who developed ce-
rebral abnormalities after US 1 were
excluded from further analysis.

Cranial US was conducted by a
team of experienced examiners includ-
ing neonatologists (F.G., L.S.d.V. and
M.J.N.L.B.) with more than 10 years
experience in neonatal cranial US as
well as medical doctors (M.J.B.) and
fellow neonatologists with at least 3
months experience in neonatal cranial
US, who performed cranial US under

Applied exclusion criteria for both
abolic disorders, perinatal asphyxia,
seizures, central nervous system infec-
tions, cranial US abnormalities other than
mild transient periventricular echoge-
nicities, and an unknown GA.

infants (GA,
,

30 weeks), included be-
tween January 2007 and June 2009, was
followed prospectively and included for
cranial US at term-equivalent age (TEA)
if no US abnormalities were observed
at sequential examinations or around
term. Neurodevelopmental outcome was
assessed at 15 months corrected age by
using the Grif
ths Mental Development
Scales. Infants with low scores accord-
ing to Ivens and Martin (corrected
developmental quotient,
,

83) were ex-
cluded to ensure that the longitudinal
reference values were based on a co-
outcome (
12
). In total, one-third of the
115 infants had to be excluded and con-
were eligible for prospective inclusion
at TEA (
Fig 1

).

Measurements

Maternal and neonatal charts were
reviewed for demographic and clinical
characteristics.

Perinatal and neonatal character-
istics considered were maternal preg-
nancy-induced hypertensive disease,
prolonged rupture of membranes (
.

24
hours), histologic diagnosis of chorio-
amnionitis, antenatal corticosteroids,
multiple birth and monochorionicity,
mode of delivery, GA, sex, birth weight,
and head circumference.

Postnatal events recorded in the
mechanical ventilation, administration
of hydrocortisone, bronchopulmonary
dysplasia de
ned as need for supple-
mental oxygen at 36 weeks postmen-
strual age, inotropic support, persistent
ductus arteriosus requiring treatment
tizing enterocolitis, sepsis de
ned as a
positive blood culture, and the infants
postmenstrual age, weight, and head
circumference at TEA.

Figure 1


Figure 1:
neonates (GA
,
and June 2009
(*)
.
cUS
= cranial US.
232

radiology.rsna.org
n
Radiology:
Volume 262: Number 1January 2012
PEDIATRIC IMAGING:
present article: Royalties from book,
Atlas of
Neonatal Ultrasonography

. Other relationships:
none to disclose.
F.G.

Financial activities related
to the present article: none to disclose. Finan-
cial activities not related to the present article:
Payment for service lecture on PHVD therapy
for Swedish Neonatology Council. Other rela-
tionships: none to disclose.
C.K.

No potential
con
icts of interest to disclose.
L.R.P.

No po-
tential con
icts of interest to disclose.
E.J.H.M.


No potential con
icts of interest to disclose.

M.J.N.L.B.

No potential con
icts of interest to
disclose.

References



1
.
Volpe

JJ
.
Intracranial hemorrhage: germi-
nal matrix-intraventricular hemorrhage of
the premature infant. Neurology of the new-
born
.
5th ed.

Philadelphia, Pa
:
Saunders
Elsevier
,
2008
.




2
.
Soul

JS
,
Eichenwald

E
,
Walter

G
,
Volpe

JJ
,

du Plessis

AJ
.
CSF removal in infantile
posthemorrhagic hydrocephalus results in
signi
cant improvement in cerebral hemo-
dynamics
.
Pediatr Res

2004
;
55
(
5
):
872

876
.




3
.
van Alfen-van der Velden

AA
,
Hopman

JC
,

Klaessens

JH
,
Feuth

T
,
Sengers

RC
,
Liem

KD
.

Cerebral hemodynamics and oxygena
tion
after serial CSF drainage in infants with
PHVD
.
Brain Dev

2007
;
29
(
10
):
623

629
.




4
.
Liao

MF
,
Chaou

WT
,
Tsao

LY
,
Nishida

H
,

Sakanoue

M
.
Ultrasound measurement of
the ventricular size in newborn infants
.

Brain Dev

1986
;
8
(
3
):
262

268
.




5
.
Mller

WD
,
Urlesberger

B
.
Correlation of
ventricular size and head circumference
later age to con
rm their favorable out-
come. Finally, no relationship was found
and the increase in ventricular size from
aging measurements may, h
owever, be
factors for ex vacuo ventriculomegaly
than two-dimensional measurements,
and therefore caution has to be taken

In conclusion, new cross-sectional
and longitudinal reference values were
established for the neonatal lateral ven-
tricles, which may allow early identi
-
cation of posthemorrhagic and ex vacuo
ventricular dilation and may offer the
opportunity for accurate timing of in-
tervention in infants with progressive
posthemorrhagic ventricular dilation.


Acknowledgments:

We are grateful to the par-
ents who consented to the participation of their
babies in the study, and we thank Madelon
Engel, MD, Marlijne Ikink, MD, Ellen van Rooijen,
MD, Ratna Tewarie, MD, Helen Torrance, MD,
PhD, and Elske Weber, MD, for their great help
with the collection of the data.


Disclosures of Potential Con
icts of Interest:
M.J.B.

Financial activities related to the present
article: Alexandre Suerman Stipendium grant
for MD PhD students from University Medical
Center Utrecht. Financial activities not related
to the present article: none to disclose. Other
relationships: none to disclose.
L.S.d.V.

Finan-
cial activities related to the present article: none
to disclose. Financial activities not related to the
to 5 years to include the required min-
imum of 25 extremely low birth weight
infants without US abnormalities per
week GA. Although all images were
reviewed and measured by a single re-
searcher, some uniformity might have
been lost due to the performance of
cranial US by several examiners and
with different US machines during this
5-year period. Measurements of third
and fourth ventricular size as well as
Doppler US measurements of cerebral
blood 
ow velocities were beyond the
scope of this study but should, h
owever,
be considered as a valuable additional
tool to evaluate the pathophysiology of
ventriculomegaly. The phenomenon of
ventricular reopening was mainly stud-
ied in term neonates born by means of
Healthy term neonates born after a vag-
inal delivery were often discharged on
the same or next day and had either
no second cranial US or a shorter time
the extent of reopening was shown to
correlate with the time interval be-
tween US scans, the in
uence of vaginal
birth on ventricular reopening in term
infants may have been underestimated.
The longitudinal reference values es-
tablished in this study were based on
premature infants with a favorable neu-
rodevelopment at 15 months corrected
age. These infants will be reassessed at

Table 5

Intra- and Interobserver Reliability for Ventricular Measurements Assessed by Repeating Cranial US in 10 Infants
Intraobserver ReliabilityInterobserver Reliability
Mean
Observer A at
US 1 *
Mean
Observer A at
US 2 *
Median
Difference


Intraclass
Correlation
Coef cient
Mean
Observer A at
US 1 *
Mean
Observer B at
US 1 *
Median
Difference


Intraclass
Correlation
Coef cient
VI right8.9 (1.0)9.0 (1.0)0.1 (
2
0.80.6)0.8968.9 (1.0)8.9 (1.1)
2
0.2 (
2
0.91.2)0.730
VI left8.7 (0.8)

9.0 (0.7)

0.4 (
2
0.40.9)0.7978.7 (0.8)9.1 (1.1)0.3 (
2
0.71.5)0.711
AHW right1.9 (0.9)2.0 (0.9)0.0 (
2
0.10.5)0.9701.9 (0.9)1.8 (0.9)
2
0.1 (
2
0.50.3)0.972
AHW left2.0 (0.6)2.1 (0.7)0.1 (
2
0.70.8)0.8342.0 (0.6)

2.4 (0.7)

0.4 (0.01.3)0.713
TOD right16.9 (3.9)16.0 (4.0)
2
0.7 (
2
5.02.2)0.87616.9 (3.9)16.5 (3.1)
2
0.7 (
2
5.24.2)0.673
TOD left17.0 (3.7)17.4 (4.1)0.7 (
2
2.12.4)0.91817.0 (3.7)

18.9 (4.6)

2.1 (
2
1.15.4)0.789
* Numbers in parentheses are standard deviation.


Numbers in parentheses are the range.


P

,
.05.


P

,
.01.
Radiology:
Volume 262: Number 1January 2012

n

radiology.rsna.org

225
PEDIATRIC IMAGING:
gestational age (GA) as well as longitu-
dinal reference values for the follow-up
30 weeks gestation.


Institutional review board approval and
written parental informed consent were
obtained for this prospective observa-
tional cohort study.

Patients

2010, 625 neonates were consecutively
enrolled, including 217 term infants ad-
mitted with the mother on the mater-
nity ward or to the medium care unit
ther the medium or intensive care unit
of the Wilhelmina Childrens Hospital.
To acquire a representative sample size,
a minimum of 25 neonates were included
per week of gestation (except for 24 and
the time from the 1st day of the last
menstrual period and con
rmed with a
crownrump length measurement in the
1st trimester. If the US date differed by
more than 6 days from the menstrual
date, the US date was used.
an infant with progressive posthemor-
rhagic ventricular dilation needs treat-
ment. Since then, others have also
reported nomograms for the neonatal
ventricles, among which the reference
known (
4,710
).

Recently, we reviewed available data
on this topic and demonstrated consid-
erable variation among reported ref-
erence curves, especially for the most
immature infants, who face the greatest
risk of developing ventricular dilation
(
11
). A substantial number of studies
were conducted more than a decade
ago with less-sophisticated US equip-
ment. Not all studies covered the entire
neonatal period, and in general, only
few extremely low birth weight infants
were included in previous research.
Furthermore, not all studies excluded
neonates with small germinal matrix
or intraventricular hemorrhages, which
may have in
uenced reported measure-
ments of ventricular size. In addition,
to our knowledge, no longitudinal refer-
ence curves are currently available for
sequential cranial US evaluation of ven-

Accurate reference values are of
importance for infants with posthemor-
rhagic ventricular dilation, as interven-
tion is started on the basis of changes
in ventricular size of only a few mil-
should enable improved identi
cation
neonates. Therefore, the aims of this
study were to establish new cross-sec-
tional reference values for the size of
the lateral ventricles in a large cohort













P
osthemorrhagic ventricular dila-
tion affects approximately 75%
severe germinal matrix-intraven
tricular
hemorrhage and represents a poten-
tial threat to the developing neonatal
brain (
1
). In infants with progressive
ventricular dilation, drainage of cerebro-
spinal 
uid has been shown to improve
cerebral hemodynamics
and oxygen-
ation and may prevent further brain
injury (
2,3
). To diagnose posthemor-
rhagic ventricular dilation and evalu-
ate the need for intervention, measure-
ment of v
entricular size by means of
cranial ultrasonography (US) has
been
shown to be superior to measur
ement
of head circumference or assessment
of clinical symptoms of raised intra-
cranial pressure (
4,5
). US measurement
of the lateral ventricles offers, in addi-
due to periventricular white matter loss.

In the early 1980s, Levene (
6
) was
the 
rst to publish reference values for
the size of the neonatal lateral ventri-
cles on cranial US images, and his curve

Implications for Patient Care

Our reference values should

n
enable early identi
cation and
accurate monitoring of posthem-
orrhagic ventricular dilation.

The longitudinal reference curves

n
enable recognition of ex vacuo
ventriculomegaly due to periven-
tricular white matter loss, which
may have prognostic
implications.

Advances in Knowledge

Cross-sectional reference curves

n
were established for the size of
the neonatal lateral ventricles at
cranial US, on the basis of a
large cohort of neonates born
gestation.

Longitudinal reference values

n
were provided for the follow-up
tional age,
,

30 weeks) until
term-equivalent age (TEA).

Ventricular reopening following

n
birth was shown to play only a
minor role, with an estimated
increase in anterior horn width
of 0.7 mm or less within the 1st
week of life.


n
enlarged ventricular size at TEA
compared with term-born infants
(ventricular index: +0.5 mm; an-
terior horn width: +1.2 mm;
thal-
amo-occipital distance: +3.5
mm).

Published online before print

10.1148/radiol.11110334
Content codes:


Radiology 2012;

262:224233
Abbreviations:
AHW = anterior horn width
GA = gestational age
TEA = term-equivalent age
TOD = thalamo-occipital distance
VI = ventricular index
Author contributions:
Guarantors of integrity of entire study, M.J.B., L.S.d.V., F.G.,
M.J.N.L.B.; study concepts/study design or data acquisition
drafting or manuscript revision for important intellectual
content, all authors; approval of  nal version of submitted
manuscript, all authors; literature research, M.J.B., L.R.P.,
M.J.N.L.B.; clinical studies, M.J.B., L.S.d.V., F.G., C.K.,
M.J.N.L.B.; statistical analysis, M.J.B., F.G., L.R.P., E.J.H.M.;
and manuscript editing, M.J.B., L.S.d.V., F.G., L.R.P.,
E.J.H.M., M.J.N.L.B.
Potential con icts of interest are listed at the end
of this article.
Radiology:
Volume 262: Number 1January 2012

n

radiology.rsna.org

233
PEDIATRIC IMAGING:
reference range
.
Arch Dis Child

1990
;
6
(
10
Spec No
):
1029

1032
.





27
.
Ichihashi

K
,
Takahashi

N
,
Honma

Y
,

Momoi

M
.
Cerebral ventricular volume assess-
ment by thr
ee-dimensional ultrasonography
.

J Perinat Med

2005
;
33
(
4
):
332

335
.





28
.
Winchester

P
,
Brill

PW
,
Cooper

R
,
Krauss

AN
,


HD
.
Prevalence of compressed
full-term neonates: sonographic study
.
AJR
Am J Roentgenol

1986
;
146
(
3
):
471

475
.





29
.
Boardman

JP
,
Counsell

SJ
,
Rueckert

D
,

.
Early growth in brain volume is pre-
.

Ann Neurol

2007
;
62
(
2
):
185

192
.





30
.
Inder

TE
,
War
eld

SK
,
Wang

H
,
Hppi

PS
,

Volpe

JJ
.
Abnormal cerebral structure is pre-
sent at term in premature infants
.
Pediatrics


2005
;
115
(
2
):
286

294
.





31
.
Nguyen The Tich

S
,
Anderson

PJ
,
Shimony


JS
,
Hunt

RW
,
Doyle

LW
,
Inder

TE
.
A novel
.
AJNR Am J
Neuror
adiol

2009
;
30
(
1
):
125

131
.





32
.
Kesler

SR
,
Ment

LR
,
Vohr

B
,
.
Volu-
.
Pediatr Neurol


2004
;
31
(
5
):
318

325
.





33
.
Nosarti

C
,
Al-Asady

MH
,
Frangou

S
,

Stewart

AL
,
Rifkin

L
,
Murray

RM
.
Adoles-
decreased brain volumes
.
Brain

2002
;
125
(
Pt 7
):
1616

1623
.





34
.

BS
,
Vohr

B
,
Staib

LH
,
.

Regional brain volume abnormalities and
infants
.
JAMA

2000
;
284
(
15
):
1939

1947
.





35
.
Maunu

J
,
Lehtonen

L
,
Lapinleimu

H
,
.

Ventricular dilatation in relation to out
come
prospective Finnish cohort study
.
Dev Med
Child Neurol

2011
;
53
(
1
):
48

54
.





36
.
Vollmer

B
,
Roth

S
,
Riley

K
,
.
Neurode-
ventricular dilatation with and without asso-
ciated haemorrhage
.
Dev Med Child Neurol


2006
;
48
(
5
):
348

352
.














16
.
Brann

BS

4th
,
Qualls

C
,
Wells

L
,
Papile

L
.

bral
ventricle in infants with posthemorrhagic
ventri
cular dilation
.
J Pediatr

1991
;
118
(
1
):

108

112
.





17
.
Grant

EG
,
Borts

FT
,
Schellinger

D
,

McCullough

DC
,
Sivasubramanian

KN
,

Smith

Y
.
Real-time ultrasonography of
neonatal intraventricular hemorrhage and
comparison with computed tomography
.

Radiology

1981
;
139
(
3
):
687

691
.





18
.
Batton

DG
,
Swails

TL
.
Isolated lateral ven-
infants: a left-sided lesion?

Am J Perinatol


1998
;
15
(
3
):
183

186
.





19
.
Enrquez

G
,
Correa

F
,
Lucaya

J
,
Piqueras

J
,

Aso

C
,
Ortega

A
.
Potential pitfalls in cra-
nial sonography
.
Pediatr Radiol

2003
;
33
(
2
):

110

117
.





20
.
Ichihashi

K
,
Iino

M
,
Eguchi

Y
,
Uchida

A
,

Honma

Y
,
Momoi

M
.
left
and right lateral ventricular sizes in neo
nates
.

Early Hum Dev

2002
;
68
(
1
):
55

64
.





21
.
Achiron

R
,
Yagel

S
,
Rotstein

Z
,
Inbar

O
,

Mashiach

S
,
Lipitz

S
.
Cerebral lateral ven-
nographic 

Gynecol

1997
;
89
(
2
):
233

237
.





22
.
Horsch

S
,
Bengtsson

J
,
Nordell

A
,

Lagercrantz

H
,
Adn

U
,
Blennow

M
.
Lat-
eral ventricular size in extremely premature
infants: 3D MRI con
rms 2D ultrasound
measurements
.
Ultrasound Med Biol

2009
;

35
(
3
):
360

366
.





23
.
Giedd

JN
,
Snell

JW
,
Lange

N
,
.
Quanti-
man
brain development: ages 4-18
.
Cereb Cortex


1996
;
6
(
4
):
551

560
.





24
.
Salomon

LJ
,
Bernard

JP
,
Ville

Y
.
Reference
nor-
mal approach
.


2007
;
30
(
1
):
61

66
.





25
.
Xenos

C
,
Sgouros

S
,
Natarajan

K
.
Ventricu-
lar volume change in childhood
.
J Neurosurg


2002
;
97
(
3
):
584

590
.





26
.
Saliba

E
,
Bertrand

P
,
Gold

F
,
Vaillant

MC
,

Laugier

J
.
Area of lateral ventricles measured
fants:
after severe intra-periventricular haemor-
.
Childs Nerv Syst


1992
;
8
(
1
):
33

35
.




6
.
Levene

MI
.
Measurement of the growth of
with real-time ultrasound
.
Arch Dis Child


1981
;
56
(
12
):
900

904
.




7
.
Davies

MW
,
Swaminathan

M
,
Chuang

SL
,


FR
.
Reference ranges for the
linear dimensions of the intracranial ven-
.
Arch Dis Child

2000
;
82
(
3
):
F218

F223
.




8
.
Perry

RN
,
Bowman

ED
,
Murton

LJ
,

Roy

RN
,
de Crespigny

LC
.
Ventricular size in
newborn infants
.
J Ultrasound Med

1985
;

4
(
9
):
475

477
.




9
.
Reeder

JD
,
Kaude

JV
,

ES
.
The occip-
ital horn of the lateral ventricles in prema-
ture infants. An ultrasonographic study
.
Eur
J Radiol

1983
;
3
(
2
):
148

150
.





10
.
Sondhi

V
,
Gupta

G
,
Gupta

PK
,
Patnaik

SK
,

Tshering

K
.
Establishment of nomograms
and reference ranges for intra-cranial ven
tric-
ular dimensions and ventriculo-hemispheric
ratio in newborns by ultrasonography
.

Acta P
aediatr

2008
;
97
(
6
):
738

744
.





11
.
Brouwer

MJ
,
de Vries

LS
,
Pistorius

L
,

Rademaker

KJ
,
Groenendaal

F
,
Benders

MJ
.

Ultrasound measurements of the lateral ven-
tricles in neonates: why, how and when? A
systematic review
.
Acta Paediatr

2010
;
99
(
9
):

1298

1306
.





12
.
Ivens

J
,
Martin

N
.
Grif
ths Scales
.
Arch Dis Child

2002
;
87
(
2
):

109

110
.





13
.
Brennan

P
,
Silman

A
.
for assessing observer variability in clinical
measures
.
BMJ

1992
;
304
(
6840
):
1491

1494
.





14
.
Royston

P
,
Wright

EM
.
How to construct
.
Ultra
sound

1998
;
11
(
1
):
30

38
.





15
.
Nelson

MD

Jr
,
Tavar

CJ
,

L
,
Kim

P
,

Gilles

FH
.
Changes in the size of the lat-
eral ventricles in the normal-term newborn
following vaginal delivery
.
Pediatr Radiol


2003
;
33
(
12
):
831

835
.

Radiology:
Volume 262: Number 1January 2012

n

radiology.rsna.org

231
PEDIATRIC IMAGING:
in the lateral ventricles and the mode of
delivery has been reported previously for
term infants (
28
). The effect of ventric-
ular reopening was small, h
owever, and
did not change the 97th percentile of the
reference curve for the AHW, which was
based on the 
rst cranial US after birth.

weeks gestation, longitudinal reference
curves were interpolated according to
ventricular US measurements following
born infants had larger ventricles than
term infants. Previous MR imaging
increase in ventricular size in the 
rst
days following birth. Others also dem-
onstrated a rapid increase in ventricular
area (
20,26
) and volume (
27
) at the end
of the 1st week and in the 2nd week af-
ter birth. We found that vaginal birth was
associated with a slightly smaller AHW
following birth and a gradual increase in
AHW in the 1st week, which supports
opening of the cerebral ventricles may
be related to the mechanical effect of a
vaginal delivery (
15
). A correlation be-
tween the amount of cerebrospinal 
uid

Figure 4


Figure 4:
Longitudinal reference curves for the
(a)
VI,
(b)
AHW, and
(c)

,
30 weeks) from birth up until TEA (
n
= 79);
presented are the estimated means and 95% reference intervals  tted to
(orange triangles).
studies have also demonstrated increased
cerebrospinal 
uid and ventricular vol-
pared with term-born control subjects
(
2931
). Differences in ventricular volume
have been shown to persist throughout
childhood and adolescence (
3234
) and
were also observed in infants (
29
) and
adolescents (
33
) without a
history of
posthemorrhagic ventric
ular dilation.
In the absence of major preceding brain
disease, ventricular enlargement in pre-
term infants seems a consequence of
more subtle brain injury with subsequent
volume loss of the adjacent periven
tricu-
lar white matter and subcortical gray
matter (
32,34
). Previous studies did
isolated ventriculomegaly and neurode-
infants. In the presence of other brain
disease, h
owever, v
entric
ular dilation was
shown to pose an addi
tional risk for cog-
nitive and motor impairments (
35,36


).

Our study is subject to several lim-
itations that need to be addressed.
The study period had to be extended
228

radiology.rsna.org
n
Radiology:
Volume 262: Number 1January 2012
PEDIATRIC IMAGING:
associated with a larger AHW at birth
but a smaller increase in AHW up until
TEA (
Table 3
). No other perinatal and
postnatal characteristics were associ-
ated with the increase in ventricular
size from birth to TEA.

The infants postmenstrual age ap-
peared to be most predictive for v
entric-
ular size. The most immature infants
(GA,
,

27 weeks) showed the largest
increase in VI and TOD throughout
the neonatal period. Male sex was

Table 1

Patient Characteristics at Birth and TEA
CharacteristicTotal Cohort (US 1) (
n
= 625)
Ventricular Reopening (US 1
and US 2) (
n
n
= 79)
Neonatal characteristics
GA at birth (wks) * 33.4 (24.742.6)30.1 (24.741.0)28.2 (25.929.9)
Birth weight (g) * 1900 (5305060)1310 (5304460)1120 (6001780)
No. of male patients365 (58)166 (55)41 (52)
Vaginal births340 (54)137 (46)42 (53)
Multiple births146 (23)89 (30)31 (39)
Monochorionicity40 (27)22 (25)4 (13)
Cranial US 1 and US 2
Interval birthUS 1 (h:min) * 14:24 (0:2395:34)3:44 (0:2244:46)1:57 (0:2243:46)
Interval birthUS 2 (h:min) * 67:49 (15:40160:35)
Interval US 1US 2 (h:min) * 54:00 (14:41154:24)

,
24 hours26 (9)
2448 hours99 (33)
4872 hours86 (29)
7296 hours49 (16)
96120 hours34 (11)


120 hours7 (2)
TEA US
Postmenstrual age (wks) *
41.6 (39.346.3)
Interval birthTEA US (wks) *
13.5 (10.017.7)
Note.Unless otherwise indicated, date are numbers of patients, and numbers in parentheses are percentages.
* Data are medians, and numbers in parentheses are ranges.

Table 2

US Examination and
MeasurementMean Right Ventricle * Mean Left Ventricle *
Median Left-Right
Difference


US 1 (
n
= 625)
VI10.3 (1.4)

10.4 (1.5)

0.0 (
2
2.12.5)
AHW1.1 (0.7)

1.2 (0.8)

0.1 (
2
2.52.5)
TOD15.6 (2.2)

16.0 (2.4)

0.3 (
2
7.06.9)
TEA US (
n
= 79)
VI12.6 (1.2)12.7 (1.4)0.0 (
2
1.43.3)
AHW2.2 (1.0)

2.6 (1.1)

0.3 (
2
2.53.4)
TOD19.0 (2.5)

19.7 (2.5)

0.8 (
2
3.55.8)
* Data in parentheses are standard deviation.


Data in parentheses are the range.


P

,
.05.


P

,
.001.

Compared with term-born infants
(
n

n

= 79) dem-
onstrated larger ventricles around TEA,
even after correction for differences in
postmenstrual age and head circumfer-
ence (
Table 4, Fig E1 [online]

).

Intra- and Interobserver Reliability

The reproducibility of ventricular mea-
surements by a single observer was con-
sidered very good (


0.81) for the AHW,
TOD, and right VI and good (0.610.80)
for the left VI according to the classi-

cation by Brennan and Silman (
13
).
observers, ventricular measure-
ments appeared also to be consistent.
Very good interobserver reliability was
observed for the right AHW and good
reliability was observed for the other
ventricular dimensions (
Table 5

).

Discussion

Sequential US measurements of the
neonatal lateral ventricles enable rec-
ognition of ventriculomegaly and further
Note: This copy is for your personal, non-commercial
use only. To order presentation-ready copies for
distribution to your colleagues or clients,

contact us at
www.rsna.org/rsnarights
.
ORIGINAL RESEARCH

n

PEDIATRIC IMAGING
224

radiology.rsna.org
n

Radiology:
Volume 262: Number 1January 2012
New Reference Values for the
Neonatal Cerebral Ventricles
1

Linda S. de Vries , MD , PhD
Floris Groenendaal , MD , PhD
Corine Koopman , MD , PhD
Lourens R. Pistorius , MD , PhD
Eduard J. H. Mulder , MSc , PhD
Manon J. N. L. Benders , MD , PhD

Purpose:
To establish new cross-sectional reference values for the
size of the lateral ventricles in a large cohort of neonates
as longitudinal reference values for the follow-up of very

Materials and
Institutional review board approval and parental written
informed consent were obtained for this prospective
cohort study of 625 neonates (58% male patients) with
a median GA of 33.4 weeks (range, 24.742.6 weeks).
All infants underwent cranial ultrasonography (US) within
4 days after birth to evaluate the size of the lateral ven-
neonates within the 1st week of life to assess the presence
of ventricu
(GA,
,

30 weeks) were prospectively included for cranial
US at term-equivalent age (TEA). US measurements were
performed of the ventricular index (VI), anterior horn
width (AHW), and thalamo-occipital distance (TOD). Sta-
tistical analysis was conducted by using a paired
t

test,
multilevel analysis, and analysis of covariance.

Results:
Cross-sectional reference values for the VI and TOD
increased with maturity, whereas the AHW remained
constant. V
aginal birth was independently associated with
a slightly smaller AHW following birth and with an increase
in AHW within the 1st week of life (
P


,

born infants showed a larger ventricular size at TEA com-
pared with term infants (
P


,

.001).

Conclusion:
New cross-sectional and longitudinal reference curves
were established for the size of the neonatal lateral ventri-
cles, which may allow for early identi
cation and quanti
-
cation of ventriculomegaly due to either posthemorrhagic
ventricular dilation or periventricular white matter loss.

q

RSNA, 2011
Supplemental material:
http://radiology.rsna.org/lookup
/suppl/doi:10.1148/radiol.
1111
0334/-/DC1


1
From the Department of Neonatology (M.J.B., L.S.d.V.,
Gynaecology (L.R.P., E.J.H.M.), University Medical Center
Utrecht/Wilhelmina Childrens Hospital, Post Box 85090,
2011; revision requested April 5; revision received May 27;
accepted June 23;  nal version accepted August 15. Sup-
ported by a grant from the Alexandre Suerman program for
MD/PhD students of the University Medical Center Utrecht,
Address correspondence to
M.J.N.L.B.
(e-mail:
[email protected]
).

q
RSNA, 2011
Radiology:
Volume 262: Number 1January 2012

n

radiology.rsna.org

227
PEDIATRIC IMAGING:


ventricle was observed both following
birth and at TEA. Absolute side-to-side
differences ranged up to more than 3 mm
for the VI and AHW and up to 7 mm
with larger (or smaller) measurements
of the other ventricular dimensions in
the ipsilateral hemisphere, and differ-
ences in left and right ventricular size
were not consistent over time. On av-
erage, ventricular size appeared slightly
larger on the left. Mean differences be-
tween lateral ventricles were, h
owever,
small and unlikely to be of clinical im-
portance (
Table 2

).

Cross-sectional Reference Curves

Cross-sectional reference curves were
designed for VI, AHW, and TOD accord-
ing to measurements at US 1 (
Fig 3

). GA
was associated with an increase in VI and
TOD, but did not in
uence the AHW.
Male neonates showed a slightly larger
ventricular size than female neonates
(
Table 3

). Birth weight, head circumfer-
ence, multiple birth, and monochorionic-
ity were not associated with ventricular
size in multiple regression analysis.

Ventricular Reopening

The presence of ventricular reopen-
ing following birth was assessed in a
subcohort of 301 neonates who had
undergone two subsequent US scans
within the 1st week (
Table 1
US 1 and US 2, an increase in AHW
was observed that was independent of
the infants GA and the presence or
duration of prolonged ruptured mem-
branes. Vaginal birth was associated
with
a slightly smaller AHW following
birth
and a gradual, compensatory increase
in the days thereafter (
Table 3
). No in-
crease in VI and TOD could be
demon-
strated in the 1st week after birth.

Longitudinal Reference Curves

Longitudinal reference curves were
designed for the VI, AHW, and TOD
(
Fig 4

) on the basis of measurements
at US 1 and TEA US in 79 prospec-
tively included infants born at less than
30 weeks gestation (
Table 1
).

The phenomenon of ventricular re-
opening following birth and its hypothe-
sized relationship with a vaginal delivery
were evaluated in a subcohort of term
gone two US scans in the 1st week of
life (
15
). The data of US 1 and US 2
were entered into a multilevel model,
as well as the presence of prolonged
ruptured membranes (
.

24 hours) and
the mode of delivery (vaginal birth vs
Caesarean section), while covarying for
the GA and sex of the infants and the
time intervals from birth to US 1 and
from US 1 to US 2.

Longitudinal reference curves for
the VI, AHW, and TOD were based on
measurements at US 1 and TEA US in a
subcohort of prospectively followed pre-
term infants (GA,
,

30 weeks). Estimates
of the mean and 95% reference intervals
were 
tted to the data using multilevel
previously described perinatal and post-
natal events and the observed increase
in ventricular size during the neonatal
period was tested in a multilevel model,
which was corrected for the infants
post
menstrual age at birth and around
infants (GA,
,

30 weeks) at TEA and
term neonates was compared by using
analysis of covariance, with postmen-
strual age and head circumference as

nal covariates.

Statistical analysis was performed
with statistical software (SPSS 18.0 for
Windows; SPSS, Chicago, Ill).

Results

Descriptive Results

A total of 625 neonates were enrolled
(
Table 1

were eligible for prospective inclusion at
TEA (
Fig 1
). US 1 was performed within
4 days after birth, mainly on day 1
(
n

= 417 [67%]) or day 2 (
n

= 130 [21%]).
US 2 was performed in 55% (301 of 547)
of the newborns in whom US 1 was con-
ducted within 48 hours after birth. In the
remaining neonates, it was not possible
to repeat cranial US within the 1st week
of life because of discharge from the hos-
pital or subsequent exclusion after US 1.

Statistical Analysis

right ventricle approached a normal dis-
tribution, a paired
t

test was performed
to assess 
rst the presence of ventric-
to be of clinical importance, measure-
ments of the left and right ventricles
were averaged in subsequent analysis
to enhance the clinical applicability and

Cross-sectional reference curves were
designed for the VI, AHW, and TOD
according to measurements at US 1.
Following logarithmic transformation of
the VI and TOD and square-root trans-
formation of the AHW, estimates of
the mean and 95% reference intervals
were calculated by using the approach
of regression modeling described by
Royston and Wright (
14
). Multiple re-
gression analysis was used to assess
the in
uence of sex, GA, birth weight,
head circumference, multiple birth,
and
monochorionicity.

Figure 2


Figure 2:
on cranial US images in the
(a)
coronal and
(b)

parasagittal plane.

Приложенные файлы

  • pdf 9590552
    Размер файла: 1 MB Загрузок: 0

Добавить комментарий