IMAGING OF THE FETAL CENTRAL NERVOUS SYSTEM

Lou Pistorius

IMAGING OF THE FETAL CENTRAL NERVOUS SYSTEM

Beeldvorming van het foetale centrale zenuwstelsel

(met een samenvatting in het Nederlands)

Proefschrift

ter verkrijging van de graad van doctor aan de Universiteit Utrecht op gezag van

de rector magnificus, prof. dr. J.C. Stoof, ingevolge het besluit van het college voor

promoties in het openbaar te verdedigen op dinsdag 11 november 2008 des middags

te 4:15 uur

door

Lourens Rasmus Pistorius

geboren op 27 april 1963 te Potchefstroom, Zuid Afrika

Promotoren: Prof.dr. G.H.A. Visser

Prof.dr. L.S. de Vries

Co-promotoren: Dr. R.H.J.M. Gooskens

Dr. F. Groenendaal

Dr. Ph. Stoutenbeek

ISBN: 978-90-393-49380

Layout and

publishing by: Gildeprint Drukkerijen B.V., Enschede, The Netherlands

Front page: Imagining the fetal brain (mola patchwork by Ronel Pistorius)

Publication of this thesis was financially supported by the University Medical Centre

Utrecht, BMA BV (Mosos), General Electric Healthcare and Toshiba Medical Systems.

Ek kry ‘n klein beiteltjie,

ek tik hom en hy klink;

toe slyp ek en ek slyp hom

totdat hy klink en blink.

N.P. van Wyk Louw

Vir Ronel, Wilma en Hantie

J

CONTENTS

Introduction 9

CHAPTER 1: Aims of the thesis 11

CHAPTER 2. Fetal neuro-imaging: ultrasound, MRI, or both? 15

Obstet Gynecol Survey 2008, in press

Physiology 45

CHAPTER 3. First trimester neurosonoembryology with automated follicle 47

tracking: Preliminary fi ndings

J Matern Fetal Neonatal Med 2008, in press

CHAPTER 4. Grade and symmetry of normal fetal cortical development: 55

a longitudinal two and three dimensional ultrasound study

Ultrasound Obstet Gynecol (provisionally accepted)

CHAPTER 5. A longitudinal ultrasound study of the size and symmetry 79

of the fetal lateral cerebral ventricles

Submitted

CHAPTER 6. Fetal cerebellar volume and symmetry on three-dimensional 95

ultrasound: volume measurement with multiplanar and VOCAL techniques

Ultrasound Med Biol 2008 (in press))

Pathology 109

CHAPTER 7. The role of prenatal ultrasound in predicting survival and 111

mental and motor functioning in children with spina bifi da

Ultrasound Obstet Gynecol 2008 (in press)

CHAPTER 8. Middle cerebral artery pulsatility index in fetal central nervous 127

system abnormalities

Ultrasound Obstet Gynecol (under revision)

CHAPTER 9. Disturbance of cerebral migration following congenital 141

Parvovirus B19 infection

Fetal Diagn Ther 2008 (in press)

CHAPTER 10. Fetal central nervous system abnormalities: ultrasound, 151

MRI or multidisciplinary discussions?

Submitted

CHAPTER 11. Lactate to Creatinine Ratio in Amniotic Fluid: a Pilot Study 163

Submitted

Summary 171

CHAPTER 12. General discussion and recommendations 173

Appendix 183

Nederlandse samenvatting 184

Afrikaanse opsomming 192

Acknowledgments 201

Curriculum vitae 203

List of publications 205

INTRODUCTION

Chapter 1AIMS OF THE THESIS

| Ch

apte

r 1

12

The fetal brain is the most complex organ and one that carries on developing in size,

intricacy and function throughout all three trimesters of pregnancy (and certainly

thereafter!) The fetal central nervous system and other fetal organs can be visualized

in increasingly fi ne detail as imaging equipment becomes ever more sophisticated. This

answers many questions, but continually poses new ones, as we have to re-evaluate

accepted knowledge in the light of new fi ndings. This forms the basis of this thesis:

an investigation of imaging of normal and abnormal development of the fetal central

nervous system.

INTRODUCTION

Chapter 2: review of the literature to evaluate the knowledge and gaps in the available

knowledge about ultrasound (US) and magnetic resonance imaging (MRI) imaging of

normal and abnormal development of the fetal central nervous system. We attempted

to identify the strengths and weaknesses of each modality from the literature, and to

arrive at practical recommendations for which conditions and at which gestational

age ultrasound, MRI or both would be most suitable.

PHYSIOLOGY

Chapter 3: to determine whether new ultrasound hardware and software can facilitate

imaging of fetal brain development in the late embryological period.

Chapter 4: evaluation of the grade and symmetry of cortical development in the latter

half of pregnancy as demonstrated in a longitudinal ultrasound study and to evaluate

whether three dimensional or two dimensional ultrasound is more reproducible or

time-consuming.

Chapter 5: assessment of the size and symmetry of the lateral cerebral ventricles in

the latter half of pregnancy as demonstrated in a longitudinal ultrasound study, and

to evaluate whether new technical developments such as volume contrast imaging are

more useful or reproducible than traditional methods.

Chapter 6: measurement of the fetal cerebellar volume and symmetry in the latter

half of pregnancy as demonstrated in a longitudinal ultrasound study, and to evaluate

which method of volume measurement has the best inter observer agreement.

| Aim

s o

f th

e th

esis

13

PATHOLOGY

Chapter 7: evaluation of the prediction of survival and mental and motor functioning

in children with spina bifi da by means of prenatal ultrasound imaging, and whether

the accuracy in determining the lesion level has improved with new, high resolution

and 3D equipment.

Chapter 8: investigation of the prognostic value of Doppler ultrasound measurement

of blood fl ow in the middle cerebral artery in fetuses with central nervous system

abnormalities.

Chapter 9: presentation of a case of a fetus who developed malformations of cortical

development after requiring intra-uterine transfusion for a severe Parvovirus B19

infection. This serves as introduction to the next aim, namely:

Chapter 10: evaluation whether the diagnosis made with ultrasound, with MRI or

during a multidisciplinary discussion was more accurate in diagnosing complex fetal

central nervous system abnormalities in a retrospective cohort.

Chapter 11: preliminary examination of the possibility to determine the degree of fetal

asphyxia by MR spectroscopy by measuring amniotic fl uid lactate levels.

Chapter 2FETAL NEURO-IMAGING:

ULTRASOUND, MRI, OR BOTH?

L. R. Pistorius1, P. M. Hellmann1, G.H.A. Visser1, G. Malinger2, D. Prayer3

1Department of Obstetrics, Division Perinatology and Gynaecology, University Medical Centre Utrecht, The Netherlands

2Prenatal Diagnosis Unit, and Fetal Neurology Clinic, Edith Wolfson Medical Center, Holon, Israel3Department of Neuroradiology, University Clinic of Radiodiagnostics, Vienna, Austria

Obstet Gynecol Survey 2008, in press

| Ch

apte

r 2

16

ABSTRACT

Much has been written recently about the relative merits and demerits of fetal imaging

with ultrasound and MRI. Unfortunately, the arguments often generate more heat

than light. We attempted to identify the strengths and weaknesses of each modality

from the literature, and to arrive at some practical recommendations on when to use

which imaging modality.

• In conclusion, combining neurosonography and MRI is mostly redundant, but

occasionally complementary.

• Both are operator dependant and neither technique obviates the need for a thorough

knowledge of normal and abnormal neuro-anatomy.

• In early pregnancy, and where repeated assessment is needed, ultrasound has the

obvious advantage.

• Where ultrasound is diffi cult, as in the obese patient or a patient with severe

oligohydramnios, better images might be obtained by MRI examination, although a

special MR system is required for patients weighing more than 150kg. MRI might also

identify fetal ischemic lesions early after an insult such as severe maternal trauma or

death of a monochorionic co-twin.

• There is a synergy between ultrasound and MRI for the diagnosis of certain conditions,

such as congenital cytomegalovirus infection or cerebellar telangiectasis.

Local conditions and expertise obviously infl uence the accuracy of both modalities.

Both ultrasound and MRI should be performed to the highest possible standard, and

the fi nal diagnosis should be made in a multidisciplinary setting.

| Fet

al n

euro

-im

agin

g: u

ltra

sou

nd

, MRI

, or

bo

th?

17

INTRODUCTION

Much has been written about the merits and demerits of fetal imaging with ultrasound

and MRI. The arguments often generate more heat than light. What is needed is an

evaluation of the strengths and weaknesses of each modality as illustrated by the

literature, and practical recommendations on when to use which modality (Table 1) .

Table 1. Recommendations: ultrasound, MRI or a combination of both in fetal brain imaging

Ultrasound preferred: Repeated examinationsScreeningMRI contra-indicated or failedEarly diagnosis (< 20w)Assessment of fetal movements (up to 3rd trimester)Assessment of cerebral blood fl owEvaluation of associated (especially cardiac) abnormalities

MRI preferred: OligohydramniosEngaged fetal head and ruptured membranesAssessment of fetal movements (late 3rd trimester)Detecting and determining age of intracranial bleedingDetecting intracranial tuberous sclerosisSchizencephalyAcute asphyxiaPostmortem brain imagingPosterior fossa abnormalitiesSevere microcephaly

Combination preferred: Suspected cerebellar telangiectasisFetal brain deathCytomegaloviral infectionIntracranial tumorsTraumaVein of Galen abnormalitiesGerminal matrix and intraventricular bleedingSuspected hemimegalencephalySepto-optic dysplasia

Either US or MRI appropriate: VentriculomegalyHoloprosencephaly (after 20 weeks)Corpus callosum abnormalities (after 20 weeks)Craniosynostosis

A detailed description of technical aspects is beyond the scope of this review, but the

basics are explained in a thesaurus (Table 2).

| Ch

apte

r 2

18

Table 2. Thesaurus

Ultrasound – see also (1:2)

• How does ultrasound work? Electrical impulses cause piezo-electrical crystals in an ultrasound transducer to produce ultrasound waves. The ultrasound image is generated by the refl ection of ultrasound waves onto the transducer, in which the piezo-electrical crystals translate sound energy again into electrical impulses. These impulses are then depicted on a display and arranged longitudinally according to the amount of time it took since the impulse was sent out, and laterally according to the place on the transducer array where the impulse was received.

• A-mode ultrasound: ultrasound imaging with a one-dimensional display, where the distance between structures producing echoes and the amplitude of the echoes are represented by a graph. This was the only available ultrasound prior to the 1970’s, and is currently used mainly in ophthalmology where precise measurements are required.

• Grey scale: a display where the amplitude of the echo is represented by the intensity of a dot of light on a screen.

• M-mode ultrasound: ultrasound imaging where A-mode ultrasound images are repeated over time and displayed consecutively in a grey scale. This is used for displaying heart muscle and valve leafl et movements.

• B-mode ultrasound: cross-sectional ultrasound imaging with a two-dimensional display – in other words, what most of us are using most of the time.

• 3D ultrasound: ultrasound imaging where ultrasound is acquired as a volume rather than cross-sectional; this can be displayed in multiplanar mode (where three orthogonal, or cross-sectional planes perpendicular to each other, are displayed) or rendered mode, where a volume is displayed as it would appear to the naked eye

• Neurosonogram: a multiplanar ultrasound examination of the central nervous system (CNS), performed where there is an increased risk of a CNS abnormality or where a screening examination was suggestive of CNS abnormality (3)

MRI – see also (4:5)

• How does MRI work?: Normally, individual MR-active atomic nuclei (such as hydrogen nuclei) are randomly aligned, and each has its own magnetic fi eld which is caused by its rotation. When an external magnetic fi eld is applied, the nuclei are aligned parallel or in opposition to the magnetic fi eld. The cumulative magnetic fi eld of all the nuclei is called the net magnetization vector. If a radiofrequency (RF) pulse is applied, the net magnetization vector is fl ipped by a certain angle, with a longitudinal and transverse component. The transverse component induces a current in a receiver coil, which is translated into a MR signal. The nuclei gradually move out of phase and the signal in the coil therefore decreases (or “decays”).

• T1 recovery involves the recovery of longitudinal magnetization because of energy released into the environment.

• T2 decay involves a decrease of the transverse magnetization because of the individual nuclei moving out of phase and interaction of the individual magnetic fi elds.

• Tissue contrast on MRI is the result of differences between T1 recovery, T2 decay and proton (hydrogen nuclei) density.

• Repetition time (TR) is the time between two different RF pulses.• Echo time (TE) is the time from the RF pulse until the MR signal is measured.• With a short TR, the difference between the T1 recovery of water and fat signals is maximal.• With a long TE, the difference between the T2 decay of water and fat signals is maximal.• For T1 weighted imaging, both the TR and TE are therefore short. Water appears hypointense and fat

hyperintense• For T2 weighted imaging, both the TR and TE are therefore long. Water appears hyperintense.• If the TR is long and the TE is short, there is little difference between the T1 recovery or T2 decay of water

and fat, and contrast is due to difference in proton density.• The two fundamental MR pulse sequences are the SE (Spin echo) and GRE (gradient echo)• FLAIR (Fluid attenuated inversion-recovery) is used to suppress the water signal, by applying a RF pulse at

a time when the net magnetization vector of the water signal is very weak therefore enhancing contrast

| Fet

al n

euro

-im

agin

g: u

ltra

sou

nd

, MRI

, or

bo

th?

19

in tissues (with a bright signal) and suppressing the water signal (which would appear dark.) STIR (short T1 inversion recovery) is used in a similar way to suppress the fat signal.

• Single shot fast spin echo (SSFSE) and half-Fourier acquisition turbo spin echo (HASTE) are different terms used by different manufacturers of MR systems for essentially the same thing: a multiple-echo variant of SE which allows imaging in a shorter time, enabling fetal MRI without sedation.

• Diffusion weighted imaging (DWI) allows the distinction between protons (or water) diffusing freely or in a restricted way. This is useful to detect anisotropy, where there is a difference of movement in a specifi c direction, such as is the case in premyelinating structures such as the corpus callosum in early pregnancy. DWI is also useful in distinguishing between vasogenic and cytotoxic oedema.

• The ADC (apparent diffusion coeffi cient) is a way of quantifying DWI by means of the ratio of two image sets; one obtained with and one without a diffusion gradient. This can be used or to predict lung maturity by detecting the anisotropy caused by vascularization of terminal tubules.

• Other factors that infl uence the quality of the image include the positioning of the coil, with the receiver ideally positioned directly over the region of interest. A small fi eld of view is optimal (as it allows maximal resolution), but might increase the risk of artefacts.

• The frame rate, or number of images obtained per second, depends on the time to complete one sequence. Dynamic sequences can be obtained at frame rates of up to 6 frames per second, which allow the observation of fetal limb, breathing and swallowing movements.

• The slice thickness refers to the thickness of the slice through the target organ which is averaged to obtain the MRI signal. The thicker the slice, the fewer slices are needed, and the less the total imaging time, but the ability to detect a small structure or lesion decreases.

• MRS (magnetic resonance spectroscopy) provides information on the chemical composition of a target volume. Each different molecule has a different electron cloud, which causes a slight shift in frequency of the magnetic fi eld. This chemical shift is a very small effect, and measured in parts per million (ppm) of the magnetic fi eld. PMRS (Proton MRS) is used to evaluate levels of inositol, choline, aspartate and lactate in the brain, and phosphorus MRS is used to evaluate products of energy metabolism, and estimate pH levels. Because of the tiny effect, the magnetic fi eld has to be calibrated (“shimmed”) very carefully.

• BOLD (blood oxygen level dependant) imaging depends on the difference of magnetic properties of oxyhemoglobin (not magnetic) and deoxyhemoglobin (magnetisable) to generate different MRI signals depending on the oxygen saturation. This can be used to map brain activity (with an increase in deoxyhemoglobin in activated parts of the brain because of an increase in local oxygen consumption).

ULTRASOUND

General aspects

Since the early days of ultrasound, it has been possible to diagnose abnormalities of the

fetal brain using A-mode ultrasound (6). With the advent of B-mode ultrasound, two-

dimensional images of the fetal brain could be obtained. Subsequent improvements

in resolution have been so dramatic that it is now possible to visualize the early

embryonic development of the central nervous system in vivo (7-10). Each new

generation of ultrasound machines is so impressive that it is easy to forget how much

of an improvement the previous generation of machines had been.

The advantages of using three-dimensional ultrasound to evaluate the fetal brain

include the ability to visualize planes which would otherwise be inaccessible and to

| Ch

apte

r 2

20

accurately measure the volumes of structures such as the cerebellum (11-13). Three

dimensional ultrasound might thus be more useful than two-dimensional diameters

in diagnosing, monitoring and understanding pathology.

The fetal neurosonogram

Traditionally, the brain was examined in axial planes. Evaluation of these planes

is widely used as a screening tool (14;15). However, examining the brain in coronal

and sagittal planes by aligning the transducer with the fontanelles is also possible,

sometimes by means of gentle manipulation of the fetus (16).

The fetal neurosonogram is a detailed, multiplanar ultrasound examination of

the central nervous system performed by someone with specifi c expertise using

sophisticated equipment. It is performed when there is an increased risk of a CNS

abnormality or when a basic or screening ultrasound examination suggests a CNS

abnormality. If the fetus is in cephalic presentation, transvaginal ultrasonography can

be used to allow higher ultrasound frequencies with a higher resolution. Although

not yet universally applied, fetal neurosonography is of greater diagnostic value than

conventional ultrasound, especially in fetuses with complex malformations (3).

Structural evaluation

It is diffi cult to imagine a part of the fetal brain for which a range of normal values has

not been described by ultrasound imaging. These include the lateral (17), third (18), and

fourth (19) ventricles, corpus callosum (20;21), cavum septum pellucidum (22), cerebellar

diameter and vermis (23;24), cisterna magna (22), pons (25), subarachnoid space (26),

and even the intracranial circulation (27) and patterns of cortical development (28-

31). Using mainly transvaginal ultrasound, it is possible to diagnose CNS pathology as

early as the late fi rst trimester. In addition to exencephaly (32), the presence of Dandy-

Walker malformations (33), an encephalocoele, hydranencephaly or holoprosencephaly

(34-36) have been diagnosed in the fi rst trimester. Later in pregnancy, the range of

brain abnormalities, including malformations of cortical development, which can be

diagnosed by ultrasound, is too great to mention in detail (37).

Evaluation of perfusion

Power or color Doppler ultrasound is useful to delineate vascular lesions such as an

aneurysm of the vein of Galen or other intracranial arteriovenous malformations

(3;38). Decreased regional fl ow in the circle of Willis or carotid artery can help to

determine the origin of pathology such as schizencephaly (39). Evaluation of middle

| Fet

al n

euro

-im

agin

g: u

ltra

sou

nd

, MRI

, or

bo

th?

21

cerebral arterial fl ow is widely used in the management of growth-restricted fetuses

and in the diagnosis of fetal anemia (40).

Functional evaluation

Fetal movements are routinely evaluated as part of the biophysical profi le, which

assesses fetal well-being (41). Evaluation of fetal movements can also be used

to asses the integrity of the fetal nervous system (42). Although the assessment

of fetal movements has not yet been adopted as a standard component of fetal

neurosonography (3), it may be used more widely in the near future (43).

Safety

Despite concerns recently raised as result of animal studies (44), there is no evidence

that the human fetus is harmed while performing diagnostic ultrasound according to

the ALARA (as low as reasonably achievable) principle (45).

MAGNETIC RESONANCE IMAGING

General aspects

Despite the advantages of ultrasound, MRI is useful when there is a relative contra-

indication to transvaginal ultrasound such as vaginismus (46) or ruptured membranes

and when it is diffi cult to obtain a good acoustic window transabdominally, such as in a

fetus with a well engaged head or an obese patient. Morbidly obese patients weighing

more than 150kg cannot be examined in a traditional MRI, and have to be examined in

a large bore cylindrical or vertical open MRI system.

MRI tissue contrast is infl uenced by the fat and water content and proton density of

tissues. The signals can be enhanced or suppressed by manipulating the repetition

time or echo time, or by using specifi c techniques to nullify the fat or water signal (4).

Fast, ultra-fast and real-time sequences have made fetal imaging possible without the

need for fetal paralysis (47;48), although sedation is still sometimes used (49-52).

T2 weighted imaging is typically used to demonstrate fetal anatomy. T1 weighted

imaging is useful to depict hemorrhages, calcifi cations or lipomas (5), as well as the

cortical plate, brain stem and basal ganglia in late gestation. Diffusion-weighted

imaging is sensitive for the detection of hypoxic-ischemic brain lesions and useful for

identifying premyelinating structures, such as the corpus callosum, in early pregnancy (5).

| Ch

apte

r 2

22

Structural evaluation

The normal development of the cerebral cortex has been well defi ned with fetal MRI

(53). A typical fetal lamination pattern is seen on T1WI from 23 to 28 weeks’ gestation

(54) and refl ects the migration of neurons from the ventricular and subventricular

zones to the cerebral cortex, in which six histological layers can be seen. Growing

pathways that cross in the periventricular region (the “periventricular crossroads”) can

be demonstrated by MRI, especially from 29 to 32 weeks, and are seen as periventricular

echodensities by ultrasound (55). Anatomically, this area is a grid of projection,

commissural and associative fi bers embedded in the extracellular matrix, which is

increased and rich in axonal guidance molecules during this stage of development.

This is the area where periventricular leukomalacia is most likely to develop (56). The

development of gyri is also well described on MRI (57-59). Although myelination occurs

mainly postnatally, white matter tracts can be observed prenatally by anisotropy on

diffusion weighted imaging. In addition, development of the hippocampus, basal

ganglia, brain stem, and cerebellum are all well described by MRI studies (53).

In case of stillbirth or neonatal demise, MRI may be a useful adjunct to autopsy,

by helping to guide sampling for post-mortem histological diagnosis, and may be

particularly useful for evaluation of the central nervous and skeletal systems (60-62).

Evaluation of perfusion, bleeding and ischemia

T1WI is a sensitive modality to detect the presence and the age of an intracranial

hemorrhage (5). An early sub acute bleed gives rise to a hyper-intense signal on T1WI

and a hypo-intense signal on T2WI (63). Hemoglobin can also be detected with gradient

echo (GRE) sequences which detect the signal loss caused by local magnetization of the

iron in hemoglobin. Although the distinction between acute (1-3 days), early subacute

(3–7 days) and late subacute (7-30 days) hemorrhages can be made with a combination

of T1WI and T2WI in the neonate and adult (64), this has not been reported in a fetus.

It is possible that the difference between fetal and adult hemoglobin would infl uence

the appearance and accuracy of antenatal imaging (5).

The apparent diffusion coeffi cient (ADC) is obtained by subtracting images with and

without the application of a diffusion gradient, and decreases within minutes to

an hour of the occurrence of ischemia (65) due to intracellular fl uid retention (66).

Combining T1WI, T2WI and ADC might be useful in determining the age of a prenatal

infarct (67-69). However, the age of a hemorrhage or infarct cannot be determined by

MRI with suffi cient accuracy for forensic or legal purposes.

| Fet

al n

euro

-im

agin

g: u

ltra

sou

nd

, MRI

, or

bo

th?

23

Functional evaluation

Fetal movement can also be evaluated by means of dynamic MRI sequences with a

large fi eld of view (FOV) and at a frame rate of 4–6 per second (5). Although not of

immediate clinical value, in the future, specialized techniques may allow determination

of brain oxygenation by means of blood oxygen level-dependent (BOLD) imaging (70)

by functional MRI (FMRI) as early as 33 weeks (71), measurement of pO2 by means

of quantitative images of fl uid oxygenation on standard single-shot fast-spin echo

(SSFSE) sequences (72), or determination of fetal brain lactate by magnetic resonance

spectroscopy (MRS).

Safety

Although widely regarded to be safe, an MRI examination does expose the fetus to noise,

albeit fi ltered by the maternal abdominal and uterine walls, and to a static magnetic

fi eld ten thousand times stronger than the earth’s (73), as well as radiofrequency and

time-varying gradient electromagnetic fi elds. As is the case with ultrasound, there is

no evidence that the human fetus is harmed while performing MRI examinations in

pregnancy (74), while using 1.5T apparatus (4). A ten-year follow-up study of infants

exposed to MRI in utero is underway (75).

ULTRASOUND OR MRI?

With two such powerful techniques available, it is no surprise that the literature

abounds with publications evaluating the added value of MRI relative to ultrasound

for the diagnosis of fetal brain abnormalities. The publications range from case reports

to case series and reviews, but only few compare high quality neurosonography with

high quality MRI. It is diffi cult to pool the data from very different articles, but the

available data are presented below, according to a classifi cation used in pediatric

neuroradiology (76).

Failure of dorsal induction

Anencephaly is reliably diagnosed by ultrasound in the late fi rst trimester (77) when

MRI is of little value. Although encephalocoeles can also be well delineated by axial

ultrasound, neurosonography and MRI may be of value in helping to diagnose possible

additional abnormalities, such as the posterior fossa abnormalities associated with

| Ch

apte

r 2

24

Joubert syndrome, before 24 weeks’ gestation (78). The intracranial signs of spina bifi da,

such as the Chiari malformation, are also diagnosed accurately by ultrasound, whether

early in pregnancy or after 24 weeks (79). MRI might be of value in delineating posterior

fossa lesions more accurately (80), and is as accurate as ultrasound for determining the

anatomical level of the spinal defect between 20 and 30 weeks. Because the functional

neurologic level might not correspond to the anatomic level, the ultimate neurologic

outcome cannot be predicted with accuracy by either modality (81).

Failure of ventral induction

Holoprosencephaly

Holoprosencephaly has been diagnosed prior to 10 weeks by means of transvaginal

ultrasound (34;35). However, in other studies using lower resolution ultrasound, both

semilobar and lobar holoprosencephaly have been correctly diagnosed by MRI after 20

weeks’ gestation in fetuses with an ultrasound diagnosis of ventriculomegaly (82;83).

Cerebellar and posterior fossa abnormalities

Several authors have found MRI useful for the evaluation of the posterior fossa.

Examples include providing additional information on areas of hemorrhage in a

tumor during the third trimester (49), excluding false positive ultrasound fi ndings (84),

diagnoses of posterior fossa cysts (85;86), or identifi cation of pontine abnormalities

after 20 weeks’ gestation (87;88). MRI fi ndings have also infl uenced the management

of pregnancies in which the fetus has a posterior fossa cyst between 21 and 39 weeks

(89;90), unilateral cerebellar defi ciency at 25 weeks (91) or Chiari malformation after

24 weeks’ gestation (82).

The posterior fossa has remained an area where MRI excels. In one study, half (6/12) of

ultrasound diagnoses of the posterior fossa were proved wrong by postnatal imaging or

autopsy (92). However, these were all transabdominal ultrasound examinations, and it

can be diffi cult to visualize the posterior fossa well on transabdominal ultrasound in the

late third trimester. Transvaginal ultrasound was necessary in two studies between 20

and 33 weeks’ gestation to accurately distinguish between conditions such as rotation

versus hypoplasia of the vermis, two conditions with very different prognoses (93-

97). In one study, prenatal MRI diagnoses of vermian agenesis, brain stem hypoplasia,

or destructive lesions of the posterior fossa at 25 to 36 weeks’ gestation correlated

well with postnatal fi ndings, but MRI incorrectly diagnosed vermian hypoplasia in

4 of 12 cases and had diffi culty identifying cerebellar hypoplasia at 32 to 36 weeks’

(98). The development of the posterior fossa can be seen by MRI (99) but structures

| Fet

al n

euro

-im

agin

g: u

ltra

sou

nd

, MRI

, or

bo

th?

25

such as the vermian fi ssures are seen by MRI as much as fi ve weeks later than their

fi rst visualization in anatomic studies. This can be explained by the fact that MRI may

fail to identify small structures or lesions situated between consecutive planes (100).

Cerebellar and pontine hypoplasia can be seen with either MRI or ultrasound, but more

detailed visualisation of the brain stem can be diffi cult prenatally with either modality

(101). Whichever technique is used, it is essential to come to a standardized anatomic-

developmental diagnosis (102).

Failure of neuronal proliferation, differentiation and histogenesis

Hemimegalencephaly

Asymmetry of the cerebral hemispheres or ventricles at 20 weeks’ gestation can lead to

the suspicion of hemimegalencephaly (103). MRI is useful for depicting the associated

accelerated myelination antenatally at 35 weeks’ gestation or postnatally (89;104),

increased cellularity at 32 weeks (105), and abnormal position of the ipsilateral ventricle

postnatally (106). However, the diagnosis of asymmetry is fraught with uncertainty.

Asymmetric ventriculomegaly can be identifi ed in less than 1% of fetuses, whereas

it is seen in 21-44% of neonates (107). This might refl ect a real difference, or the fact

that usually only one ventricle is visualized clearly prenatally (108). However, in vitro

anatomic studies show that the left cerebral hemisphere may be up to 20% larger

than the right (109;110), and there is some evidence of structural (111) and functional

(112) asymmetry identifi ed by ultrasound. In cases with pathology, there seems to be

a larger degree of asymmetry of the ventricular volume when measured by MRI (113).

Asymmetry of the development of the temporal lobe has also been demonstrated

with MRI, with the right side developing earlier (Unpublished data, Kasprian, G).

Tumors

Doppler ultrasound can identify tumors with increased blood fl ow at 35 weeks (114).

Hemorrhage into a tumor has also been suggested by a sudden increase in tumor

size or the development of a mass within the mass at 39 weeks’ gestation (115).

MRI can also be used to determine cellularity (66), exclude areas of hemorrhage or

necrosis, distinguish tumors from hematomas, and confi rm the presence of fat in

lipomas (116). The intracranial nodules of tuberous sclerosis have been consistently

found with greater sensitivity by MRI, both antenatally between 20 and 32 weeks’

gestation or postnatally, usually in fetuses in which an intracardiac mass indicative of

a rhabdomyoma was found (117-122).

| Ch

apte

r 2

26

Microcephaly

Microcephaly has traditionally been defi ned as a head circumference more than three

standard deviations below the mean. In cases in which the head is not quite this small,

it may be useful to assess the (relative) size of the frontal lobe (123;124) and to assess

cerebral blood fl ow with Doppler ultrasound (125). In severe cases of microcephaly, in

which the acoustic window is very limited due to small fontanelles, MRI may be of use.

Microcephaly, like ventriculomegaly, is anatomic fi nding rather than a diagnosis, and

should prompt a search for the underlying cause.

Failure of neuronal migration

Schizencephaly, lissencephaly, pachygria, polymicrogyria and neuronal heterotopias

MRI often adds to the diagnosis of malformations of cortical development between 21

and 32 weeks’ gestation (82;85;119;126-131). This might be partially due to an inherent

advantage of MRI. Although it is feasible to evaluate cortical gyration and sulcation by

means of either MRI or ultrasound in the second half of pregnancy (28-31;132;133),

myelination and migration can only be evaluated by MRI (134-136). Myelination occurs

predominantly after birth, but the protein and lipid signals of the white matter sheath

can be seen by MRI from 30 weeks onward, and the premyelinating formation of white

matter tracts can be demonstrated antenatally with diffusion weighted imaging

(53;137). The dynamic changes in the layered appearance of the cortex also help to

confi rm normal migration (138). However, development as seen by MRI can lag behind

anatomic studies by up to two weeks due to factors such as slice thickness or MRI planes

which are not quite orthogonal (135;139). Ultrasound can also identify malformations

of cortical development, such as an increased echogenicity found at 25 weeks’

gestation in areas where polymicrogyria later became evident (140), ventriculomegaly,

midline cysts, and abnormal opercular formation found at 24 to 32 weeks’ gestation

(141), or abnormal appearing gyri and sulci (142). A retrospective study confi rmed that

it is possible to identify delayed cortical development on ultrasound images obtained

between 20 and 33 weeks’ gestation in fetuses that were subsequently proved to

have Miller-Dieker syndrome (143). In another, prospective study, 23 out of 24 cases

of abnormal cortical development were diagnosed by ultrasound between 18 and 40

weeks. The 24th case had periventricular nodular heterotopia, which was missed by

both ultrasound and prenatal MRI (37). Visualization of polymicrogyria may be age

dependent, as such malformations may develop secondary to vascular problems and

thus will not manifest before the end of the second trimester (144). MRI may also

| Fet

al n

euro

-im

agin

g: u

ltra

sou

nd

, MRI

, or

bo

th?

27

be better than ultrasound for differentiating schizencephaly from other differential

diagnoses prenatally (32 - 38 weeks) and postnatally (65;145).

Corpus callosum

Although the callosum can be measured in detail by ultrasound from 16 weeks onward

(20;21), and the premyelinated corpus callosum does not show up well on MRI before

late gestation (47), it is interesting that MRI is better in diagnosing partial or complete

absence of the corpus callosum in many studies done between 21 and 34 weeks’

gestation (50;82;83;85-87;92;119;122;129;146;147). However, this probably refl ects

poorly performed ultrasound studies rather than an inherent technical disadvantage

of ultrasound. In studies in which a midsagittal view was used to evaluate the corpus

callosum, ultrasound was shown to be as good (95;148) or better (149) than MRI for

the diagnosis of corpus callosum abnormalities between 21 and 34 weeks’ gestation.

In addition to indirect signs such as colpocephaly, the corpus callosum can be visualised

directly on multiplanar neurosonography to diagnose complete or partial absence of

the corpus callosum (148). Three dimensional and Doppler ultrasound (95) may also be

of benefi t for visualizing the midline structures and the pericallosal arteries (148).

The optic chiasm is well visualized by MRI from 19 weeks onward (150) and by three

dimensional ultrasound from 22 weeks (11). Olfactory bulbs and sulci are visible by

MRI at 30 to 34 weeks and by ultrasound from 24 weeks (29).

Failure of myelination

Failure of myelination may be a sign of a metabolic defect. However, during fetal life

the maternal metabolism can compensate for metabolic defects in the fetus, with the

result that abnormal myelination as a morphological manifestations of these disorders

is rarely seen prenatally or at birth. Some metabolic disorders may be morphologically

identifi ed prenatally because they are associated with different pathology, such as

parenchchymal lesions, which are found in pyruvate dehydrogenase defi ciency, or

cysts and impaired cortical development, which are common in Zellwegers disease

(144). Defi nitive prenatal diagnosis requires chorionic villus sampling or amniocentesis

for molecular genetic testing, and is generally done when the fetus is at increased risk

because there is a family history of the disease (151).

| Ch

apte

r 2

28

Acquired Lesions

Ischemia

Ischemia might be demonstrated by MRI minutes to hours after an insult in the third

trimester, if it results in a hyperintense DWI signal and reduced ADC (66;67;152). Chronic

changes are more commonly seen, and appear as increased T2WI and decreased T1WI

signals due to subcortical leukomalacia (51). In one study (154), intracranial ultrasound

abnormalities were seen sonographically at 21 to 28 weeks’ gestation within hours to

two weeks after the death of a monochorial co-twin, and indicated the development

of cortical abnormalities. However, in another case at 23 weeks (126), the same

diagnosis was not suspected after axial ultrasound but was made by MRI. In cases of

fetal stroke, MRI has demonstrated hemorrhagic lesions in 90%, and porencephaly in

13% of affected fetuses (153). In a severe case, a combination of cardiotocography,

Doppler ultrasound, and MRI made it possible to diagnose fetal brain death at 23 weeks’

gestation (154). MRI has also added to the clinical information about fetuses with a

sonographic diagnosis of porencephaly, schizencephaly, periventricular leukomalacia

and other lesions of the brain parenchyma between 21 and 34 weeks’ (49;50;83;86;

126;129;155).

Infections

Intra-uterine infection with cytomegalovirus can lead to microcephaly, ventriculomegaly

and porencephaly, which are easily seen by ultrasound at 29 weeks’ gestation (156).

More subtle signs of infection include borderline ventriculomegaly with focal areas of

increased echogenicity and small cysts which are situated symmetrically in the lateral

periventricular areas at 22 to 29 weeks’ gestation (157;158). The combination of this

typical ultrasound fi nding with an MRI study showing hyper-intense germinal matrices

bulging into the lateral ventricles at 33 to 35 weeks is very specifi c to the diagnosis of

congenital cytomegalovirus infection (159), and is another example of how ultrasound

and MRI can complement each other. Another case demonstrating synergy between

ultrasound and MRI and the dynamic nature of fetal diagnosis involved a fetus in whom

ventriculomegaly and a large posterior fossa were initially diagnosed by ultrasound.

MRI demonstrated cerebellar hypoplasia and inappropriate cortical gyrations at 34

weeks’ gestation. Days later, parenchymal calcifi cations were seen by ultrasound

(49) and congenital cytomegalovirus infection was confi rmed. In a similar case, MRI

demonstrated lissencephaly and cerebellar atrophy at 37 to 39 weeks’ gestation (160).

In the rare case of congenital herpes infection, severe parenchymal destruction might

| Fet

al n

euro

-im

agin

g: u

ltra

sou

nd

, MRI

, or

bo

th?

29

be seen on MRI. Diffusion weighted imaging might be useful for the detection and

monitoring of brain abscesses and herpetic encephalopathy from 35 weeks’ onward

(161).

Trauma

Although our knowledge of the prevalence and evolution of fetal neurological damage

resulting from maternal trauma is limited, fetal central nervous system pathology

such as schizencephaly, vermian haemorrhage (162) and epidural haemorrhage (163)

have been noted by both ultrasound and MRI after maternal trauma as early as at 23

weeks’ gestation.

Germinal matrix bleeding

Germinal matrix and intraventricular bleeding (GMIVH) is another example of an

abnormality for which accurate diagnosis is more likely when both ultrasound and

MRI are used. Ventriculomegaly resulting from GMIVH is visible by ultrasound, with

blood appearing as hyper- or isoechogenic material inside the ventricle. Using MRI,

this material appears as a structure with a central hypointense area surrounded by a

hyperintense signal on T1WI and mixed hypo- and isointensity on T2WI. Hemosiderin

might also be visualized in the parenchyma as a hypointense signal on T2WI from 24

to 33 weeks’ gestation. MRI can identify hemorrhage, even without a mass effect such

as seen with grade I hemorrhage (52;164).

Brainstem watershed injuries

Watershed injuries of the brainstem might cause neurologic abnormalities that can

lead to conditions like Möbius syndrome or the Pierre-Robin sequence, but are beyond

the resolution of antenatal ultrasound or MRI (165). However, when both ultrasound

and MRI have been used at 24 to 29 weeks’ gestation to evaluate an echogenic area in

the cerebellum, hemorrhage (166) could be distinguished from telangiectasis (167).

Ventriculomegaly

Ventriculomegaly is a description, not a diagnosis. Aqueduct stenosis can be inferred

from the presence of dilated lateral and third ventricles with a small fourth ventricle,

along with the absence of any sign of intraventricular bleeding or other intracranial

pathology (168). Sonographic screening for ventriculomegaly is usually done by

evaluating the width of the ventricle at the atrium, in the axial plane (3;169). There is

| Ch

apte

r 2

30

a good correlation between ultrasound and MRI measurements of the ventricles (170).

Normal values have been established for measurements of the lateral, third and fourth

ventricles from 12 to 40 weeks’ gestation (17-19;171). When MRI is done at 21 to 37

weeks for the sonographic diagnosis of fetal ventriculomegaly, additional fi ndings have

included lissencephaly (172), holoprosencephaly (83), meningocoele (50) and evidence

of intraventricular hemorrhage (49;63;90;173;174). Conversely, ventriculomegaly has

been ruled out in fetuses between 19 and 33 weeks’ gestation after an ultrasound

exam has suggested mild asymmetrical ventriculomegaly (49;90;91;149;173). MRI

fi ndings that have prompted a change in management include periventricular

leukomalacia or other evidence of ischemic brain injury (83;91), lissencephaly,

polymicrogyria or cortical heterotopia at 25 to 37 weeks’ gestation (83;90;119;174-

177), hydranencephaly (82), holoprosencephaly (82;83), Chiari malformation (82),

corpus callosum agenesis (83;90;119), haemorrhage (83), vein of Galen malformation

(83), Dandy Walker malformation (83) and tuberous sclerosis at 20 to 30 weeks. Because

other reports have confi rmed that all these conditions can be diagnosed by ultrasound

(12;114;140;172;174;178-182), it is unclear whether MRI is intrinsically better for the

diagnosis of conditions associated with ventriculomegaly, or whether the difference

lies in the detail and expertise with which the ultrasound is performed (181).

Craniosynostosis

Craniosynostosis can be diagnosed by ultrasound between 20 and 32 weeks’ gestation

(183;184). Three dimensional ultrasound has been used to aid the diagnosis at 26 weeks’,

and helps to visualize the frontal and metopic sutures (185-188). MRI has also been

found to be accurate in the diagnosis of craniosynostosis at 26 to 37 weeks (189).

| Fet

al n

euro

-im

agin

g: u

ltra

sou

nd

, MRI

, or

bo

th?

31

CONCLUSIONS

1. Neurosonography and MRI are often redundant, but occasionally complementary

tests. Both should be performed to a high standard. Although it is appropriate to

use a technique in which one is experienced to supplement another technique in

which one is less experienced, that cannot be taken as evidence that one technique

is inherently superior to the other.

2. Both neurosonography and MRI is operator dependant, and neither obviates the

need for a thorough knowledge of normal and abnormal neuroanatomy, and a

multidisciplinary approach to diagnosis.

3. There are some conditions in which either ultrasound or MRI is clearly the

diagnostic method of choice. In early pregnancy, and where repeated assessment

is needed, ultrasound has an obvious advantage. When ultrasound is diffi cult, as

in the obese patient or a patient with severe oligohydramnios, an MRI examination

may be more informative. MRI may also identify fetal ischemic lesions very quickly

after an insult, such as severe maternal trauma or death of a monochorionic co-

twin, and help to guide sampling for post-mortem histological diagnosis.

4. There is synergy between ultrasound and MRI for the diagnosis of certain conditions,

such as congenital cytomegalovirus infection or cerebellar telangiectasis.

| Ch

apte

r 2

32

REFERENCES

(1) Pasto ME, Kurtz AB. Ultrasonography of the normal fetal brain. Neuroradiology 1986; 28(5-6):380-385.

(2) Whittingham TA. Medical diagnostic applications and sources. Prog Biophys Mol Biol 2007; 93(1-3):84-110.

(3) Malinger G, Monteagudo A, Pilu G, Timor-Tritsch IE, Toi A. Sonographic examination of the fetal central nervous

system: guidelines for performing the ‘basic examination’ and the ‘fetal neurosonogram’. Ultrasound Obstet

Gynecol 2007; 29(1):109-116.

(4) Bitar R, Leung G, Perng R, Tadros S, Moody AR, Sarrazin J et al. MR pulse sequences: what every radiologist wants

to know but is afraid to ask. Radiographics 2006; 26(2):513-537.

(5) Brugger PC, Stuhr F, Lindner C, Prayer D. Methods of fetal MR: beyond T2-weighted imaging. Eur J Radiol 2006;

57(2):172-181.

(6) Donald I. Diagnostic sonar in obstetrics and gynecology. Obstet Gynecol Annu 1972; 1:245-271.

(7) Blaas HG, Eik-Nes SH, Kiserud T, Hellevik LR. Early development of the hindbrain: a longitudinal ultrasound study

from 7 to 12 weeks of gestation. Ultrasound Obstet Gynecol 1995; 5(3):151-160.

(8) Blaas HG, Eik-Nes SH, Berg S, Torp H. In-vivo three-dimensional ultrasound reconstructions of embryos and early

fetuses. Lancet 1998; 352(9135):1182-1186.

(9) Salihagic-Kadic A, Kurjak A, Medic M, Andonotopo W, Azumendi G. New data about embryonic and fetal

neurodevelopment and behavior obtained by 3D and 4D sonography. J Perinat Med 2005; 33(6):478-490.

(10) Timor-Tritsch IE, Peisner DB, Raju S. Sonoembryology: an organ-oriented approach using a high-frequency

vaginal probe. J Clin Ultrasound 1990; 18(4):286-298.

(11) Bault JP. Visualization of the fetal optic chiasma using three-dimensional ultrasound imaging. Ultrasound Obstet

Gynecol 2006; 28(6):862-864.

(12) Kalache KD, Eder K, Esser T, Proquitte H, Stoltenburg-Didinger G, Hartung JP et al. Three-dimensional

ultrasonographic reslicing of the fetal brain to assist prenatal diagnosis of central nervous system anomalies. J

Ultrasound Med 2006; 25(4):509-514.

(13) Monteagudo A, Timor-Tritsch IE, Mayberry P. Three-dimensional transvaginal neurosonography of the fetal

brain: ‘navigating’ in the volume scan. Ultrasound Obstet Gynecol 2000; 16(4):307-313.

(14) Pilu G, Hobbins JC. Sonography of fetal cerebrospinal anomalies. Prenat Diagn 2002; 22(4):321-330.

(15) Reece EB, Goldstein I. Three-level view of fetal brain imaging in the prenatal diagnosis of congenital anomalies.

J Matern Fetal Med 1999; 8(6):249-252.

(16) Timor-Tritsch IE, Monteagudo A. Transvaginal fetal neurosonography: standardization of the planes and sections

by anatomic landmarks. Ultrasound Obstet Gynecol 1996; 8(1):42-47.

(17) Monteagudo A, Timor-Tritsch IE, Moomjy M. Nomograms of the fetal lateral ventricles using transvaginal

sonography. J Ultrasound Med 1993; 12(5):265-269.

(18) Hertzberg BS, Kliewer MA, Freed KS, McNally PJ, DeLong DM, Bowie JD et al. Third ventricle: size and appearance

in normal fetuses through gestation. Radiology 1997; 203(3):641-644.

| Fet

al n

euro

-im

agin

g: u

ltra

sou

nd

, MRI

, or

bo

th?

33

(19) Goldstein I, Makhoul IR, Tamir A, Rajamim BS, Nisman D. Ultrasonographic nomograms of the fetal fourth

ventricle: additional tool for detecting abnormalities of the posterior fossa. J Ultrasound Med 2002; 21(8):849-

856.

(20) Achiron R, Achiron A. Development of the human fetal corpus callosum: a high-resolution, cross-sectional

sonographic study. Ultrasound Obstet Gynecol 2001; 18(4):343-347.

(21) Malinger G, Zakut H. The corpus callosum: normal fetal development as shown by transvaginal sonography. AJR

Am J Roentgenol 1993; 161(5):1041-1043.

(22) Serhatlioglu S, Kocakoc E, Kiris A, Sapmaz E, Boztosun Y, Bozgeyik Z. Sonographic measurement of the fetal

cerebellum, cisterna magna, and cavum septum pellucidum in normal fetuses in the second and third trimesters

of pregnancy. J Clin Ultrasound 2003; 31(4):194-200.

(23) Malinger G, Ginath S, Lerman-Sagie T, Watemberg N, Lev D, Glezerman M. The fetal cerebellar vermis: normal

development as shown by transvaginal ultrasound. Prenat Diagn 2001; 21(8):687-692.

(24) Zalel Y, Seidman DS, Brand N, Lipitz S, Achiron R. The development of the fetal vermis: an in-utero sonographic

evaluation. Ultrasound Obstet Gynecol 2002; 19(2):136-139.

(25) Achiron R, Kivilevitch Z, Lipitz S, Gamzu R, Almog B, Zalel Y. Development of the human fetal pons: in utero

ultrasonographic study. Ultrasound Obstet Gynecol 2004; 24(5):506-510.

(26) Malinger G, Lerman-Sagie T, Achiron R, Lipitz S. The subarachnoid space: normal fetal development as

demonstrated by transvaginal ultrasound. Prenat Diagn 2000; 20(11):890-893.

(27) Pooh RK, Pooh KH, Nakagawa Y, Maeda K, Fukui R, Aono T. Transvaginal Doppler assessment of fetal intracranial

venous fl ow. Obstet Gynecol 1999; 93(5 Pt 1):697-701.

(28) Bernard C, Droulle P, Didier F, Gerard H, Larroche JC, Plenat F et al. [Echographic aspects of cerebral sulci in the

ante- and perinatal period]. J Radiol 1988; 69(8-9):521-532.

(29) Cohen-Sacher B, Lerman-Sagie T, Lev D, Malinger G. Sonographic developmental milestones of the fetal cerebral

cortex: a longitudinal study. Ultrasound Obstet Gynecol 2006; 27(5):494-502.

(30) Monteagudo A, Timor-Tritsch IE. Development of fetal gyri, sulci and fi ssures: a transvaginal sonographic study.

Ultrasound Obstet Gynecol 1997; 9(4):222-228.

(31) Toi A, Lister WS, Fong KW. How early are fetal cerebral sulci visible at prenatal ultrasound and what is the normal

pattern of early fetal sulcal development? Ultrasound Obstet Gynecol 2004; 24(7):706-715.

(32) Kennedy KA, Flick KJ, Thurmond AS. First-trimester diagnosis of exencephaly. Am J Obstet Gynecol 1990;

162(2):461-463.

(33) Fong KW, Toi A, Salem S, Hornberger LK, Chitayat D, Keating SJ et al. Detection of fetal structural abnormalities

with US during early pregnancy. Radiographics 2004; 24(1):157-174.

(34) Sepulveda W, Dezerega V, Be C. First-trimester sonographic diagnosis of holoprosencephaly: value of the

“butterfl y” sign. J Ultrasound Med 2004; 23(6):761-765.

(35) Sepulveda W, Lutz I, Be C. Holoprosencephaly at 9 weeks 6 days in a triploid fetus: two- and 3-dimensional

sonographic fi ndings. J Ultrasound Med 2007; 26(3):411-414.

| Ch

apte

r 2

34

(36) Zalen-Sprock RM, van Vugt JM, van Geijn HP. First and early second trimester diagnosis of anomalies of the

central nervous system. J Ultrasound Med 1995; 14(8):603-610.

(37) Malinger G, Kidron D, Schreiber L, Ben Sira L, Hoffmann C, Lev D et al. Prenatal diagnosis of malformations of

cortical development by dedicated neurosonography. Ultrasound Obstet Gynecol 2007; 29(2):178-191.

(38) Hartung J, Heling KS, Rake A, Zimmer C, Chaoui R. Detection of an aneurysm of the vein of Galen following signs

of cardiac overload in a 22-week old fetus. Prenat Diagn 2003; 23(11):901-903.

(39) Suchet IB. Schizencephaly: antenatal and postnatal assessment with colour-fl ow Doppler imaging. Can Assoc

Radiol J 1994; 45(3):193-200.

(40) Baschat AA, Hecher K. Fetal growth restriction due to placental disease. Semin Perinatol 2004; 28(1):67-80.

(41) Manning FA. Fetal biophysical profi le: a critical appraisal. Clin Obstet Gynecol 2002; 45(4):975-985.

(42) Visser GH, Mulder EJ, Prechtl HF. Studies on developmental neurology in the human fetus. Dev Pharmacol Ther

1992; 18(3-4):175-183.

(43) de Vries JI, Fong BF. Changes in fetal motility as a result of congenital disorders: an overview. Ultrasound Obstet

Gynecol 2007; 29(5):590-599.

(44) Ang ES, Jr., Gluncic V, Duque A, Schafer ME, Rakic P. Prenatal exposure to ultrasound waves impacts neuronal

migration in mice. Proc Natl Acad Sci U S A 2006; 103(34):12903-12910.

(45) Abramowicz JS. Prenatal exposure to ultrasound waves: is there a risk? Ultrasound Obstet Gynecol 2007;

29(4):363-367.

(46) Clement S, Candy B, Heath V, To M, Nicolaides KH. Transvaginal ultrasound in pregnancy: its acceptability to

women and maternal psychological morbidity. Ultrasound Obstet Gynecol 2003; 22(5):508-514.

(47) Levine D, Cavazos C, Kazan-Tannus JF, McKenzie CA, Dialani V, Robson CD et al. Evaluation of real-time single-

shot fast spin-echo MRI for visualization of the fetal midline corpus callosum and secondary palate. AJR Am J

Roentgenol 2006; 187(6):1505-1511.

(48) Revel MP, Pons JC, Lelaidier C, Vial M, Frydman R, Musset D et al. Surface coil magnetic resonance imaging of the

fetal brain. Lancet 1992; 340(8812):176.

(49) Breysem L, Bosmans H, Dymarkowski S, Schoubroeck DV, Witters I, Deprest J et al. The value of fast MR imaging

as an adjunct to ultrasound in prenatal diagnosis. Eur Radiol 2003; 13(7):1538-1548.

(50) Gerards FA, Stoutenbeek P, Gooskens RH, Beek FJ, Groenendaal F. [Diagnostic value of prenatal MRI in fetus with

intracranial anomalies diagnosed by ultrasonography]. Ned Tijdschr Geneeskd 2001; 145(4):179-184.

(51) Girard N, Gire C, Sigaudy S, Porcu G, D’ercole C, Figarella-Branger D et al. MR imaging of acquired fetal brain

disorders. Childs Nerv Syst 2003; 19(7-8):490-500.

(52) Morioka T, Hashiguchi K, Nagata S, Miyagi Y, Mihara F, Hikino S et al. Fetal germinal matrix and intraventricular

hemorrhage. Pediatr Neurosurg 2006; 42(6):354-361.

(53) Prayer D, Kasprian G, Krampl E, Ulm B, Witzani L, Prayer L et al. MRI of normal fetal brain development. Eur J

Radiol 2006; 57(2):199-216.

| Fet

al n

euro

-im

agin

g: u

ltra

sou

nd

, MRI

, or

bo

th?

35

(54) Girard N, Raybaud C, Poncet M. In vivo MR study of brain maturation in normal fetuses. AJNR Am J Neuroradiol

1995; 16(2):407-413.

(55) Gelder-Hasker MR, Wezel-Meijler G, van Geijn HP, de Vries JI. Ultrasonography of the peri- and intraventricular

areas of the fetal brain between 26 and 36 weeks’ gestational age; a comparison with neonatal ultrasound.

Ultrasound Obstet Gynecol 2001; 17(1):34-41.

(56) Rados M, Judas M, Kostovic I. In vitro MRI of brain development. Eur J Radiol 2006; 57(2):187-198.

(57) Garel C, Chantrel E, Brisse H, Elmaleh M, Luton D, Oury JF et al. Fetal cerebral cortex: normal gestational

landmarks identifi ed using prenatal MR imaging. AJNR Am J Neuroradiol 2001; 22(1):184-189.

(58) Garel C, Chantrel E, Elmaleh M, Brisse H, Sebag G. Fetal MRI: normal gestational landmarks for cerebral biometry,

gyration and myelination. Childs Nerv Syst 2003; 19(7-8):422-425.

(59) van der Knaap MS, Wezel-Meijler G, Barth PG, Barkhof F, Ader HJ, Valk J. Normal gyration and sulcation in

preterm and term neonates: appearance on MR images. Radiology 1996; 200(2):389-396.

(60) Nicholl RM, Balasubramaniam VP, Urquhart DS, Sellathurai N, Rutherford MA. Postmortem brain MRI with

selective tissue biopsy as an adjunct to autopsy following neonatal encephalopathy. Eur J Paediatr Neurol 2007;

11(3):167-174.

(61) Whitby EH, Paley MN, Cohen M, Griffi ths PD. Postmortem MR imaging of the fetus: an adjunct or a replacement

for conventional autopsy? Semin Fetal Neonatal Med 2005; 10(5):475-483.

(62) Whitby EH, Paley MN, Cohen M, Griffi ths PD. Post-mortem fetal MRI: what do we learn from it? Eur J Radiol 2006;

57(2):250-255.

(63) Canapicchi R, Cioni G, Strigini FA, Abbruzzese A, Bartalena L, Lencioni G. Prenatal diagnosis of periventricular

hemorrhage by fetal brain magnetic resonance imaging. Childs Nerv Syst 1998; 14(12):689-692.

(64) Allkemper T, Tombach B, Schwindt W, Kugel H, Schilling M, Debus O et al. Acute and subacute intracerebral

hemorrhages: comparison of MR imaging at 1.5 and 3.0 T--initial experience. Radiology 2004; 232(3):874-881.

(65) Barnes PD. Neuroimaging and the timing of fetal and neonatal brain injury. J Perinatol 2001; 21(1):44-60.

(66) Mascalchi M, Filippi M, Floris R, Fonda C, Gasparotti R, Villari N. Diffusion-weighted MR of the brain: methodology

and clinical application. Radiol Med (Torino) 2005; 109(3):155-197.

(67) Baldoli C, Righini A, Parazzini C, Scotti G, Triulzi F. Demonstration of acute ischemic lesions in the fetal brain by

diffusion magnetic resonance imaging. Ann Neurol 2002; 52(2):243-246.

(68) Righini A, Bianchini E, Parazzini C, Gementi P, Ramenghi L, Baldoli C et al. Apparent diffusion coeffi cient

determination in normal fetal brain: a prenatal MR imaging study. AJNR Am J Neuroradiol 2003; 24(5):799-804.

(69) Bui T, Daire JL, Chalard F, Zaccaria I, Alberti C, Elmaleh M et al. Microstructural development of human brain

assessed in utero by diffusion tensor imaging. Pediatr Radiol 2006; 36(11):1133-1140.

(70) Levine D. Science to practice: can MR oxygenation imaging be used to assess at-risk pregnancies? Radiology

2006; 238(3):765-766.

(71) Jardri R, Pins D, Houffl in-Debarge V, Chaffi otte C, Rocourt N, Pruvo JP et al. Fetal cortical activation to sound at

33 weeks of gestation: A functional MRI study . Neuroimage. In press.

| Ch

apte

r 2

36

(72) Zaharchuk G, Busse RF, Rosenthal G, Manley GT, Glenn OA, Dillon WP. Noninvasive oxygen partial pressure

measurement of human body fl uids in vivo using magnetic resonance imaging. Acad Radiol 2006; 13(8):1016-

1024.

(73) Lowery CL, Eswaran H, Murphy P, Preissl H. Fetal magnetoencephalography. Semin Fetal Neonatal Med 2006;

11(6):430-436.

(74) Kanal E. Pregnancy and the safety of magnetic resonance imaging. Magn Reson Imaging Clin N Am 1994;

2(2):309-317.

(75) Myers C, Duncan KR, Gowland PA, Johnson IR, Baker PN. Failure to detect intrauterine growth restriction

following in utero exposure to MRI. Br J Radiol 1998; 71(845):549-551.

(76) van der Knaap MS, Valk J. Classifi cation of congenital abnormalities of the CNS. AJNR Am J Neuroradiol 1988;

9(2):315-326.

(77) Johnson SP, Sebire NJ, Snijders RJ, Tunkel S, Nicolaides KH. Ultrasound screening for anencephaly at 10-14 weeks

of gestation. Ultrasound Obstet Gynecol 1997; 9(1):14-16.

(78) Doherty D, Glass IA, Siebert JR, Strouse PJ, Parisi MA, Shaw DW et al. Prenatal diagnosis in pregnancies at risk for

Joubert syndrome by ultrasound and MRI. Prenat Diagn 2005; 25(6):442-447.

(79) Appasamy M, Roberts D, Pilling D, Buxton N. Antenatal ultrasound and magnetic resonance imaging in localizing

the level of lesion in spina bifi da and correlation with postnatal outcome. Ultrasound Obstet Gynecol 2006;

27(5):530-536.

(80) Mangels KJ, Tulipan N, Tsao LY, Alarcon J, Bruner JP. Fetal MRI in the evaluation of intrauterine myelomeningocele.

Pediatr Neurosurg 2000; 32(3):124-131.

(81) Aaronson OS, Hernanz-Schulman M, Bruner JP, Reed GW, Tulipan NB. Myelomeningocele: prenatal evaluation--

comparison between transabdominal US and MR imaging. Radiology 2003; 227(3):839-843.

(82) Twickler DM, Magee KP, Caire J, Zaretsky M, Fleckenstein JL, Ramus RM. Second-opinion magnetic resonance

imaging for suspected fetal central nervous system abnormalities. Am J Obstet Gynecol 2003; 188(2):492-496.

(83) Whitby EH, Paley MN, Sprigg A, Rutter S, Davies NP, Wilkinson ID et al. Comparison of ultrasound and magnetic

resonance imaging in 100 singleton pregnancies with suspected brain abnormalities. BJOG 2004; 111(8):784-

792.

(84) Poutamo J, Vanninen R, Partanen K, Ryynanen, Kirkinen P. Magnetic resonance imaging supplements

ultrasonographic imaging of the posterior fossa, pharynx and neck in malformed fetuses. Ultrasound Obstet

Gynecol 1999; 13(5):327-334.

(85) Sonigo PC, Rypens FF, Carteret M, Delezoide AL, Brunelle FO. MR imaging of fetal cerebral anomalies. Pediatr

Radiol 1998; 28(4):212-222.

(86) Ismail KM, Ashworth JR, Martin WL, Chapman S, McHugo J, Whittle MJ et al. Fetal magnetic resonance imaging

in prenatal diagnosis of central nervous system abnormalities: 3-year experience. J Matern Fetal Neonatal Med

2002; 12(3):185-190.

| Fet

al n

euro

-im

agin

g: u

ltra

sou

nd

, MRI

, or

bo

th?

37

(87) Frates MC, Kumar AJ, Benson CB, Ward VL, Tempany CM. Fetal anomalies: comparison of MR imaging and US for

diagnosis. Radiology 2004; 232(2):398-404.

(88) Glenn OA, Goldstein RB, Li KC, Young SJ, Norton ME, Busse RF et al. Fetal magnetic resonance imaging in the

evaluation of fetuses referred for sonographically suspected abnormalities of the corpus callosum. J Ultrasound

Med 2005; 24(6):791-804.

(89) Yuh WT, Nguyen HD, Fisher DJ, Tali ET, Gao F, Simonson TM et al. MR of fetal central nervous system abnormalities.

AJNR Am J Neuroradiol 1994; 15(3):459-464.

(90) Simon EM, Goldstein RB, Coakley FV, Filly RA, Broderick KC, Musci TJ et al. Fast MR imaging of fetal CNS anomalies

in utero. AJNR Am J Neuroradiol 2000; 21(9):1688-1698.

(91) Coakley FV, Hricak H, Filly RA, Barkovich AJ, Harrison MR. Complex fetal disorders: effect of MR imaging on

management--preliminary clinical experience. Radiology 1999; 213(3):691-696.

(92) Wald M, Lawrenz K, Deutinger J, Weninger M. Verifi cation of anomalies of the central nervous system detected

by prenatal ultrasound. Ultraschall Med 2004; 25(3):214-217.

(93) Paladini D, Volpe P. Posterior fossa and vermian morphometry in the characterization of fetal cerebellar

abnormalities: a prospective three-dimensional ultrasound study. Ultrasound Obstet Gynecol 2006; 27(5):482-

489.

(94) Pilu G, Visentin A, Valeri B. The Dandy-Walker complex and fetal sonography. Ultrasound Obstet Gynecol 2000;

16(2):115-117.

(95) Pilu G, Segata M, Ghi T, Carletti A, Perolo A, Santini D et al. Diagnosis of midline anomalies of the fetal brain with

the three-dimensional median view. Ultrasound Obstet Gynecol 2006; 27(5):522-529.

(96) Vinals F, Munoz M, Naveas R, Shalper J, Giuliano A. The fetal cerebellar vermis: anatomy and biometric assessment

using volume contrast imaging in the C-plane (VCI-C). Ultrasound Obstet Gynecol 2005; 26(6):622-627.

(97) Zalel Y, Gilboa Y, Gabis L, Ben Sira L, Hoffman C, Wiener Y et al. Rotation of the vermis as a cause of enlarged

cisterna magna on prenatal imaging. Ultrasound Obstet Gynecol 2006; 27(5):490-493.

(98) Tilea B, Delezoide AL, Khung-Savatovski S, Guimiot F, Vuillard E, Oury JF et al. Comparison between magnetic

resonance imaging and fetopathology in the evaluation of fetal posterior fossa non-cystic abnormalities.

Ultrasound Obstet Gynecol 2007.

(99) Triulzi F, Parazzini C, Righini A. MRI of fetal and neonatal cerebellar development. Semin Fetal Neonatal Med

2005; 10(5):411-420.

(100) Chong BW, Babcook CJ, Pang D, Ellis WG. A magnetic resonance template for normal cerebellar development in

the human fetus. Neurosurgery 1997; 41(4):924-928.

(101) McCann E, Pilling D, Hesseling M, Roberts D, Subhedar N, Sweeney E. Pontomedullary disconnection: fetal and

neonatal considerations. Pediatr Radiol 2005; 35(8):812-814.

(102) Guibaud L, des P, V. Plea for an anatomical approach to abnormalities of the posterior fossa in prenatal diagnosis.

Ultrasound Obstet Gynecol 2006; 27(5):477-481.

| Ch

apte

r 2

38

(103) Ramirez M, Wilkins I, Kramer L, Slopis J, Taylor SR. Prenatal diagnosis of unilateral megalencephaly by real-time

ultrasonography. Am J Obstet Gynecol 1994; 170(5 Pt 1):1384-1385.

(104) Yagishita A, Arai N, Tamagawa K, Oda M. Hemimegalencephaly: signal changes suggesting abnormal myelination

on MRI. Neuroradiology 1998; 40(11):734-738.

(105) Agid R, Lieberman S, Nadjari M, Gomori JM. Prenatal MR diffusion-weighted imaging in a fetus with

hemimegalencephaly. Pediatr Radiol 2006; 36(2):138-140.

(106) Barkovich AJ, Kuzniecky RI. Neuroimaging of focal malformations of cortical development. J Clin Neurophysiol

1996; 13(6):481-494.

(107) Achiron R, Yagel S, Rotstein Z, Inbar O, Mashiach S, Lipitz S. Cerebral lateral ventricular asymmetry: is this a

normal ultrasonographic fi nding in the fetal brain? Obstet Gynecol 1997; 89(2):233-237.

(108) Monteagudo A. Fetal neurosonography: should it be routine? Should it be detailed? Ultrasound Obstet Gynecol

1998; 12(1):1-5.

(109) de Lacoste MC, Horvath DS, Woodward DJ. Possible sex differences in the developing human fetal brain. J Clin

Exp Neuropsychol 1991; 13(6):831-846.

(110) Weinberger DR, Luchins DJ, Morihisa J, Wyatt RJ. Asymmetrical volumes of the right and left frontal and occipital

regions of the human brain. Ann Neurol 1982; 11(1):97-100.

(111) Hering-Hanit R, Achiron R, Lipitz S, Achiron A. Asymmetry of fetal cerebral hemispheres: in utero ultrasound

study. Arch Dis Child Fetal Neonatal Ed 2001; 85(3):F194-F196.

(112) Hepper PG, Shahidullah S, White R. Handedness in the human fetus. Neuropsychologia 1991; 29(11):1107-

1111.

(113) Grossman R, Hoffman C, Mardor Y, Biegon A. Quantitative MRI measurements of human fetal brain development

in utero. Neuroimage 2006; 33(2):463-470.

(114) Pinto V, Meo F, Loiudice L, D’Addario V. Prenatal sonographic imaging of an immature intracranial teratoma.

Fetal Diagn Ther 1999; 14(4):220-222.

(115) Lee DY, Kim YM, Yoo SJ, Cho BK, Chi JG, Kim IO et al. Congenital glioblastoma diagnosed by fetal sonography.

Childs Nerv Syst 1999; 15(4):197-201.

(116) Woodward PJ, Sohaey R, Kennedy A, Koeller KK. From the archives of the AFIP: a comprehensive review of fetal

tumors with pathologic correlation. Radiographics 2005; 25(1):215-242.

(117) Kelekci S, Yazicioglu HF, Yilmaz B, Aygun M, Omeroglu RE. Cardiac rhabdomyoma with tuberous sclerosis: a case

report. J Reprod Med 2005; 50(7):550-552.

(118) Levine D, Barnes P, Korf B, Edelman R. Tuberous sclerosis in the fetus: second-trimester diagnosis of subependymal

tubers with ultrafast MR imaging. AJR Am J Roentgenol 2000; 175(4):1067-1069.

(119) Levine D, Barnes PD, Robertson RR, Wong G, Mehta TS. Fast MR imaging of fetal central nervous system

abnormalities. Radiology 2003; 229(1):51-61.

| Fet

al n

euro

-im

agin

g: u

ltra

sou

nd

, MRI

, or

bo

th?

39

(120) Mitra AG, Dickerson C. Central nervous system tumor with associated unilateral ventriculomegaly: unusual

prenatal presentation of subsequently diagnosed tuberous sclerosis. J Ultrasound Med 2000; 19(9):651-654.

(121) Sonigo P, Elmaleh A, Fermont L, Delezoide AL, Mirlesse V, Brunelle F. Prenatal MRI diagnosis of fetal cerebral

tuberous sclerosis. Pediatr Radiol 1996; 26(1):1-4.

(122) Revel MP, Pons JC, Lelaidier C, Fournet P, Vial M, Musset D et al. Magnetic resonance imaging of the fetus: a study

of 20 cases performed without curarization. Prenat Diagn 1993; 13(9):775-799.

(123) Bahado-Singh RO, Wyse L, Dorr MA, Copel JA, O’Connor T, Hobbins JC. Fetuses with Down syndrome have

disproportionately shortened frontal lobe dimensions on ultrasonographic examination. Am J Obstet Gynecol

1992; 167(4 Pt 1):1009-1014.

(124) Persutte WH, Coury A, Hobbins JC. Correlation of fetal frontal lobe and transcerebellar diameter measurements:

the utility of a new prenatal sonographic technique. Ultrasound Obstet Gynecol 1997; 10(2):94-97.

(125) Pilu G, Falco P, Milano V, Perolo A, Bovicelli L. Prenatal diagnosis of microcephaly assisted by vaginal sonography

and power Doppler. Ultrasound Obstet Gynecol 1998; 11(5):357-360.

(126) Glenn OA, Norton ME, Goldstein RB, Barkovich AJ. Prenatal diagnosis of polymicrogyria by fetal magnetic

resonance imaging in monochorionic cotwin death. J Ultrasound Med 2005; 24(5):711-716.

(127) Kubik-Huch RA, Huisman TA, Wisser J, Gottstein-Aalame N, Debatin JF, Seifert B et al. Ultrafast MR imaging of

the fetus. AJR Am J Roentgenol 2000; 174(6):1599-1606.

(128) Parazzini C, Righini A, Lalatta F, Bianchini E, Triulzi F. Frontal bilateral megalencephaly: fetal and autopsy MR

evaluation of an unclassifi ed malformation. Prenat Diagn 2005; 25(6):489-491.

(129) Vimercati A, Greco P, Vera L, Resta M, Selvaggi L. The diagnostic role of “in utero” magnetic resonance imaging.

J Perinat Med 1999; 27(4):303-308.

(130) Bargallo N, Puerto B, De Juan C, Martinez-Crespo JM, Lourdes OM. Hereditary subependymal heterotopia

associated with mega cisterna magna: antenatal diagnosis with magnetic resonance imaging. Ultrasound

Obstet Gynecol 2002; 20(1):86-89.

(131) Fogliarini C, Chaumoitre K, Chapon F, Fernandez C, Levrier O, Figarella-Branger D et al. Assessment of cortical

maturation with prenatal MRI: part II: abnormalities of cortical maturation. Eur Radiol 2005; 15(9):1781-1789.

(132) Mittal P, Goncalves LF, Kusanovic JP, Espinoza J, Lee W, Nien JK et al. Objective evaluation of sylvian fi ssure

development by multiplanar 3-dimensional ultrasonography. J Ultrasound Med 2007; 26(3):347-353.

(133) Ruiz A, Sembely-Taveau C, Paillet C, Sirinelli D. [Sonographic cerebral sulcal pattern in normal fetuses]. J Radiol

2006; 87(1):49-55.

(134) Chung HW, Chen CY, Zimmerman RA, Lee KW, Lee CC, Chin SC. T2-Weighted fast MR imaging with true FISP

versus HASTE: comparative effi cacy in the evaluation of normal fetal brain maturation. AJR Am J Roentgenol

2000; 175(5):1375-1380.

(135) Levine D, Barnes PD. Cortical maturation in normal and abnormal fetuses as assessed with prenatal MR imaging.

Radiology 1999; 210(3):751-758.

| Ch

apte

r 2

40

(136) Chong BW, Babcook CJ, Salamat MS, Nemzek W, Kroeker D, Ellis WG. A magnetic resonance template for normal

neuronal migration in the fetus. Neurosurgery 1996; 39(1):110-116.

(137) Prayer D, Prayer L. Diffusion-weighted magnetic resonance imaging of cerebral white matter development. Eur

J Radiol 2003; 45(3):235-243.

(138) Kostovic I, Judas M, Rados M, Hrabac P. Laminar organization of the human fetal cerebrum revealed by

histochemical markers and magnetic resonance imaging. Cereb Cortex 2002; 12(5):536-544.

(139) Chi JG, Dooling EC, Gilles FH. Gyral development of the human brain. Ann Neurol 1977; 1(1):86-93.

(140) Delle Urban LA, Righini A, Rustico M, Triulzi F, Nicolini U. Prenatal ultrasound detection of bilateral focal

polymicrogyria. Prenat Diagn 2004; 24(10):808-811.

(141) Jan W, Lacey NA, Langford KS, Bewley SJ, Maxwell DJ. The antenatal diagnosis of migration disorders: a series of

four cases. Clin Radiol 2003; 58(3):247-250.

(142) Malinger G, Lerman-Sagie T, Watemberg N, Rotmensch S, Lev D, Glezerman M. A normal second-trimester

ultrasound does not exclude intracranial structural pathology. Ultrasound Obstet Gynecol 2002; 20(1):51-56.

(143) Fong KW, Ghai S, Toi A, Blaser S, Winsor EJ, Chitayat D. Prenatal ultrasound fi ndings of lissencephaly associated

with Miller-Dieker syndrome and comparison with pre- and postnatal magnetic resonance imaging. Ultrasound

Obstet Gynecol 2004; 24(7):716-723.

(144) Prayer D, Brugger PC, Kasprian G, Witzani L, Helmer H, Dietrich W et al. MRI of fetal acquired brain lesions. Eur J

Radiol 2006; 57(2):233-249.

(145) Oh KY, Kennedy AM, Frias AE, Jr., Byrne JL. Fetal schizencephaly: pre- and postnatal imaging with a review of the

clinical manifestations. Radiographics 2005; 25(3):647-657.

(146) D’ercole C, Girard N, Cravello L, Boubli L, Potier A, Raybaud C et al. Prenatal diagnosis of fetal corpus callosum

agenesis by ultrasonography and magnetic resonance imaging. Prenat Diagn 1998; 18(3):247-253.

(147) Patel TR, Bannister CM, Thorne J. A study of prenatal ultrasound and postnatal magnetic imaging in the diagnosis

of central nervous system abnormalities. Eur J Pediatr Surg 2003; 13 Suppl 1:S18-S22.

(148) Volpe P, Paladini D, Resta M, Stanziano A, Salvatore M, Quarantelli M et al. Characteristics, associations and

outcome of partial agenesis of the corpus callosum in the fetus. Ultrasound Obstet Gynecol 2006; 27(5):509-

516.

(149) Malinger G, Ben Sira L, Lev D, Ben Aroya Z, Kidron D, Lerman-Sagie T. Fetal brain imaging: a comparison between

magnetic resonance imaging and dedicated neurosonography. Ultrasound Obstet Gynecol 2004; 23(4):333-

340.

(150) Celentano C, Prefumo F, Liberati M, Tartaro A, Gallo G, Lattanzio G et al. Prenatal diagnosis of septal agenesis

with normal pituitary function. Prenat Diagn 2006; 26(11):1075-1077.

(151) De Biasio P, Prefumo F, Casagrande V, Stroppiano M, Venturini PL, Filocamo M. First-trimester fetal nuchal

translucency and inherited metabolic disorders. Prenat Diagn 2006; 26(1):77-80.

(152) Drobyshevsky A, Derrick M, Prasad PV, Ji X, Englof I, Tan S. Fetal brain magnetic resonance imaging response

acutely to hypoxia-ischemia predicts postnatal outcome. Ann Neurol 2007; 61(4):307-314.

| Fet

al n

euro

-im

agin

g: u

ltra

sou

nd

, MRI

, or

bo

th?

41

(153) Ozduman K, Pober BR, Barnes P, Copel JA, Ogle EA, Duncan CC et al. Fetal stroke. Pediatr Neurol 2004; 30(3):151-

162.

(154) Otsubo Y, Yoneyama Y, Sawa R, Suzuki S, Araki T. Fetal brain death and Dandy-Walker malformation. Prenat

Diagn 1999; 19(8):777-779.

(155) Wagenvoort AM, Bekker MN, Go AT, Vandenbussche FP, van Buchem MA, Valk J et al. Ultrafast scan magnetic

resonance in prenatal diagnosis. Fetal Diagn Ther 2000; 15(6):364-372.

(156) Moinuddin A, McKinstry RC, Martin KA, Neil JJ. Intracranial hemorrhage progressing to porencephaly as a result

of congenitally acquired cytomegalovirus infection--an illustrative report. Prenat Diagn 2003; 23(10):797-800.

(157) Malinger G, Lev D, Zahalka N, Ben Aroia Z, Watemberg N, Kidron D et al. Fetal cytomegalovirus infection of the

brain: the spectrum of sonographic fi ndings. AJNR Am J Neuroradiol 2003; 24(1):28-32.

(158) Tassin GB, Maklad NF, Stewart RR, Bell ME. Cytomegalic inclusion disease: intrauterine sonographic diagnosis

using fi ndings involving the brain. AJNR Am J Neuroradiol 1991; 12(1):117-122.

(159) Guibaud L, Attia-Sobol J, Buenerd A, Foray P, Jacquet C, Champion F et al. Focal sonographic periventricular pattern

associated with mild ventriculomegaly in foetal cytomegalic infection revealing cytomegalic encephalitis in the

third trimester of pregnancy. Prenat Diagn 2004; 24(9):727-732.

(160) de Vries LS, Gunardi H, Barth PG, Bok LA, Verboon-Maciolek MA, Groenendaal F. The spectrum of cranial ultrasound

and magnetic resonance imaging abnormalities in congenital cytomegalovirus infection. Neuropediatrics 2004;

35(2):113-119.

(161) Duin LK, Willekes C, Baldewijns MM, Robben SG, Offermans J, Vles J. Major brain lesions by intrauterine herpes

simplex virus infection: MRI contribution. Prenat Diagn 2007; 27(1):81-84.

(162) Mancini J, Lethel V, Hugonenq C, Chabrol B. Brain injuries in early foetal life: consequences for brain development.

Dev Med Child Neurol 2001; 43(1):52-55.

(163) Ellestad SC, Shelton S, James AH. Prenatal diagnosis of a trauma-related fetal epidural hematoma. Obstet

Gynecol 2004; 104(6):1298-1300.

(164) Ramenghi LA, Fumagalli M, Righini A, Triulzi F, Kustermann A, Mosca F. Thrombophilia and fetal germinal matrix-

intraventricular hemorrhage: does it matter? Ultrasound Obstet Gynecol 2005; 26(5):574-576.

(165) Sarnat HB. Watershed infarcts in the fetal and neonatal brainstem. An aetiology of central hypoventilation,

dysphagia, Moibius syndrome and micrognathia. Eur J Paediatr Neurol 2004; 8(2):71-87.

(166) Gorincour G, Rypens F, Lapierre C, Costa T, Audibert F, Robitaille Y. Fetal magnetic resonance imaging in the

prenatal diagnosis of cerebellar hemorrhage. Ultrasound Obstet Gynecol 2006; 27(1):78-80.

(167) Guibaud L, Garel C, Annie B, Pascal G, Francois V, Vavasseur C et al. Prenatal diagnosis of capillary telangiectasia

of the cerebellum--ultrasound and MRI features. Prenat Diagn 2003; 23(10):791-796.

(168) D’Addario V, Pinto V, Di Cagno L, Pintucci A. Sonographic diagnosis of fetal cerebral ventriculomegaly: an update.

J Matern Fetal Neonatal Med 2007; 20(1):7-14.

(169) Cardoza JD, Goldstein RB, Filly RA. Exclusion of fetal ventriculomegaly with a single measurement: the width of

the lateral ventricular atrium. Radiology 1988; 169(3):711-714.

| Ch

apte

r 2

42

(170) Garel C, Alberti C. Coronal measurement of the fetal lateral ventricles: comparison between ultrasonography

and magnetic resonance imaging. Ultrasound Obstet Gynecol 2006; 27(1):23-27.

(171) Twickler DM, Reichel T, McIntire DD, Magee KP, Ramus RM. Fetal central nervous system ventricle and cisterna

magna measurements by magnetic resonance imaging. Am J Obstet Gynecol 2002; 187(4):927-931.

(172) Greco P, Resta M, Vimercati A, Dicuonzo F, Loverro G, Vicino M et al. Antenatal diagnosis of isolated lissencephaly

by ultrasound and magnetic resonance imaging. Ultrasound Obstet Gynecol 1998; 12(4):276-279.

(173) Valsky DV, Ben Sira L, Porat S, Yanai N, Lewin A, Nadjari M et al. The role of magnetic resonance imaging in the

evaluation of isolated mild ventriculomegaly. J Ultrasound Med 2004; 23(4):519-523.

(174) Hashimoto I, Tada K, Nakatsuka M, Nakata T, Inoue N, Takata M et al. Fetal hydrocephalus secondary to

intraventricular hemorrhage diagnosed by ultrasonography and in utero fast magnetic resonance imaging. A

case report. Fetal Diagn Ther 1999; 14(4):248-253.

(175) de Laveaucoupet J, Audibert F, Guis F, Rambaud C, Suarez B, Boithias-Guerot C et al. Fetal magnetic resonance

imaging (MRI) of ischemic brain injury. Prenat Diagn 2001; 21(9):729-736.

(176) Kojima K, Suzuki Y, Seki K, Yamamoto T, Sato T, Tanaka T et al. Prenatal diagnosis of lissencephaly (type II) by

ultrasound and fast magnetic resonance imaging. Fetal Diagn Ther 2002; 17(1):34-36.

(177) Righini A, Zirpoli S, Mrakic F, Parazzini C, Pogliani L, Triulzi F. Early prenatal MR imaging diagnosis of polymicrogyria.

AJNR Am J Neuroradiol 2004; 25(2):343-346.

(178) Aubry MC, Aubry JP, Dommergues M. Sonographic prenatal diagnosis of central nervous system abnormalities.

Childs Nerv Syst 2003; 19(7-8):391-402.

(179) Langer B, Boudier E, Gasser B, Christmann D, Messer J, Schlaeder G. Antenatal diagnosis of brain damage in the

survivor after the second trimester death of a monochorionic monoamniotic co-twin: case report and literature

review. Fetal Diagn Ther 1997; 12(5):286-291.

(180) Malinger G, Lev D, Lerman-Sagie T. Is fetal magnetic resonance imaging superior to neurosonography for

detection of brain anomalies? Ultrasound Obstet Gynecol 2002; 20(4):317-321.

(181) Malinger G, Lev D, Lerman-Sagie T. Fetal central nervous system: MR imaging versus dedicated US--need for

prospective, blind, comparative studies. Radiology 2004; 232(1):306-307.

(182) Timor-Tritsch IE, Monteagudo A. Magnetic resonance imaging versus ultrasound for fetal central nervous

system abnormalities. Am J Obstet Gynecol 2003; 189(4):1210-1211.

(183) Ghi T, Perolo A, Banzi C, Contratti G, Valeri B, Savelli L et al. Two-dimensional ultrasound is accurate in the

diagnosis of fetal craniofacial malformation. Ultrasound Obstet Gynecol 2002; 19(6):543-551.

(184) Delahaye S, Bernard JP, Renier D, Ville Y. Prenatal ultrasound diagnosis of fetal craniosynostosis. Ultrasound

Obstet Gynecol 2003; 21(4):347-353.

(185) Faro C, Benoit B, Wegrzyn P, Chaoui R, Nicolaides KH. Three-dimensional sonographic description of the fetal

frontal bones and metopic suture. Ultrasound Obstet Gynecol 2005; 26(6):618-621.

(186) Faro C, Chaoui R, Wegrzyn P, Levaillant JM, Benoit B, Nicolaides KH. Metopic suture in fetuses with Apert

syndrome at 22-27 weeks of gestation. Ultrasound Obstet Gynecol 2006; 27(1):28-33.

| Fet

al n

euro

-im

agin

g: u

ltra

sou

nd

, MRI

, or

bo

th?

43

(187) Benacerraf BR, Spiro R, Mitchell AG. Using three-dimensional ultrasound to detect craniosynostosis in a fetus

with Pfeiffer syndrome. Ultrasound Obstet Gynecol 2000; 16(4):391-394.

(188) Krakow D, Santulli T, Platt LD. Use of three-dimensional ultrasonography in differentiating craniosynostosis

from severe fetal molding. J Ultrasound Med 2001; 20(4):427-431.

(189) Fjortoft MI, Sevely A, Boetto S, Kessler S, Sarramon MF, Rolland M. Prenatal diagnosis of craniosynostosis: value

of MR imaging. Neuroradiology 2007; 49(6):515-521.

PHYSIOLOGY

Chapter 3FIRST TRIMESTER

NEUROSONOEMBRYOLOGY WITH

AUTOMATED FOLLICLE TRACKING:

PRELIMINARY FINDINGS

L.R. Pistorius, Ph. Stoutenbeek, G.H.A. Visser

L.R. Pistorius, Ph. Stoutenbeek, G.H.A. VisserDepartment of Obstetrics, Division Perinatology and Gynaecology, University Medical Centre Utrecht,

The Netherlands

J Matern Fetal Neonatal Med 2008, in press

| Ch

apte

r 3

48

ABSTRACT

We evaluated the feasibility of demonstration of the embryonic cerebral ventricles

using three dimensional transvaginal ultrasonography and software which has been

developed to automatically calculate the volume of ovarian follicles after ovarian

stimulation. In four out of six subsequent ultrasound examinations which were done

for determination of gestational age from eight to ten gestational weeks, suffi cient

resolution was obtained to demonstrate the embryonic cerebral ventricles with

automated volume calculation software, yielding striking images of a dramatic stage

of embryonic development.

| Firs

t trim

este

r neu

roso

noem

bryo

logy

with

aut

omat

ed fo

llicl

e tra

ckin

g: P

relim

inar

y fi n

ding

s

49

INTRODUCTION

The study of the development of the human embryo has become possible in vivo more

than two decades ago with the introduction of real-time ultrasonography, allowing

the study of fetal movements and development of the motor component of the fetal

central nervous system (1) as well as the structural development of the embryonic

central nervous system (2). The fi rst three-dimensional reconstructions were reported

a decade later using an experimental 7,5MHz annular array transvaginal transducer

(3). To improve on the resolution obtainable with commercially available transvaginal

transducers, intrauterine sonography with a 20MHz transducer has been performed

to describe the normal development of cerebral ventricles prior to fi rst trimester

pregnancy termination (4).

Commercially, transvaginal transducers with a frequency of 6 – 12MHz and harmonic

frequencies up to 18MHz are now available. Imaging software include automated

volume calculation (SonoAVC, General Electrical Medical Systems, Zipf, Austria) which

was developed to eliminate observer bias and reduce the time needed for volume

measurements of ovarian follicles. As suggested recently, this technology might be

useful for obstetrical applications such as the demarcation of fetal cerebral ventricles

(5).

METHODS

We decided to evaluate the applicability of the combination of three dimensional

ultrasound with a high-frequency transvaginal transducer and automated follicle

tracking in normal clinical practice by evaluating six subsequent patients presenting

for fi rst trimester ultrasound for determination of gestational age.

The ultrasound examination was performed with a General Electric Voluson E8 fi tted

with the RIC6-12-D 6-12 MHz transvaginal probe with high harmonic frequency

and with the optional CRI (crossbeam imaging) and SRI (speckle reduction imaging)

activated. The three-dimensional data were aquired at “high-2”or “max” quality, and

subsequently analysed with 4DView version 7.0 (General Electrical Medical Systems,

Zipf, Austria). The orientation of the images were standardized as described by Merz

et al (6) to facilitate the identifi cation of the right and left side. Measurements were

taken of the ventricles of the telencephalon, diencephalon, mesencephalon and

| Ch

apte

r 3

50

rhombencephalon on sectional planes, and the volumes of these ventricles were

calculated with automated volume calculation by right-clicking the separate ventricles,

and by merging parts of the lateral ventricles anterior and posterior to choroid plexus

volumes into a single volume. The “growth” algorithm was set to “max”, and the

“separation” algorithm set to “mid” or “max”, depending on the best visual result.

Figure 1 CRL=19mm

8w3dCRL=22mm

8w6dCRL=28mm

9w2d

Lateral

Oblique

Frontal

Telencephalon

Diencephalon

Mesencephalon

Rhombencephalon

1 cm

| Firs

t trim

este

r neu

roso

noem

bryo

logy

with

aut

omat

ed fo

llicl

e tra

ckin

g: P

relim

inar

y fi n

ding

s

51

RESULTS

In all six pregnancies, it was easy to reconstruct the appropriate plane to measure

the crown rump length. In four out of six pregnancies, the cerebral ventricles could

be visualized with suffi cient resolution for two dimensional measurements and

evaluation with automated volume calculation. In one pregnancy, there was an

acutely retroverted uterus with too large a distance between the ultrasound probe

and the embryo, and in another pregnancy there was a set of triplets, where suffi cient

resolution for the SonoAVC software could not be obtained. The mechanical index (MI)

ranged from 0.5 to 1.2, and the thermal index (TI) from <0.1 to 0.2.

The images of three embryos are displayed to scale in Figure 1, showing the rapid

development of cerebral ventricles in less than a week of gestational age, and the

identifi cation of the lateral ventricles developing from the telencephalon, third

ventricle developing from the diencephalon, future aqueduct developing from the

mesencephalon and the fourth ventricle developing from the rhombencephalon.

DISCUSSION

The visual results compare favourably with those obtained in a research setting with

an experimental 7.5MHz transvaginal probe (3) and with intra-uterine sonography

with a 20MHz ultrasound probe (4)

Unfortunately, it is not possible to choose the particular colour which is used to display

a particular ventricle, which would make longitudinal assessment visually more

intuitive.

The software has also been developed for automated follicle measurement after

ovarian hyperstimulation, and has not been validated for volume calculation of

the irregular shape and smaller size of the embryonic brain ventricles. However,

the algorithm used for automated calculation is based on the number of coloured

voxels and does not assume a spheric shape which should yield an accurate volume

measurement in addition to a visual display which allows subjective assessment of

normality. The volumes of the different ventricles were often less than the minimum

volume of 0.01ml which is displayed by the automated software, and this would need

to be improved in order to use this technology to generate a normal data set this early

in pregnancy.

| Ch

apte

r 3

52

The rapid development of the embryonic cerebral ventricles is a graphic reminder of

the intricacy of brain development. Although ultrasound is deemed safe, also in the

fi rst trimester, the recent fi nding that exposure to ultrasound for more than 30 minutes

can impair neuronal migration in mice fetuses, was another prompt to reconsider

unbridled use of ultrasound. In this regard, contrary to previous comments that three

dimensional ultrasound adds to scan time when evaluating embryonic development

(7), one can actually decrease the scan time by obtaining a three dimensional volume.

From this volume, the required images (such as that needed for measuring the crown

rump length or evaluating the ventricular system) can then be reconstructed. The

(short) extra time spent in analysis is more than compensated for by the decrease in

actual scan time.

In conclusion, in this small pilot study it was easy to obtain very satisfactory images of

the embryonic ventricular system at eight weeks of gestational age with commercially

available equipment. In addition, a single three-dimensional sweep could also minimize

the exposure of the rapidly developing embryo to ultrasound energy.

POSTSCRIPT

Subsequent to this report, a report on the use of inverse mode rendering of the fetal

cerebral ventricles has appeared with similar results to those obtained with automatic

volume calculation (8).

| Firs

t trim

este

r neu

roso

noem

bryo

logy

with

aut

omat

ed fo

llicl

e tra

ckin

g: P

relim

inar

y fi n

ding

s

53

REFERENCES

(1) de Vries JI, Visser GH, Prechtl HF. The emergence of fetal behaviour. I. Qualitative aspects. Early Hum Dev 1982;

7(4):301-322.

(2) Timor-Tritsch IE, Farine D, Rosen MG. A close look at early embryonic development with the high-frequency

transvaginal transducer. Am J Obstet Gynecol 1988; 159(3):676-681.

(3) Blaas HG, Eik-Nes SH, Berg S, Torp H. In-vivo three-dimensional ultrasound reconstructions of embryos and

early fetuses. Lancet 1998; 352(9135):1182-1186.

(4) Tanaka H, Senoh D, Yanagihara T, Hata T. Intrauterine sonographic measurement of embryonic brain vesicle.

Hum Reprod 2000; 15(6):1407-1412.

(5) Raine-Fenning N, Jayaprakasan K., Clewes J. Automated follicle tracking facilitates standardization and may

improve work fl ow. Ultrasound Obstet Gynecol 2008; 30:1015-1018.

(6) Merz E, Benoit B, Blaas HG, Baba K, Kratochwil A, Nelson T et al. Standardization of three-dimensional images

in obstetrics and gynecology: consensus statement. Ultrasound Obstet Gynecol 2007; 29(6):697-703.

(7) Centini G, Rosignoli L., Faldini E, Tonni G, Azzoni D, LaSala GB. Correlation between transvaginal 2D and 3D fi rst

trimester neuroimaging and embryonic development according to Carnagie stadiation. Ultrasound Obstet

Gynecol 2004;371.

(8) Kim MS, Jeanty P, Turner C, Benoit B. Three-dimensional sonographic evaluations of embryonic brain

development. J Ultrasound Med 2008; 27(1):119-124.

Chapter 4GRADE AND SYMMETRY OF NORMAL FETAL

CORTICAL DEVELOPMENT: A LONGITUDINAL TWO

AND THREE DIMENSIONAL ULTRASOUND STUDY

L.R. Pistorius1, Ph. Stoutenbeek1, F. Groenendaal2, L.S. de Vries2, G.T.R. Manten1, E.J.H. Mulder1, G.H.A. Visser1

1Department Obstetrics, Division Perinatology and Gynaecology, University Medical Centre Utrecht, Utrecht, The Netherlands

2Department Neonatology, Division Perinatology and Gynaecology, University Medical Centre Utrecht, Utrecht, The Netherlands

Ultrasound Obstet Gynecol (provisionally accepted)

| Ch

apte

r 4

56

ABSTRACT

Introduction

Recent studies have shown the capability of ultrasound to demonstrate when fetal

cortical sulci appear, but information is lacking on the grade of development and left-

right differences at different gestational ages.

Objectives

A longitudinal study was undertaken to evaluate the reproducibility of a simple scoring

method to grade fetal cortical development and to study physiological asymmetry.

Methods

A cohort of 28 patients was examined by 2D and 3D ultrasound from 20 until 40

weeks. Cortical development was graded from 0 (no development) to 5 (maximum

development). One examination per week of gestation was randomly selected for

evaluation of intra- and interobserver variation.

Results

215 ultrasound examinations were performed in 28 patients. The intra-observer

variation showed a correlation (Spearman’s rho) of 0,933 (2D) and 0,927 (3D) and a

Kappa of 0.6 (2D) and 0.63 (3D). The inter-observer variation showed a correlation

(Spearman’s rho) of 0,904 (2D) and 0,921 (3D) and a Kappa of 0.56 (2D) and 0.57 (3D).

Asymmetry was seen in the parieto-occipital, calcarine and cingulate sulci and

occipital area, especially between 24 and 28 weeks, and especially in female fetuses.

The right parieto-occipital sulcus tended to be more advanced and the left calcarine

and cingulate sulci tended to be more advanced.

Conclusions

Evaluation of cortical development by means of a simple grading score is feasible and

reproducible. Asymmetrical development of sulci, previously only described in vitro,

has now also been demonstrated in vivo. 3D ultrasound was equally accurate and

repeatable as 2D ultrasound.

| Gra

de

and

sym

met

ry o

f n

orm

al f

etal

cor

tica

l dev

elop

men

t: a

lon

gitu

din

al t

wo

and

th

ree

dim

ensi

onal

ult

raso

un

d s

tud

y

57

INTRODUCTION

The gestational age at which different sulci and gyri develop has been well described

anatomically (1-3) as well as with antenatal MRI examination (4-8) and antenatal

transabdominal (9;10) and transvaginal ultrasound (10-12) Indeed, neurosonography

has been shown to be as useful as MRI in diagnosing malformations of cortical

development (13). Although the appearance of cortical development has been

described, quantifi cation of the development of sulci and gyri is still lacking. Also,

asymmetrical cortical development has been demonstrated anatomically (1;3), but this

has not been documented nor systematically studied antenatally (2;4-7;9;10;12;14;15)

As malformation of cortical development might be localized, and might present with

advanced or delayed gyral formation(16;17), it is important to determine the degree of

asymmetry which might represent normal individual physiological variation.

As ultrasound lends itself eminently to the visualization of structural cortical

development, a study was undertaken to describe the progression and symmetry of

fetal cortical development. It was also evaluated whether 2D or 3D ultrasound would

be more suitable for assessing fetal cortical development.

METHODS

Twenty eight healthy pregnant women were included after a screening ultrasound

examination at 20 weeks had demonstrated a singleton pregnancy with no abnormal

fi ndings. Other inclusion criteria were confi rmation of pregnancy duration by

ultrasound during the fi rst trimester, and absence of risk factors which might infl uence

fetal growth or development, such as maternal disease or previous intra-uterine

growth restriction. The study was approved by the local Medical Ethics Committee of

the University Medical Centre Utrecht.

Ultrasound examinations were carried out at two weekly intervals with a General

Electric Voluson 730 Expert (General Electric Healthcare, London). Biometry (including

fetal head circumference, abdominal circumference and femur length) was performed,

followed by neurosonography if the biometry conformed to the gestational age.

Transabdominal neurosonography was performed with the GE M7C-H H40412LS 2.6-

7.7MHz 2D ultrasound probe and the GE RAB4-8L H48621Z 4.0 -8.5MHz 3D abdominal

probe. Prior to the fi rst examination, it was randomly assigned (by means of an internet

| Ch

apte

r 4

58

based random number generator www.random.org/integers) to start with either

two or three dimensional ultrasound examination. Subsequently, the examination

was started alternatively with two or three dimensional ultrasound. All ultrasound

examinations were done by one observer (LP). On 2D ultrasound, it was attempted

to obtain the standard axial planes (18), the coronal and sagittal neurosonographic

planes (19) and a plane cephalad to the transventricular plane to demonstrate the

parieto-occipital fi ssure (10). Three images of each plane were stored digitally for

subsequent evaluation. The position of the fetus (right side or left side anterior) was

noted. On 3D ultrasound, three static volumes were obtained with high to maximum

quality (depending on the amount of fetal movement): the fi rst starting with the

axial transventricular plane (20), the second starting with the axial transcerebellar

plane, and the third starting with the midcoronal plane. If a midsagittal plane could

be obtained, a fourth volume was obtained starting with this plane. The angle was

chosen to include the maximum amount of fetal brain that could be visualized, but

kept as narrow as possible to shorten scanning time. If fetal movement was observed

during the volume acquisition, the volume was acquired again. The times used for two

dimensional and three dimensional ultrasound examinations were recorded.

A subjective score ranging from 0 (no development) to 5 (maximum development)

(Figure 1) was modifi ed from scores used for MRI (8) and ultrasound analysis (21) of

cortical development and used to assess the development of the Sylvian, parieto-

occipital, calcarine, superior temporal, central and cingulate sulci, and the frontal,

parietal, temporal, occipital and mesial cortical areas. The development of the parieto-

occipital and central sulci and frontal and parietal areas were evaluated in the axial

plane cephalad to the transventricular plane, the Sylvian and superior temporal sulci,

frontal and temporal areas in the transthalamic axial plane, the cingulate sulcus and

mesial area in the midcoronal planes, the calcarine sulcus and occipital areas in the

coronal transcerebellar plane, and the central sulcus, frontal and parietal areas in

the parasagittal oblique plane (Figure 2) on the recorded 2D images and 3D volumes

separately. Evaluation of stored images and volumes was done using 4DView version

7.0 (General Electric Healthcare, London). 3D images were reoriented to facilitate

identifi cation of the right and left hand sides (Figure 3) (22).

| Gra

de

and

sym

met

ry o

f n

orm

al f

etal

cor

tica

l dev

elop

men

t: a

lon

gitu

din

al t

wo

and

th

ree

dim

ensi

onal

ult

raso

un

d s

tud

y

59

1.1 Sulcus: Parieto-occipital, calcarine, cingular, superior temporal and centralGrade Definition & diagram Example

0 None visible

1 Earliest changes (shallow indentation or echogenic dot)

2 Broad V (width > depth)

3 Y or narrow V (depth = or > width

4 I-or J-haped

5 Branched

< 60o

Figure 1. Cortical grading (continued)

| Ch

apte

r 4

60

1.2 Sylvian fissure

Grade Definition & diagram Example

0 Not visible

1 Shallow indentation

2 Obtuse angular shape

3 Acute angles, < 50% operculisation

4 ≥50% operculisation

5 Complete operculisation

x

y

| Gra

de

and

sym

met

ry o

f n

orm

al f

etal

cor

tica

l dev

elop

men

t: a

lon

gitu

din

al t

wo

and

th

ree

dim

ensi

onal

ult

raso

un

d s

tud

y

61

1.3 Gyration: Frontal, parietal, temporal, occipital and mesial

Grade Definition & diagram Example0 None visible

1 Shallow undulation

2 Gyral width > depth

3 Gyral width = depth

4 Gyral depth > width

5 Branched gyri / sulci

| Ch

apte

r 4

62

Axial plane cephalad to transventricular plane Parieto-occipital fissure Central sulcus Frontal area Parietal area

Axial transthalamic plane Sylvian fissure Superior temporal fissure Frontal area Temporal area

Coronal transcaudate and transthalamic plane Cingulate sulcus Mesial area

Coronal transcerebellar plane Calcarine sulcus Occipital area

Parasagittal oblique plane Central sulcus Frontal area Parietal area

Frontal Parietal

Central

Parieto-occipital

Frontal Temporal

SylvianSuperiortemporal

CingulateMesial

CalcarineOccipital

Frontal

Parietal

Central

Figure 2. Axial, coronal and sagittal planes

| Gra

de

and

sym

met

ry o

f n

orm

al f

etal

cor

tica

l dev

elop

men

t: a

lon

gitu

din

al t

wo

and

th

ree

dim

ensi

onal

ult

raso

un

d s

tud

y

63

Prim

ary

imag

e (t

ran

scer

ebel

lar)

Reor

ien

ted

imag

e

Prim

ary

imag

e (t

ran

sth

alam

ic)

Reor

ien

ted

imag

e

Fig

ure

3. O

rien

tati

on o

f 3D

imag

e

| Ch

apte

r 4

64

An attempt was made to grade both left and right hand side at each examination.

The quality of the images was graded subjectively (0=not performed or impossible

to evaluate, 1=barely acceptable, 2=suboptimal, 3=average, 4=better than average,

5=excellent), and the time spent on the analyses was also recorded. Where both (left

and right) sulci and cortical areas could be seen in at least 70% of examinations per

fetus, the difference in grade between the two sides was noted. Where one or the other

side predominated, it was noted as asymmetry. If both left and right sulci and cortical

areas could not be seen in at least 70% of examinations per fetus, asymmetry would

be defi ned as a signifi cant difference (by the Mann-Whitney u test) in gestational age

at which each grade of cortical development was fi rst seen provided that the specifi c

grade was seen in at least 70% of fetuses on each side.

One examination per week was randomly selected for analysis of intra- and

interobserver variation. The analysis was done by a second observer (GM) and redone

by the fi rst observer, without being aware of the fi rst grades.

Data management and statistical analysis (including multilevel analysis mixed model

option to identify variables with an independent effect on the time course of cortical

development) were performed with SPSS version 16.0 for Windows (SPSS Inc. Chicago,

Illinois). Mixor version 2 (Donald Hedeker, Chicago, Illinois) was used for analysis of

categorical variables. The proportion of “missing values” or structures not visualized

well enough to be graded reliably on 2D and 3D ultrasound examination respectively

was compared with Fisher’s exact test. The Kappa statistic and Spearman’s rho were

calculated for the intra- and interobserver variation. The study would be also able

to detect an 8% difference in the proportion of structures visualized reliably with

2D or 3D ultrasound examination with a power of 80% and α 0.05 (two tailed). The

distribution of different grades of cortical development seen at different gestational

ages and gestational age at the fi rst appearance of each grade were calculated. The

time spent on 2D and 3D ultrasound examination and analysis was compared with a

paired t test. The study would be able to detect a 5 minute difference in duration of

2D or 3D ultrasound examination with a power of 90% and α 0.05 (two tailed) (Instat,

Graphpad Software 1993).

| Gra

de

and

sym

met

ry o

f n

orm

al f

etal

cor

tica

l dev

elop

men

t: a

lon

gitu

din

al t

wo

and

th

ree

dim

ensi

onal

ult

raso

un

d s

tud

y

65

RESULTS

215 ultrasound examinations were performed in 28 patients at a gestational age from

a median of 22 (range: 20 – 25) to a median of 37 (range: 33 – 40) weeks. Two patients

were excluded from the study and replaced with other patients: the fi rst when multiple

choroid plexus cysts were found at the fi rst examination, and the second because of

hospitalization in a different hospital due to symptomatic placenta previa. Delivery of

infants in the study occurred at a median of 39 weeks 6 days (range 37w 0 days – 42

weeks 0 days). The median birth weight was 3563g (range 2360 – 4100g). There were

12 male and 16 female infants, one of whom developed MEN IIb syndrome, and the

other of whom are all alive and well.

68% of structures could be seen with 2D and 70% with 3D (Fisher’s exact test p=0.1).

The quality of the 2D and 3D scans both had a mean score of 3 (standard deviation 1),

and were signifi cantly correlated (Spearman rho 0.528, p<0.01). The means were not

signifi cantly different (paired t test p=0.89). The examinations randomly selected for

the interobserver variation also had a mean quality score of 3 for both 2D and 3D.

Grading by 2D and 3D ultrasound were highly correlated (Spearman’s rho 0.92 – 0.97 for

the different sulci and 0.95 – 0.97 for the different cortical areas for the fi rst observer,

n=215, and 0.92 – 1.0 and 0.98 – 1.0 respectively for the second observer, n=19). The

gestational age at the fi rst appearance of different grades is presented in Figure 4.

With multilevel analysis, neither fetal gender nor laterality was found to have an

independent effect on cortical development. The only combination that nearly

reached signifi cance (p= 0.06) was the combination of female gender and laterality in

the development of the calcarine sulcus. Predictive curves are shown in Figure 5.

Intra- and inter-observer variation

The intra-observer variation showed a correlation (Spearman’s rho) of 0,933 (2D) and

0,927 (3D) and a Kappa of 0.6 (2D) and 0.63 (3D). The inter-observer variation showed

a correlation (Spearman’s rho) of 0,904 (2D) and 0,921 (3D) and a Kappa of 0.56 (2D)

and 0.57 (3D).

| Ch

apte

r 4

66

Sulcus Sylvian

Parieto-occipital

Calcarine

Superiortemporal

Central

Cingulate

Figure 4. First appearance of different grades of cortical development (continued)

| Gra

de

and

sym

met

ry o

f n

orm

al f

etal

cor

tica

l dev

elop

men

t: a

lon

gitu

din

al t

wo

and

th

ree

dim

ensi

onal

ult

raso

un

d s

tud

y

67

Cortical area Frontal

Parietal

Temporal

Occipital

Mesial

| Ch

apte

r 4

68

5.1 Sulci

Sylvian Superior temporal

Parieto-occipital Central

Calcarine Cingulate

Figure 5. Prediction of cortical grade by multilevel analysis (continued)

| Gra

de

and

sym

met

ry o

f n

orm

al f

etal

cor

tica

l dev

elop

men

t: a

lon

gitu

din

al t

wo

and

th

ree

dim

ensi

onal

ult

raso

un

d s

tud

y

69

5.2 Cortical areas

Frontal Occipital

Parietal Mesial

Temporal

| Ch

apte

r 4

70

Asymmetry

It was possible to visualize the parieto-occipital, calcarine and cingulate sulci and

occipital and mesial cortical areas on both sides in more than 90% of examinations. An

example of asymmetrical sulci is shown in Figure 6.

Symmetrical parieto-occipital fissure (27w)

Asymmetrical parieto-occipital fissure (33w)

Symmetrical calcarine sulcus (27w) Asymmetrical calcarine sulcus (28w)

Figure 6. Symmetrical and asymmetrical sulci

In total, 12% of examinations (64/514) were asymmetrical. In 8 fetuses all examinations

were symmetrical at all times, in 7 fetuses there was asymmetry once, in 6 fetuses

twice, and in 7 fetuses three of four times. The parieto-occipital sulcus was always

symmetrical in 15 fetuses, more advanced by one grade in 10 fetuses on the right hand

side and in 2 fetuses on the left hand side, and by two grades in 1 fetus on the right

hand side. The calcarine sulcus was always symmetrical in 13 fetuses, more advanced

by one grade in 4 fetuses on the right hand side and in 7 fetuses on the left hand side,

and by two grades in 1 fetus on the right hand side and in 3 fetuses on the left hand

| Gra

de

and

sym

met

ry o

f n

orm

al f

etal

cor

tica

l dev

elop

men

t: a

lon

gitu

din

al t

wo

and

th

ree

dim

ensi

onal

ult

raso

un

d s

tud

y

71

side. The cingulate sulcus was always symmetrical in 21 fetuses, more advanced by

one grade in 3 fetus on the right hand side and in 6 fetuses on the left hand side. The

occipital area was always symmetrical in 18 fetuses, more advanced by one grade in

2 fetuses on the right hand side and in 7 fetuses on the left hand side, and by two

grades in 1 fetus on the right hand side. There was a signifi cant correlation between

asymmetry in the calcarine sulcus and occipital area (Spearman’s rho = 0.84, p<0.001).

There was a signifi cant association between female gender and asymmetry of the

calcarine sulcus (Fisher’s exact test p=0.02), and a signifi cant association between

asymmetry of the parieto-occipital and calcarine sulci (Fisher’s exact test p=0.003),

although opposite sides tended to be more advanced. There was a signifi cant (p<0.01)

negative correlation between gestational age and degree of asymmetry in the parieto-

occipital fi ssure (Spearman’s rho=-0.35), the calcarine sulcus (Spearman’s rho=-0.48)

and the occipital area (Spearman’s rho=-0.46). Most instances of asymmetry occurred

before 28, and it was rare after 32 gestational weeks (Figure 7).

Figure 7. Asymmetry at different gestational ages

Most instances of asymmetry occurred at a maximal cortical grade of 3, with 53% of

examinations being asymmetrical. No asymmetry was seen with a maximal cortical

| Ch

apte

r 4

72

grade of 1, 10% of examinations with a maximal cortical grade of 2 was asymmetrical,

23% of grade 4, and 1% of grade 5.

The time spent with 2D and 3D respectively is depicted in Table 1.

Table 1. Scanning, analysis and total time with two dimensional (2D) and three dimensional (3D) ultrasound.

2D 3D Paired t test

Observer 1 Observer 2 Observer 1 Observer 2 Observer 1 Observer 2

Scanning time 5.2 (2.3) - 1.7 (0.9) - p<0.01 -

Analysis time 2.3 (1.0) 9.3 (2.9) 4.3 (1.4) 12.7 (2.8) p<0.01 <0.001

Total time 7.4 (2.6) 15.9 (8.1) 5.9 (1.9) 14.9 (3.9) p<0.01 0.5

Values in minutes (standard deviation)

DISCUSSION

Cortical development could be graded with excellent intra- and interobserver variability

during the latter half of pregnancy. Although 30-32% of sulci or cortical areas were

not visualized, this should represent the “worst-case” scenario, as the images were

evaluated after the scanning procedure and care was taken to limit the amount of

examination time in healthy volunteers. With more time and especially the use of

transvaginal ultrasound, it should be possible to visualize a greater percentage of sulci,

and visualize these with greater resolution.

Another disadvantage of transabdominal ultrasound is that the proximal hemisphere

of the fetal brain is partially obscured by reverberation artifacts (23;24). Due to this, it

has not been possible to determine the degree of asymmetry in many sulci and cortical

areas in this study. The reverberation artifacts have fortunately to a large degree been

solved by newer generation apparatus, but too late for this study (Figure 8) .

It is reassuring and not unexpected that there is a high degree of correlation between

grading with 2D and 3D ultrasound. With 3D ultrasound, asymmetry could be

confi rmed with greater confi dence, as it is possible to ensure a perfectly aligned plane

with an orthogonal display.

| Gra

de

and

sym

met

ry o

f n

orm

al f

etal

cor

tica

l dev

elop

men

t: a

lon

gitu

din

al t

wo

and

th

ree

dim

ensi

onal

ult

raso

un

d s

tud

y

73

Reverberation obscuring Sylvian fissure and lateral ventricle in proximal hemisphere(GE Voluson 730 Expert, 2005)

Good visualization of Sylvian fissure and lateral ventricle in proximal hemisphere (GE Voluson E8, 2008)

Figure 8. Visualization of proximal hemisphere

The shorter scanning decreases the time that the fetus is exposed to ultrasound, and

the total time (scanning and analysis time) was not increased with 3D ultrasound in

this study. Although the amount of ultrasound exposure during normal clinical use is a

small fraction of the amount found to disturb cortical migration in the mouse embryo

(25), it still makes sense to limit the amount of exposure (26). It is also reassuring that

3D ultrasound does indeed shorten the total scanning and analysis time, irrespective

of the amount of experience of the ultrasonographer.

In comparing the gestational age at which the different sulci become visible with

the results of other studies (Table 2), the fi rst appearance of sulci such as the Sylvian,

parieto-occipital and calcarine sulci in this study seemed to lag behind the results of

other ultrasound, MRI and anatomical studies. This is most likely a result of the later

gestational age at which this study started, where the Sylvian fi ssure was seen in all

patients in comparison with other studies, that started earlier in gestation.

The fi rst appearance of sulci developing later, such as the central or superior temporal

sulci, was seen in this study at a gestational age comparable with other antenatal

imaging studies. It is also noteworthy how different the gestational ages in two

different anatomical studies were. This might be partially explained that the study

reporting later development also examined brains from fetuses after 22 weeks, and

there might be some inaccuracy in the determination of the gestational ages (1;3).

| Ch

apte

r 4

74

Tab

le 2

. Ges

tati

onal

age

at

det

ecti

on o

f su

lci

2a. C

omp

aris

ion

wit

h a

nat

omic

al a

nd

MRI

stu

die

s: g

esta

tion

al a

ge (w

eeks

)

Stru

ctu

reA

nat

omic

al

app

eara

nce

(G

ovae

rt)

An

atom

ical

ap

pea

ran

ce

(Ch

i)

An

atom

ical

ap

pea

ran

ce

(Dor

ovin

i)

MR

I: D

etec

tab

le

in >

75%

of

bra

ins

(Lev

ine)

MR

I: A

pp

eara

nce

(L

an)

MR

I: D

etec

tab

le in

25

-75%

/ >

75%

of

bra

ins

(Gar

el)

Ult

raso

un

d:

Det

ecta

ble

in

25-7

5% /

>75%

of

bra

ins

(th

is s

tud

y)

Sylv

ian

fi ss

ure

1214

14-1

524

-25

/ 29

22 /

22

Pari

eto-

occi

pit

al fi

ssu

re16

1622

18-1

922

/ 2

3

Cal

cari

ne fi

ssu

re16

1622

18-1

922

-23

/ 24

-25

22 /

23

Cin

gula

te s

ulc

us

1618

2426

-27

3022

-23

/ 24

-25

23 /

26

Cen

tral

su

lcu

s20

2024

26-2

824

-26

24-2

5 /

2726

/ 2

6

Sup

erio

r te

mp

oral

su

lcu

s28

2328

28-2

924

-26

26 /

27

25 /

26

2b. C

omp

aris

ion

wit

h o

ther

ult

raso

un

d s

tud

ies:

ges

tati

onal

age

(wee

ks)

Sulc

us

Firs

t ap

pea

ran

ceFi

rst

app

eara

nce

/ d

etec

ted

in a

llD

etec

ted

in 2

5-75

% /

>75

%

Ber

nar

dM

onte

agu

do

Toi

This

stu

dy

Coh

en-S

ach

erTh

is s

tud

y

Sylv

ian

1918

22 /

23

22 /

22

Pari

eto-

occi

pit

al23

1818

/ 2

021

/ 2

418

/ 2

022

/ 2

3

Cal

cari

ne fi

ssu

re23

1818

/22

21 /

26

20 /

22

22 /

23

Cin

gula

te s

ulc

us

22-2

426

23/2

422

/ 2

720

/ 2

423

/ 2

6

Cen

tral

su

lcu

s23

/ 2

726

/ 2

726

/28

26 /

27

Sup

erio

r te

mp

oral

su

lcu

s26

- 28

24 /

29

28 /

3025

/ 2

6

| Gra

de

and

sym

met

ry o

f n

orm

al f

etal

cor

tica

l dev

elop

men

t: a

lon

gitu

din

al t

wo

and

th

ree

dim

ensi

onal

ult

raso

un

d s

tud

y

75

In the sulci and cortical areas where it was possible to determine asymmetry, this was

seen most clearly in the calcarine sulcus and occipital areas, and to a lesser extent in

the parieto-occipital and cingulate sulcus. This was seen more commonly in female

fetuses, but not related to the position of the fetus, and disappeared after 32 weeks.

Part of the decrease in the degree of asymmetry with advancing gestational age might

be related to the grading, which reached a plateau from around 34 weeks, but the

maximum asymmetry was seen well before this, namely before 28 weeks. Previously,

asymmetry in the hemisphere diameter was found at 20 – 22 weeks in both female

and male fetuses (27). Asymmetry in the development of the temporal lobe has

been demonstrated anatomically (28;29) and with MRI (unpublished data, D. Prayer

and G. Kasprian) with the right hand side developing earlier than the left. No other

antenatal imaging studies could be found which systematically evaluated symmetry

of cortical development. When the results of this study is compared with those of

anatomical studies (1;3), it is interesting that asymmetry was found in sulci where it

has not previously been described anatomically, and the calcarine sulcus which tended

to be more advanced on the right hand side in an anatomical study (1) was tended to

be more advanced on the left hand side in our study. The transience of asymmetry,

and the relationship with female gender in this study might explain some of these

differences. Multilevel analysis with Mixor (considering the grades as categorical

variables) and SPSS (considering the grades as continuous variables) yielded similar

results, and the predictive curves are shown as continuous lines generated from the

SPSS analyses. Advantages of multilevel analysis include that it takes dependence of

measurements in hierarchically structured data into account, and that it allows for

missing values and variation in frequency of measurements. Only the strongest effect

of laterality and gender, namely on the development of the calcarine sulcus, was born

out by multilevel analysis, demonstrating that asymmetry is an individual rather than

a systematic effect.

This study has also shown that it is feasible and not time-consuming to grade cortical

development accurately on transabdominal 2D and 3D ultrasound, and that a degree of

asymmetry can occur physiologically. It should be expanded to increase the number of

evaluations earlier in gestation, and done with apparatus where it is mostly possible to

visualize both hemispheres. Clinically, this technique could be applied as part of a fetal

neurosonographical examination to assist the assesment of cortical development.

| Ch

apte

r 4

76

REFERENCES

(1) Chi JG, Dooling EC, Gilles FH. Gyral development of the human brain. Ann Neurol 1977 Jan;1(1):86-93.

(2) Govaert P, de Vries L. Sulci and gyri. An atlas of neonatal brain sonography.London: Cambridge University Press;

2005. p. 1-6.

(3) Dorovini-Zis K, Dolman CL. Gestational development of brain. Arch Pathol Lab Med 1977 Apr;101(4):192-5.

(4) Garel C. The role of MRI in the evaluation of the fetal brain with an emphasis on biometry, gyration and

parenchyma. Pediatr Radiol 2004 Sep;34(9):694-9.

(5) Girard N, Raybaud C, Gambarelli D, Figarella-Branger D. Fetal brain MR imaging. Magn Reson Imaging Clin N Am

2001 Feb;9(1):19-56, vii.

(6) Lan LM, Yamashita Y, Tang Y, Sugahara T, Takahashi M, Ohba T, et al. Normal fetal brain development: MR imaging

with a half-Fourier rapid acquisition with relaxation enhancement sequence. Radiology 2000 Apr;215(1):205-

10.

(7) Levine D, Barnes PD. Cortical maturation in normal and abnormal fetuses as assessed with prenatal MR imaging.

Radiology 1999 Mar;210(3):751-8.

(8) Van der Knaap MS, Wezel-Meijler G, Barth PG, Barkhof F, Ader HJ, Valk J. Normal gyration and sulcation in

preterm and term neonates: appearance on MR images. Radiology 1996 Aug;200(2):389-96.

(9) Bernard C, Droulle P, Didier F, Gerard H, Larroche JC, Plenat F, et al. [Echographic aspects of cerebral sulci in the

ante- and perinatal period]. J Radiol 1988 Aug;69(8-9):521-32.

(10) Toi A, Lister WS, Fong KW. How early are fetal cerebral sulci visible at prenatal ultrasound and what is the normal

pattern of early fetal sulcal development? Ultrasound Obstet Gynecol 2004 Dec;24(7):706-15.

(11) Cohen-Sacher B, Lerman-Sagie T, Lev D, Malinger G. Sonographic developmental milestones of the fetal cerebral

cortex: a longitudinal study. Ultrasound Obstet Gynecol 2006 May;27(5):494-502.

(12) Monteagudo A, Timor-Tritsch IE. Development of fetal gyri, sulci and fi ssures: a transvaginal sonographic study.

Ultrasound Obstet Gynecol 1997 Apr;9(4):222-8.

(13) Malinger G, Kidron D, Schreiber L, Ben Sira L, Hoffmann C, Lev D, et al. Prenatal diagnosis of malformations of

cortical development by dedicated neurosonography. Ultrasound Obstet Gynecol 2007 Feb;29(2):178-91.

(14) Cohen-Sacher B, Sagie-Lerman T, Lev D, Malinger G. Sonographic characterization of the fetal cortex: a

longitudinal study(Abstracts of the 15th World Congress on Ultrasound in Obstetrics and Gynecology).

Ultrasound Obstet Gynecol 2005 Sep;26(4):309-471.

(15) Huisman TA, Martin E, Kubik-Huch R, Marincek B. Fetal magnetic resonance imaging of the brain: technical

considerations and normal brain development. Eur Radiol 2002 Aug;12(8):1941-51.

(16) Malinger G, Lerman-Sagie T, Watemberg N, Rotmensch S, Lev D, Glezerman M. A normal second-trimester

ultrasound does not exclude intracranial structural pathology. Ultrasound Obstet Gynecol 2002 Jul;20(1):51-6.

| Gra

de

and

sym

met

ry o

f n

orm

al f

etal

cor

tica

l dev

elop

men

t: a

lon

gitu

din

al t

wo

and

th

ree

dim

ensi

onal

ult

raso

un

d s

tud

y

77

(17) Malinger G, Lev D, Lerman-Sagie T. Abnormal sulcation as an early sign for migration disorders. Ultrasound

Obstet Gynecol 2004 Dec;24(7):704-5.

(18) Malinger G, Monteagudo A, Pilu G, Timor-Tritsch IE, Toi A. Sonographic examination of the fetal central nervous

system: guidelines for performing the ‘basic examination’ and the ‘fetal neurosonogram’. Ultrasound Obstet

Gynecol 2007 Jan;29(1):109-16.

(19) Timor-Tritsch IE, Monteagudo A. Transvaginal fetal neurosonography: standardization of the planes and sections

by anatomic landmarks. Ultrasound Obstet Gynecol 1996 Jul;8(1):42-7.

(20) Pilu G, Ghi T, Carletti A, Segata M, Perolo A, Rizzo N. Three-dimensional ultrasound examination of the fetal

central nervous system. Ultrasound Obstet Gynecol 2007 Aug;30(2):233-45.

(21) Quarello E., Stirnemann J., Guibaud L., Ville Y. Assessment of the Sylvia fi ssure operculization (SFO) at between

22 and 32 weeks of gestation: a subjective approach (part I). Ultrasound Obstet Gynecol 2007;30:418.

(22) Merz E, Benoit B, Blaas HG, Baba K, Kratochwil A, Nelson T, et al. Standardization of three-dimensional images in

obstetrics and gynecology: consensus statement. Ultrasound Obstet Gynecol 2007 Jun;29(6):697-703.

(23) Pilu G, Nicolaides K. Central nervous system. Diagnosis of fetal abnormalities: The 18 - 23 week scan.London:

Parthenon; 1999. p. 5-18.

(24) Monteagudo A. Fetal neurosonography: should it be routine? Should it be detailed? Ultrasound Obstet Gynecol

1998 Jul;12(1):1-5.

(25) Ang ES, Jr., Gluncic V, Duque A, Schafer ME, Rakic P. Prenatal exposure to ultrasound waves impacts neuronal

migration in mice. Proc Natl Acad Sci U S A 2006 Aug 22;103(34):12903-10.

(26) Abramowicz JS. Prenatal exposure to ultrasound waves: is there a risk? Ultrasound Obstet Gynecol 2007

Apr;29(4):363-7.

(27) Hering-Hanit R, Achiron R, Lipitz S, Achiron A. Asymmetry of fetal cerebral hemispheres: in utero ultrasound

study. Arch Dis Child Fetal Neonatal Ed 2001 Nov;85(3):F194-F196.

(28) Bossy J, Godlewski G, Maurel JC. [Study of right-left asymmetry of the temporal planum in the fetus]. Bull Assoc

Anat (Nancy ) 1976 Jun;60(169):253-8.

(29) Teszner D, Tzavaras A, Gruner J, Hecaen H. [Right-left asymmetry of the planum temporale; apropos of the

anatomical study of 100 brains]. Rev Neurol (Paris) 1972 Jun;126(6):444-9.

Chapter 5A LONGITUDINAL ULTRASOUND STUDY OF THE

SIZE AND SYMMETRY OF THE FETAL LATERAL

CEREBRAL VENTRICLES

L. Pistorius1, E.J.H. Mulder1, M. Rutten1, S. Kuc1, M. Benders2, G.H.A. Visser1

1Department Obstetrics, Division Perinatology and Gynaecology, University Medical Centre Utrecht, Utrecht, The Netherlands

2Department Neonatology, Division Perinatology and Gynaecology, University Medical Centre Utrecht, Utrecht, The Netherlands

Submitted

| Ch

apte

r 5

80

ABSTRACT

Introduction

On the standard transventricular view, usually only one cerebral ventricle can

be visualized. We undertook a study to see whether it is possible to visualize both

ventricles on the standard view and in the midcoronal plane with three dimensional

(3D) ultrasound and with static volume contrast imaging (VCI), to generate longitudinal

growth curves and to determine the prevalence of ventricular asymmetry.

Methods

A cohort of 28 patients was examined every two weeks from 20 until 40 weeks by 2D

and 3D ultrasound. The ventricular atrium, index (VI) and frontal horn width (FHW)

were measured by two observers. Interobserver variation was calculated. Multilevel

analysis was used to evaluate the effect of gestational age and gender on ventricular

measurements.

Results

Both ventricles could be evaluated in the axial plane signifi cantly more often using

3D (42%) and VCI (55%) than with 2D (5%). Both frontal horns could be evaluated

in more than 60% of examinations with 2D, 3D and VCI. Spearman’s rho calculated

for the correlation between measurements with 2D, 3D and VCI ranged from 0.31-

0.67 (p<0.001) and Spearman’s rho calculated for inter-observer correlation ranged

from 0.36-0.80. Asymmetry of the ventricles was seen in 0-4% of measurements.

Longitudinal growth curves based on individual developmental trajectories were

generated.

Conclusion

The midcoronal view and the axial view with VCI proved useful to evaluate both left

and right lateral cerebral ventricles. There was a good interobserver correlation, and

the differences between 2D, 3D and VCI measurements were all small. Asymmetry

was uncommon, but more common than previously reported.

| A lo

ngi

tudi

nal

ult

raso

un

d s

tud

y o

f th

e si

ze a

nd

sym

met

ry o

f th

e fe

tal l

ater

al c

ereb

ral v

entr

icle

s

81

INTRODUCTION

Evaluation of the lateral cerebral ventricles by assessment of the atrium width on the

transventricular axial plane is a routine part of structural fetal ultrasound examination

(1) (and has been for decades (2)). This evaluation has the advantage that the upper

limit of the atrial measurement (four standard deviations above the mean) remains

around 10mm throughout gestation, but it is not without its drawbacks. Reverberation

artefacts may cause a fuzzy view of the hemisphere closest to the transducer on

transabdominal examination (3). Although it has improved with top-end ultrasound

machines, it is often diffi cult to diagnose asymmetrical ventriculomegaly on this view,

especially if the ventricle distal to the transducer is of normal diameter. The fact that

asymmetrical ventriculomegaly is a very rare fi nding antenatally with a prevalence of

less than 1% (4) as compared to postnatal asymmetry in excess of 40% (5), might at

least in part be due to this ascertainment bias.

There are at least two possible solutions for this problem. The fi rst is to assess the frontal

horns of the lateral ventricles on a midcoronal plane (6), which is the view commonly

used by neonatologists (7;8) and usually allows visualization of both ventricles. A

possible disadvantage of this approach is that mild ventriculomegaly affecting only the

posterior horns might be missed on this view, although the frontal horn confi guration

should also be abnormal in true colpocephaly (9). Another approach would be to

reconstruct the axial plane with three dimensional (3D) ultrasound from different

acquisition planes which might be less prone to obscuring of one lateral ventricle. A

possible disadvantage to this approach is that the resolution of a reconstructed plane

is always less than that of the acquisition plane, but this may be compensated by using

volume contrast imaging (VCI) (10).

We therefore decided to perform a longitudinal study to assess our ability to visualize

both ventricular atria and frontal horns with antenatal two (2D) and three dimensional

ultrasound, the latter with and without VCI.

METHODS

Twenty eight healthy pregnant women were included after a screening ultrasound

examination at 20 weeks had demonstrated a singleton pregnancy with no abnormal

fi ndings. Other inclusion criteria were confi rmation of pregnancy duration by ultrasound

| Ch

apte

r 5

82

during the fi rst trimester, and absence of risk factors which might infl uence fetal

growth or development, such as maternal disease or previous intra-uterine growth

restriction. The study was approved by the Medical Ethics Committee of the University

Medical Centre Utrecht.

Ultrasound examinations were carried out at two weekly intervals with a General

Electric Voluson 730 Expert (General Electric Healthcare, London). Biometry (including

fetal head circumference, abdominal circumference and femur length) was performed,

followed by neurosonography if the biometry conformed to the gestational age.

Transabdominal neurosonography was performed by one of the authors (LP) with a

2.6-7.7MHz 2D ultrasound probe and a 4.0-8.5MHz 3D probe. On 2D ultrasound, it was

attempted to obtain the standard neurosonographic planes (6). Three images of each

plane were stored digitally for subsequent evaluation. The position of the fetus (right

side or left side anterior) was noted. On 3D ultrasound, three static volumes were

obtained with high to maximum quality (depending on the amount of fetal movement):

the fi rst starting with the axial transventricular plane, the second starting with the

axial transcerebellar plane, and the third starting with the midcoronal plane. An angle

of 45°-65° was chosen to include the maximum amount of fetal brain that could be

visualized, but kept as narrow as possible to shorten scanning time. If fetal movement

was observed during the volume acquisition, the volume was acquired again.

Image J version 1.38 (NIH, USA) was used to measure the recorded 2D images. 4DView

version 7.0 (General Electric Healthcare, London) was used to measure the 3D volumes.

Measurement of the ventricles was done by two observers (LP and SK) without being

aware of each others’ measurements. Each measurement was done three times, and

the median used for analysis. 3D images were reoriented to facilitate identifi cation

of the right and left sides (11). Measurements on 3D volumes were done at maximal

magnifi cation with and without static volume contrast imaging (VCI) set at slice

thickness 3mm. An attempt was made to measure both left and right sides at each

examination. The ventricular atria were measured on the transventricular plane in a

standardized fashion. The (inner-inner) width of the frontal horns (frontal horn width;

FHW) and the distance from the inner side of the lateral edge of the frontal horns to

the midline (ventricular index, VI) were measured in a coronal plane at the foramina of

Munro, in front of the choroid plexus in the third ventricle (7;12) (Figure 1).

Data management and statistical analysis (including multilevel analysis mixed

model option to identify variables with an independent effect on the time course of

ventricular growth) were performed with SPSS version 15.0 for Windows (SPSS Inc.

| A lo

ngi

tudi

nal

ult

raso

un

d s

tud

y o

f th

e si

ze a

nd

sym

met

ry o

f th

e fe

tal l

ater

al c

ereb

ral v

entr

icle

s

83

Chicago, Illinois). The number of structures visualized well enough to be graded reliably

with 2D, 3D and VCI, respectively, was compared with the chi square test.

We calculated Spearman’s rho to determine the correlation between the measurements

of each observer and we performed a Bland-Altman analysis to plot the difference

between the measurements of the two observers against their mean.

To obtain the best-fi t equation for multilevel analysis to compare the ventricular

measurements with gestational age, we used a model with random slope and

random intercept. We used the linear mixed model option with maximum likelihood

procedure. 2D ultrasound: axial transventricular plane 2D ultrasound: midcoronal plane

3D ultrasound: plane of acquisition 3D ultrasound: reoriented view

3D ultrasound: volume contrast imaging (VCI ) Note increased contrast (e.g. vermis)

3D ultrasound: measurement of ventricular atrium

3D ultrasound: measurement of frontal horns

1= left ventricular index (V I) 2= right V I 3=right frontal horn width (FHW) 4=left FHW

Figure 1. Evaluation of ventricles

| Ch

apte

r 5

84

Asymmetry was defi ned as a difference of more than two standard deviations

between the measurement on the left and on the right hand side (4). The occurrence

of asymmetry was calculated as percentage of the total number of fetuses in whom

both left and right ventricles could be visualized, with confi dence intervals defi ned by

the modifi ed Wald method (13).

RESULTS

215 ultrasound examinations were performed in 28 patients at a gestational age from

a median of 22 (range: 20 – 25) to a median of 37 (range: 33 – 40) weeks. Delivery of

infants in the study occurred at a median of 39 weeks 6 days (range 37w 0 days – 42

weeks 0 days). The median birth weight was 3563g (range 2360 – 4100g). There were

12 male and 16 female infants, one of whom developed MEN IIb syndrome. All other

infants were alive and well in the neonatal period.

The percentage and number of measurements that was possible with the three

methods is depicted in Figure 2 and Table 1.

Figure 2. Visualization with 2D, 3D and VCI

| A lo

ngi

tudi

nal

ult

raso

un

d s

tud

y o

f th

e si

ze a

nd

sym

met

ry o

f th

e fe

tal l

ater

al c

ereb

ral v

entr

icle

s

85

f

Table 1. VisualizationNumber of examinations where it was possible to measure neither side / one side only / both sides

2D 3D VCI Total

Atria 10 / 194 / 11 14 / 111 / 90 6 / 90 / 119 301 / 3951 / 2201

VI 50 / 31 / 134 49 / 16 / 150 25 / 43 / 147 124 / 90 / 431

FHW 51 / 32 / 132 41 / 46 / 128 25 / 50 / 140 117 / 128 / 400

Total 111 / 2572 / 2772 104 / 173 / 368 563 / 183 / 4064 271 / 613 / 1051

Chi square:1p<0.0001 (compared to VI and FHW) 2p<0.0001 (compared to 3D and VCI) 3p<0.0001 (compared to 2D and 3D)4p=0.03 (compared to 3D)

Measurements with 2D and 3D ultrasound with and without VCI were signifi cantly

correlated and the 95% confi dence interval was less than 1 mm for all comparisons

(Table 2). The inter-observer variation correlations and differences are shown in Table

3 and Figure 3.

Table 2. Correlations and differences between 2D, 3D and VCI

2D-3D 2D-VCI 3D-VCI

Spearman p Difference (mm)

Spearman p Difference (mm)

Spearman p Difference (mm)

Atria 0.51 <0.001 -0.3(-0.6 - -0.1)

0.67 <0.001 0.0(-0.2 – 0.2)

0.51 <0.001 0.1(-0.5 – 0.7)

VI 0.51 <0.001 -0.4(-0.7 - -0.2)

0.59 <0.001 0.3(0.1 – 0.5)

0.61 <0.001 0.6(0.4 – 0.8)

FHW 0.32 <0.001 0.5(0.3 – 0.6)

0.52 <0.001 0.3(0.2 – 0.4)

0.31 <0.001 0.0(-0.1 – 0.1)

2D=two dimensional ultrasound 3D=three dimensional ultrasoundVCI=volume contrast imaging Atrium=ventricular atrium (axial plane)VI=ventricular index (coronal plane) FHW=frontal horn width (coronal plane)Measurements as mm (95% confi dence intervals)

Table 3. Interobserver correlation: 2D, 3D and VCI

2D 3D VCI

Spearman p Spearman p Spearman p

Atria 0.80 <0.001 0.65 <0.001 0.58 0.001

VI 0.36 0.054 0.52 0.001 0.52 0.001

FHW 0.66 <0.001 0.42 0.01 0.38 0.02

2D=two dimensional ultrasound 3D=three dimensional ultrasoundVCI=volume contrast imaging Atrium=ventricular atrium (axial plane)VI=ventricular index (coronal plane) FHW=frontal horn width (coronal plane)

| Ch

apte

r 5

86

VCI measurements were used for further analysis. The atrial diameter was

normally distributed; the VI and FHW were normally distributed after log naturalis

transformation. Using multilevel analysis, neither gestational age (p=0.91) nor fetal

gender (p=0.93) were predictors of atrial diameter. The gestational age (p<0.0001)

and squared gestational age (p<0.0001) had main effects on VI, and there was a fetal

gender by gestational age interaction that attained borderline signifi cance (p=0.058).

For the FHW, we found a main effect of the squared gestational age (p=0.002), but not

gestational age (p=0.46) or gender (p=0.77) and a fetal gender by squared gestational

age interaction effect that showed a trend towards signifi cance (p=0.097). The normal

growth curves and measured values are shown in Figures 4 and 5.

Two dimensional ultrasound

Three dimensional ultrasound

Volume contrast imaging

Atrium

Ventricle index

Frontal horn width

Figure 3. Bland-Altman plotsLines represent mean +/- 2 standard deviationsY axes = difference between two measurementsX axes = mean of two measurements

| A lo

ngi

tudi

nal

ult

raso

un

d s

tud

y o

f th

e si

ze a

nd

sym

met

ry o

f th

e fe

tal l

ater

al c

ereb

ral v

entr

icle

s

87

Five instances of asymmetry of atrial measurements were noted out of 119

measurements (4%, 95% confi dence intervals (CI)=2-10%), no instances of asymmetry

of VI measurements (95% CI 0-3%) and four instances of asymmetry of FHW

measurements (3%, 95% CI 1-7%). Asymmetry was not related with the measurement

modality (2D, 3D or VCI), gestational age, fetal gender or fetal position.

Figure 4. Atrial diameter Lines represent mean (atrial diameter = 5.87mm), 5th and 95th centiles

| Ch

apte

r 5

88

Figure 5. Ventricle index (VI) and frontal horn width (FHW) Lines represent mean, 5th and 95th centilesVI (triangles) = exp(0,46+0,0987*GA-0,001279*GA2)FHW (circles) = exp((0,802-0,000182* GA2)

DISCUSSION

The usefulness of static VCI and C-plane VCI (VCI-C) to visualize cerebral midline

structures has previously been demonstrated (10;14) VCI is a technique which

enhances contrast by fi ltering out “noise” pixels and enhancing informative pixels. It

can either be used live in the A or the C-plane during 3D scanning (VCI-A or VCI-C), or

subsequently on a stored 3D volume (static VCI) (15). Our study has demonstrated that

static VCI is useful to visualize both cerebral ventricles, with signifi cantly fewer cases

where only one or no lateral ventricle could be visualized as compared to 2D or 3D.

Both frontal horns could be visualized in more than 60% of fetuses, whether 2D, 3D

| A lo

ngi

tudi

nal

ult

raso

un

d s

tud

y o

f th

e si

ze a

nd

sym

met

ry o

f th

e fe

tal l

ater

al c

ereb

ral v

entr

icle

s

89

or VCI was used. However, when analyzing the 3D volumes which were obtained from

a midcoronal plane, it was also obvious that the plane obtained on 2D was often not

truly coronal. A small difference in angulation should not make a big difference in

measurement, though, as the ventricles have a fairly regular shape. Another advantage

of 3D is that it is easy to identify the right and left hand side by reorienting the image

in a standardized fashion (11). The small difference in measurements between 2D and

VCI is reassuring and indicated that VCI can be used to increase the contrast resolution

without changing the measurements. We cannot extrapolate these results, however,

to instances where very small differences in measurement could have a large impact,

such as in nuchal translucency measurement (16). The fact that the inter-observer

agreement with VCI was at least as good as with 2D or 3D, while allowing the greatest

chance of visualizing both ventricles, makes VCI the modality of choice to identify and

measure the cerebral ventricles.

Using VCI to optimize the visualization of both ventricles, the incidence of asymmetry

was still low, although much higher than the 0,29% previously reported (4). However,

that study was performed using 2D ultrasound, and it was not specifi ed in how many

fetuses both ventricles were seen. If this was as low as in our 2D examinations, the

“true” percentage might also have been similar to our fi ndings of symmetry in up to

4% of examinations. The neonatal occurrence is still 5 to 10 times higher (5), which

might be the result of intrapartum events or simply the effect of gravity.

The results of this longitudinal study using individual growth trajectories confi rm

previous cross-sectional studies which showed a stable atrial diameter throughout

gestational age (17;18). If the cut-off of 10mm for the atrial diameter is used, only

three out of 328 measurements would have been classifi ed as abnormal (0.9%; 95%

confi dence interval 0.18-2.8%) (Figure 6). We demonstrated only a small effect of

gestational age on the FHW. If a cut-off is calculated for the FHW in a similar way as for

the atrium (17), 4mm would classify two out of 329 measurements as abnormal (0.6%;

95% confi dence interval 0.02-2.3%) (Figure 7). This latter value would need to be tested

prospectively to determine whether it discriminates equally well between normal and

abnormal ventricle diameters as does an atrial diameter of 10mm. There is unlikely

to be a single measurement which gives an optimal distinction between normality

and pathology, as can be seen from the high percentage of infants with a normal

neurological development after an antenatal diagnosis of mild ventriculomegaly (19).

The differences between male and female fetuses were small: no signifi cant difference

could be found using multilevel analysis. The absolute differences were also small. The

| Ch

apte

r 5

90

atrium was on average 0.2mm smaller, the VI 0.4mm larger, and the FHW 0.05mm

larger in male than in female fetuses. Even a statistically signifi cant difference would

therefore have been clinically insignifi cant.

Figure 6. Maximum atrial diameter

It has been shown that it is relatively easy to obtain a midfrontal sagittal view with

2D, and 3D could then be used whenever this is not feasible (14). Our results show

that the midcoronal view can also be obtained in the majority of fetuses with 2D. VCI

or 3D could be used where visualization with 2D is inadequate. When examining the

fetal heart, it would be considered the most basic examination to view only a four-

chamber view. In a similar way, in any but the most basic examination of the fetal

brain, we believe that the midsagittal and midcoronal views which are usually easy to

obtain, offer a wealth of additional information, and should form part of the normal

repertoire of planes in addition to axial views.

| A lo

ngi

tudi

nal

ult

raso

un

d s

tud

y o

f th

e si

ze a

nd

sym

met

ry o

f th

e fe

tal l

ater

al c

ereb

ral v

entr

icle

s

91

Figure 7. Maximum frontal horn width

The usefulness of the measurements of the frontal horn width to distinguish between

fetuses with intracranial pathology and those with a variation of the normal would

need to be evaluated prospectively. An added bonus of this view is that it corresponds

to neonatal practice, and should facilitate direct comparison between antenatal and

postnatal examinations.

| Ch

apte

r 5

92

REFERENCES

(1) Malinger G, Monteagudo A, Pilu G, Timor-Tritsch IE, Toi A. Sonographic examination of the fetal central nervous

system: guidelines for performing the ‘basic examination’ and the ‘fetal neurosonogram’. Ultrasound Obstet

Gynecol 2007; 29(1):109-116.

(2) Filly RA, Goldstein RB, Callen PW. Fetal ventricle: importance in routine obstetric sonography. Radiology 1991;

181(1):1-7.

(3) Monteagudo A. Fetal neurosonography: should it be routine? Should it be detailed? Ultrasound Obstet Gynecol

1998; 12(1):1-5.

(4) Achiron R, Yagel S, Rotstein Z, Inbar O, Mashiach S, Lipitz S. Cerebral lateral ventricular asymmetry: is this a

normal ultrasonographic fi nding in the fetal brain? Obstet Gynecol 1997; 89(2):233-237.

(5) Shen EY, Huang FY. Sonographic fi nding of ventricular asymmetry in neonatal brain. Arch Dis Child 1989;

64(5):730-732.

(6) Timor-Tritsch IE, Monteagudo A. Transvaginal fetal neurosonography: standardization of the planes and sections

by anatomic landmarks. Ultrasound Obstet Gynecol 1996; 8(1):42-47.

(7) Davies MW, Swaminathan M, Chuang SL, Betheras FR. Reference ranges for the linear dimensions of the

intracranial ventricles in preterm neonates. Arch Dis Child Fetal Neonatal Ed 2000; 82(3):F218-F223.

(8) Levene MI. Measurement of the growth of the lateral ventricles in preterm infants with real-time ultrasound.

Arch Dis Child 1981; 56(12):900-904.

(9) Levine D, Trop I, Mehta TS, Barnes PD. MR imaging appearance of fetal cerebral ventricular morphology.

Radiology 2002; 223(3):652-660.

(10) Vinals F, Munoz M, Naveas R, Giuliano A. Transfrontal three-dimensional visualization of midline cerebral

structures. Ultrasound Obstet Gynecol 2007; 30(2):162-168.

(11) Merz E, Benoit B, Blaas HG, Baba K, Kratochwil A, Nelson T et al. Standardization of three-dimensional images in

obstetrics and gynecology: consensus statement. Ultrasound Obstet Gynecol 2007; 29(6):697-703.

(12) Monteagudo A, Timor-Tritsch IE, Moomjy M. Nomograms of the fetal lateral ventricles using transvaginal

sonography. J Ultrasound Med 1993; 12(5):265-269.

(13) Agresti A, Coull BA. Approximate is better than “Exact” for interval estimation of binomial proportions. American

Statistician 1998; 52:119-126.

(14) Pilu G, Segata M, Ghi T, Carletti A, Perolo A, Santini D et al. Diagnosis of midline anomalies of the fetal brain with

the three-dimensional median view. Ultrasound Obstet Gynecol 2006; 27(5):522-529.

(15) Merz E. Current 3D/4D ultrasound technology in prenatal diagnosis. Eur Clinics Obstet Gynaecol 2005; 1(3):184-

193.

(16) Paul C, Krampl E, Skentou C, Jurkovic D, Nicolaides KH. Measurement of fetal nuchal translucency thickness by

three-dimensional ultrasound. Ultrasound Obstet Gynecol 2001; 18(5):481-484.

| A lo

ngi

tudi

nal

ult

raso

un

d s

tud

y o

f th

e si

ze a

nd

sym

met

ry o

f th

e fe

tal l

ater

al c

ereb

ral v

entr

icle

s

93

(17) Cardoza JD, Goldstein RB, Filly RA. Exclusion of fetal ventriculomegaly with a single measurement: the width of

the lateral ventricular atrium. Radiology 1988; 169(3):711-714.

(18) Hilpert PL, Hall BE, Kurtz AB. The atria of the fetal lateral ventricles: a sonographic study of normal atrial size and

choroid plexus volume. AJR Am J Roentgenol 1995; 164(3):731-734.

(19) Ouahba J, Luton D, Vuillard E, Garel C, Gressens P, Blanc N et al. Prenatal isolated mild ventriculomegaly: outcome

in 167 cases. BJOG 2006; 113(9):1072-1079.

Chapter 6FETAL CEREBELLAR VOLUME AND SYMMETRY ON

THREE-DIMENSIONAL ULTRASOUND:

VOLUME MEASUREMENT WITH MULTIPLANAR

AND VOCAL TECHNIQUES

M.J. Rutten1, L.R. Pistorius1, E.J.H. Mulder1, Ph. Stoutenbeek1, L.S. de Vries2, G.H.A. Visser1

1Department Obstetrics, Division Perinatology and Gynaecology, University Medical Centre Utrecht, Utrecht, The Netherlands

2Department Neonatology, Division Perinatology and Gynaecology, University Medical Centre Utrecht, Utrecht, The Netherlands

Ultrasound Med Biol 2008 (in press)

| Ch

apte

r 6

96

ABSTRACT

The purpose of this prospective longitudinal study was to evaluate the growth of the

fetal cerebellar volume by means of three dimensional ultrasound, to evaluate whether

there is a difference between the volumes of the left and right cerebellar hemispheres,

and to evaluate the intra- and inter-observer reliability of two different techniques of

volume measurement.

Three-dimensional ultrasound examinations were performed every two to three

weeks on 27 fetuses between 20 and 40 weeks’ gestation. Measurements of the total

cerebellar volume and of the left and right cerebellar hemispheres were done using

the multiplanar technique. Multilevel analysis was used to determine the growth of

cerebellar volume based on individual developmental trajectories and to compare

the volume of the right and left hemispheres of the cerebellum. The intra- and

interobserver reliability was calculated for the multiplanar and VOCALTM techniques in

a subgroup of ten fetuses. A growth curve of cerebellar volume in normal pregnancy

was generated.

The left cerebellar hemisphere was signifi cantly larger (12.3%, p<0.01) than the right.

The intraclass correlation coeffi cient for the measurements by the two techniques

was 0.99. Intra-observer reliability: the intraclass correlation coeffi cient for the

measurements using the multiplanar technique was 0.96 and 0.97 and for VOCALTM

it was 0.98 and 0.97 for the two observers respectively. Inter-observer reliability:

the intraclass correlation coeffi cient for the measurements using the multiplanar

technique was 0.97 and for VOCALTM 0.98. Longitudinal growth curves based on

individual developmental trajectories were generated for the cerebellar volume.

The left fetal cerebellar hemisphere was found to be signifi cantly larger than the

right. Both multiplanar and VOCALTM techniques had a good intra- and inter-observer

reliability and yielded very similar results.

| Fet

al c

ereb

ella

r vo

lum

e an

d s

ymm

etry

on

th

ree-

dim

ensi

onal

ult

raso

un

d

97

INTRODUCTION

Three-dimensional (3D) ultrasound has made it possible to measure volumes of organs

with an irregular shape more accurately. Previous studies have demonstrated that the

cerebellar volume is highly correlated with gestational age (GA) and with the transverse

cerebellar diameter (TCD). However, there were obvious differences in the volumes

obtained in different studies with the cerebellar volume from Oriental studies (1;2) 9

to 15% smaller at 32 weeks than in a Brazilian study (3).

Asymmetry of the cerebral hemispheres has been demonstrated with prenatal

ultrasound (4), but possible asymmetry of the cerebellar volume has not been studied

systematically in the fetus. Postnatal cerebellar asymmetry as a consequence of

atrophy has been described (5), and it would be useful to know whether and to what

extent physiological asymmetry of the fetal cerebellar hemispheres is present.

Two commonly used techniques of volume measurement are the multiplanar method

and the Virtual Organ Computer-Aided-Analysis (VOCALTM). The latter technique

involves rotating the data volume obtained by ultrasound around a fi xed axis in

sequential steps. The multiplanar technique entails scrolling through one plane while

manually measuring parallel slices in an orthogonal plane (Figure 1). The methods

have been validated in experimental and gynecological applications (6-9), but few

data are available on the intra- and inter-observer reliability of both methods in fetal

ultrasound (10).

We therefore decided to develop a growth curve of the total fetal cerebellar volume

and in a Dutch population, to evaluate asymmetry of the cerebellar hemispheres and

to evaluate the intra- and inter-observer reliability of the multiplanar and the VOCALTM

technique in measuring cerebellar volume.

METHODS

Twenty seven healthy pregnant women were included after a screening ultrasound

examination at 20 weeks had demonstrated no abnormal fi ndings in a singleton

pregnancy. Other inclusion criteria were confi rmation of pregnancy duration by

ultrasound during the fi rst trimester and absence of risk factors which might infl uence

fetal growth or development, such as maternal disease or previous intra-uterine growth

restriction. The study was approved by the Medical Ethics Committee of the University

| Ch

apte

r 6

98

Medical Hospital Utrecht. Informed consent was obtained from each participant prior

to inclusion.

VOCALTM: rotational measurement of

volume around a fixed axis through a

number of sequential steps.

Multiplanar technique: measurement

of volume per slice along axis

Figure 1. Schematic illustration of measurement techniques

Ultrasound examinations were carried out by the same observer (LP) at an interval

of two to three weeks beginning at a gestational age between 20 and 24 weeks.

Firstly, normal biometry of the head circumference, transverse cerebellar diameter,

abdominal circumference and femur length was confi rmed. 3D ultrasound of the

intracranial contents was then carried out using a GE Voluson 730 Expert (General

Electric Healthcare, London) with a 4.0-8.5MHz abdominal probe (RAB4-8) when the

fetus was at rest. If fetal movement was observed during the volume acquisition

or movement artefacts were visible in the B-plane of the orthogonal display of the

scanned volume, the volume was discarded and replaced with another volume. All 3D

volumes were saved for offl ine examination with 4DView version 5.3 (General Electric

Healthcare, London). Data management and statistical analyses were performed using

SPSS 12.0 (SPSS Inc. Chicago, Illinois).

The display of the orthogonal planes from the 3D volume was standardized to facilitate

identifi cation of the right and left hand sides (11). (Figure 2) When the outlines of the

cerebellum could not be identifi ed accurately the examination was excluded.

For the multiplanar measurements we used the sagittal plane to outline the external

surface of the cerebellum and measured the cerebellar area slice by slice at 1 mm

intervals along the transverse axis. The program then calculated the volume making use

| Fet

al c

ereb

ella

r vo

lum

e an

d s

ymm

etry

on

th

ree-

dim

ensi

onal

ult

raso

un

d

99

of the outlined circumferences. All measurements were done three times. The volume

of the hemispheres were recorded separately by using the falx to aid identifi cation of

the midline.

Figure 2. Standardized display of orthogonal planes

To determine the intra-and inter-observer reliability the volumes of the fi rst ten fetuses

were measured by two observers (MR and LP) three times each for the multiplanar

and VOCALTM technique. For the VOCALTM measurements we used the axial plane as

reference plane. After defi ning the lateral poles of the cerebellum with the calipers

we outlined the external surface of the organ manually at rotations of 15° for a

gestational age up to 30 weeks and 9° from 30 weeks onwards. By the end of the

rotational process the program calculated the volume automatically and provided a

3D image of the organ volume.

The intra-observer reliability was evaluated by calculating the intraclass correlation

coeffi cient for the minimum and maximum of the three measurements that were

obtained of each volume with each measurement method. We determined the inter-

observer reliability by calculating the intraclass correlation coeffi cient for the median

of the three measurements of each observer for each measurement method. We

constructed Bland-Altman plots using the median measurement from both observers

to plot the difference between against the mean of the two medians. The intraclass

correlation was also used to compare the median values obtained with the multiplanar

| Ch

apte

r 6

100

and VOCALTM techniques.

To develop growth curves the median of the three volume measurements with

the multiplanar technique by one observer (MR) and the median of the volume

measurements with VOCALTM of both observers were used. To obtain the best-fi t

equation for multilevel analysis to compare the cerebellar volume with gestational age,

we used the log naturalis in a model with random slope and random intercept, using

the linear mixed model option with restricted maximum likelihood procedure. We

composed growth lines with a 95% confi dence interval. Possible effects of symmetry,

fetal gender, lie (cephalic, breech or transverse) and position (right or left hand side

towards the observer) were also evaluated with multilevel analysis.

RESULTS

A total of 188 ultrasound examinations were carried out on 27 fetuses between 20

and 39 weeks. Forty three volume blocks (22.9%) were excluded due to insuffi cient

quality for accurate interpretation (13.8% of volume blocks up to 32 weeks and 43%

of the volume blocks after 32 weeks), leaving 145 volume blocks which were suitable

for analysis of the whole cerebellum and the left and right hemispheres, resulting

in a median of 6 (range 2-8) suitable volume blocks per fetus. These blocks were all

measured using the multiplanar technique. The VOCALTM technique was used in 58

volumes in ten fetuses, with a median of 6 (range 3-8) volume blocks per fetus.

With multilevel analysis, the growth curve of the cerebellar volume from our data

obtained with the multiplanar technique was predicted by ln(volume, mm3) =

1.502+0.052(GA)-0.000076(GA)2 (GA=gestational age in days) (Figure 3). For the

measurements obtained with VOCALTM the growth curve was ln(volume, mm3) =

0.047+0.064(GA)-0.0001(GA)2, yielding a very similar curve (Figure 4).

The main right-left effect was described by ß=-0.39(SE 0.13); p=0.003 and right-left by

gestational age interaction effect ß=0.0013 (SE 0.00064) p=0.0047. The left hemisphere

had a signifi cantly larger volume than the right hemisphere (median difference 12.3%,

range -48 – 54%) p< 0.001, paired t-test) which was more pronounced earlier in

pregnancy. (The median percentage difference until 32 weeks was 15.2% (-6% – +38%)

and after 32 weeks was 6.8% (-24% - +39%). After 36 weeks, the median percentage

difference was 5,4% (right-left difference not statistically signifi cant). With multilevel

analysis, the fetal gender, lie and position had no infl uence on the total cerebellar

| Fet

al c

ereb

ella

r vo

lum

e an

d s

ymm

etry

on

th

ree-

dim

ensi

onal

ult

raso

un

d

101

volume or asymmetry. The intraclass correlation coeffi cient for the measurements by

the two techniques was 0.99 (95% CI 0.98 – 0.99).

Figure 3. Cerebellar volume by gestational age (median and 95% confi dence intervals)

Figure 4. Comparison between cerebellar volume by multiplanar measurement (grey lines; median and 95% confi dence intervals) and VOCALTM measurements (median; solid black line)

| Ch

apte

r 6

102

Intra-observer reliability: the intraclass correlation coeffi cient for the measurements

using the multiplanar technique was 0.96 (95% confi dence intervals CI 0.74 – 0.98) and

0.97 (95% CI 0.67-0.99) and for the measurements using VOCALTM it was 0.98 (95% CI

0.84-0.995) and 0.97 (95% CI 0.63-0.99) for the two observers respectively.

Inter-observer reliability: the intraclass correlation coeffi cient for the measurements

using the multiplanar technique was 0.97 (95% confi dence intervals CI 0.96 – 0.99) and

for the measurements using VOCALTM it was 0.98 (95% CI 0.97-0.99). Figure 5 displays

the Bland- Altman plots for both methods.

Figure 5. Bland-Altman plots

DISCUSSION

Only few studies have been published about the development of fetal cerebellar

volume obtained by 3D ultrasound in normal pregnancy. These include a cross-

sectional study by Chang on 231 fetuses (1), and longitudinal studies by Araujo Junior

(3) with two measurements each of 55 fetuses and by Hata (2) with a mean of 4.2

(±2.2) measurements each of 13 fetuses. All used polynomial regression analysis

rather than multilevel analysis. Multilevel analysis derives curves which are based

on individual developmental trajectories rather than separate data points and is

therefore eminently suited to the analysis of longitudinal data, whereas traditional

regression analysis presupposes the independence of observations. Multilevel analysis

can also compensate for missing observations and variation of the interval between

| Fet

al c

ereb

ella

r vo

lum

e an

d s

ymm

etry

on

th

ree-

dim

ensi

onal

ult

raso

un

d

103

observation (12). When comparing the results of the other studies with ours (Figure 6)

there are obvious differences such as the lack of data in the late third trimester in Araujo

Junior’s study (3), and the much smaller cerebellar size in the late third trimester in the

studies from Taiwan (1;2). These differences might be related to different statistical

methods, difference in measurement technique, or, as previously suggested, ethnic

differences (13). A recent Dutch study (14) also found larger measurements of the

transverse cerebellar diameter than previously found in studies performed in Britain

(15) and the United States (16). The similarity we found in the growth curves derived

from our data using two different measurement techniques and in different sample

sizes is a reassurance that the technique itself does not make a difference, but that one

should interpret cerebellar volumes with caution when using normal values derived

from a different population or ethnic group.

Figure 6. Comparison of the Utrecht data (grey lines; median and 95% confi dence intervals) with median values from earlier published studies (Araujo, solid black line; Chang, dot-dash black line; Hata, dashed black line)

We did fi nd an increase in the percentage of volume blocks which were not interpretable

at higher gestational ages. This might raise concerns about the clinical applicability of

cerebellar volume measurement in the late third trimester. On the other hand, while

| Ch

apte

r 6

104

we took care to limit the examination time in this group of healthy volunteers, we could

still evaluate more than half of the volume blocks in fetuses older than 32 weeks. In a

patient where it would be important to measure the cerebellum, more volume blocks

could be obtained using different approaches, including transvaginal ultrasonography

(17;18).

Few prenatal studies have compared VOCALTM and multiplanar techniques of volume

assessment. An in vitro study suggested VOCALTM to be more precise than then

multiplanar technique for volume measurement of objects with irregular surfaces

(9). However, in prenatal assessment of lung volume the multiplanar technique was

deemed superior to VOCALTM (10). Satisfactory intra- and inter-observer reliability has

also been demonstrated for measuring fetal cerebellar volume using the multiplanar

technique (1;2). As far as we know the present study is the fi rst one to compare the

multiplanar and VOCALTM technique for fetal cerebellar measurement. Although the

Bland-Altman plots might suggest better reliability of the VOCALTM technique, there

is a high intraclass correlation coeffi cient for different measurements by the same

observer, by different observers and using both techniques. This would lead us to

conclude that either technique could be used to measure fetal cerebellar volume.

Although lateralization of structural (4;19) and functional (20) development of the

fetal brain is undoubtedly interesting, the importance of our fi nding of a degree of

lateral asymmetry in cerebellar volume would be of more importance where there

is concern about unilateral cerebellar atrophy. The latter has been described in the

neonate as the result of extreme prematurity (21;22) and is associated with poor

neurodevelopmental outcome, but has not been described in the fetus.

The next step would be evaluating whether cerebellar volume measurement improves

the accuracy of prenatal ultrasound to diagnose conditions where cerebellar growth

is impaired, such as pontocerebellar dysplasia (23). The advantage of being able to

measure the volume accurately might be offset by the disadvantage that it was not

always possible to measure the volume in late pregnancy. This needs to be evaluated

prospectively.

Our study included only singleton pregnancies. No signifi cant differences in cerebellar

growth between single and multiple pregnancies had been found in a study using two-

dimensional measurements(24), but it would need to be confi rmed that assessment

of growth using volume measurement is also equivalent in single and multiple

pregnancies. We only included patients from 20 weeks’ gestation, but the cerebellum

can be reliably visualized with high-resolution transvaginal ultrasound from the end

| Fet

al c

ereb

ella

r vo

lum

e an

d s

ymm

etry

on

th

ree-

dim

ensi

onal

ult

raso

un

d

105

of the fi rst trimester (25).

In conclusion, we generated longitudinal growth curves for cerebellar volume based

on individual developmental trajectories. The left fetal cerebellar hemisphere was

found to be signifi cantly larger than the right, especially before 32 weeks’ gestation.

There was an acceptable intra- and inter-observer reliability with both multiplanar

and VOCALTM techniques. Either technique could be used for measuring the cerebellar

volume, but volume measurements should be interpreted with caution when using

normal values derived from a different population.

| Ch

apte

r 6

106

REFERENCES

(1) Chang CH, Chang FM, Yu CH, Ko HC, Chen HY. Assessment of fetal cerebellar volume using three-dimensional

ultrasound. Ultrasound Med Biol 2000; 26(6):981-988.

(2) Hata T, Kuno A, Dai S-Y, Inubashiri E, Hanaoka U, Kanenishi K et al. Three-dimensional sonographic volume

measurement of the fetal cerebellum. J Med Ultrasonics 2007; 34(1):17-21.

(3) Araujo JE, Pires CR, Nardozza LM, Filho HA, Moron AF. Correlation of the fetal cerebellar volume with other fetal

growth indices by three-dimensional ultrasound. J Matern Fetal Neonatal Med 2007; 20(8):581-587.

(4) Hering-Hanit R, Achiron R, Lipitz S, Achiron A. Asymmetry of fetal cerebral hemispheres: in utero ultrasound

study. Arch Dis Child Fetal Neonatal Ed 2001; 85(3):F194-F196.

(5) Johnsen SD, Bodensteiner JB, Lotze TE. Frequency and nature of cerebellar injury in the extremely premature

survivor with cerebral palsy. J Child Neurol 2005; 20(1):60-64.

(6) Martins WP, Ferriani RA, Barra DA, dos Reis RM, Bortolieiro MA, Nastri CO et al. Reliability and validity of tissue

volume measurement by three-dimensional ultrasound: an experimental model. Ultrasound Obstet Gynecol

2007; 29(2):210-214.

(7) Martins WP, Ferriani RA, Nastri CO, Filho FM. First trimester fetal volume and crown-rump length: comparison

between singletons and twins conceived by in vitro fertilization. Ultrasound Med Biol 2008; 34(9):1360-1364.

(8) Raine-Fenning NJ, Campbell BK, Clewes JS, Johnson IR. The interobserver reliability of ovarian volume

measurement is improved with three-dimensional ultrasound, but dependent upon technique. Ultrasound Med

Biol 2003; 29(12):1685-1690.

(9) Raine-Fenning NJ, Clewes JS, Kendall NR, Bunkheila AK, Campbell BK, Johnson IR. The interobserver reliability and

validity of volume calculation from three-dimensional ultrasound datasets in the in vitro setting. Ultrasound

Obstet Gynecol 2003; 21(3):283-291.

(10) Kalache KD, Espinoza J, Chaiworapongsa T, Londono J, Schoen ML, Treadwell MC et al. Three-dimensional

ultrasound fetal lung volume measurement: a systematic study comparing the multiplanar method with the

rotational (VOCAL) technique. Ultrasound Obstet Gynecol 2003; 21(2):111-118.

(11) Merz E, Benoit B, Blaas HG, Baba K, Kratochwil A, Nelson T et al. Standardization of three-dimensional images in

obstetrics and gynecology: consensus statement. Ultrasound Obstet Gynecol 2007; 29(6):697-703.

(12) Singer JD, Whitlett JB. Applied longitudinal analysis. 2 ed. New York: Oxford University Press, 2003.

(13) Jacquemyn Y, Sys SU, Verdonk P. Fetal transverse cerebellar diameter in different ethnic groups. J Perinat Med

2000; 28(1):14-19.

(14) Verburg BO, Steegers EA, De Ridder M, Snijders RJ, Smith E, Hofman A et al. New charts for ultrasound dating of

pregnancy and assessment of fetal growth: longitudinal data from a population-based cohort study. Ultrasound

Obstet Gynecol 2008; 31(4):388-396.

(15) Snijders RJ, Nicolaides KH. Fetal biometry at 14-40 weeks’ gestation. Ultrasound Obstet Gynecol 1994; 4(1):34-

48.

| Fet

al c

ereb

ella

r vo

lum

e an

d s

ymm

etry

on

th

ree-

dim

ensi

onal

ult

raso

un

d

107

(16) Goldstein I, Reece EA. Cerebellar growth in normal and growth-restricted fetuses of multiple gestations. Am J

Obstet Gynecol 1995; 173(4):1343-1348.

(17) Malinger G, Ginath S, Lerman-Sagie T, Watemberg N, Lev D, Glezerman M. The fetal cerebellar vermis: normal

development as shown by transvaginal ultrasound. Prenat Diagn 2001; 21(8):687-692.

(18) Zalel Y, Seidman DS, Brand N, Lipitz S, Achiron R. The development of the fetal vermis: an in-utero sonographic

evaluation. Ultrasound Obstet Gynecol 2002; 19(2):136-139.

(19) Chi JG, Dooling EC, Gilles FH. Gyral development of the human brain. Ann Neurol 1977; 1(1):86-93.

(20) de Vries JI, Fong BF. Normal fetal motility: an overview. Ultrasound Obstet Gynecol 2006; 27(6):701-711.

(21) Limperopoulos C, Bassan H, Gauvreau K, Robertson RL, Jr., Sullivan NR, Benson CB et al. Does cerebellar injury in

premature infants contribute to the high prevalence of long-term cognitive, learning, and behavioral disability

in survivors? Pediatrics 2007; 120(3):584-593.

(22) Messerschmidt A, Fuiko R, Prayer D, Brugger PC, Boltshauser E, Zoder G et al. Disrupted cerebellar development

in preterm infants is associated with impaired neurodevelopmental outcome. Eur J Pediatr 2008.

(23) McCann E, Pilling D, Hesseling M, Roberts D, Subhedar N, Sweeney E. Pontomedullary disconnection: fetal and

neonatal considerations. Pediatr Radiol 2005; 35(8):812-814.

(24) Goldstein I, Reece EA, Tamir A. Cerebellar growth in normal fetuses of multiple gestations. J Matern Fetal

Neonatal Med 2002; 11(3):188-191.

(25) Blaas HG, Eik-Nes SH, Kiserud T, Hellevik LR. Early development of the hindbrain: a longitudinal ultrasound study

from 7 to 12 weeks of gestation. Ultrasound Obstet Gynecol 1995; 5(3):151-160.

PATHOLOGY

Chapter 7THE ROLE OF PRENATAL ULTRASOUND IN

PREDICTING SURVIVAL AND MENTAL AND MOTOR

FUNCTIONING IN CHILDREN WITH SPINA BIFIDA

S. van der Vossen1, L.R. Pistorius2, E.J.H. Mulder2, M. Platenkamp1, Ph. Stoutenbeek MD2, G.H.A. Visser2, R.H.J.M. Gooskens1

1Department of Neurology and Neurosurgery, Rudolf Magnus Institute of Neuroscience, Utrecht2Department Obstetrics, Division of Perinatology and Gynecology, University Medical Centre Utrecht,

The Netherlands

Ultrasound Obstet Gynecol 2008 (in press)

| Ch

apte

r 7

112

ABSTRACT

Background

Spina bifi da is one of the most common malformations and is detected effectively

by antenatal ultrasound screening. Once it is detected, well-founded counselling is

imperative.

Objective

To determine which prenatal ultrasound fi ndings can predict survival and mental and

motor functioning in children with spina bifi da.

Methods

Prenatal ultrasound examinations of all live born children who were prenatally di-

agnosed with spina bifi da from 1997 through 2002 (n=41) were retrospectively re-

viewed for lesion level, head circumference, ventriculomegaly, scoliosis and talipes.

These measures were correlated with postnatal anatomical and functional lesion lev-

els, survival and motor and mental outcome at fi ve years of age. The ability of prenatal

ultrasound to determine lesion level was also assessed in all fetuses diagnosed with

spina bifi da from 2006-2007 (n=18).

Results

Nineteen infants died before the age of fi ve years. Multivariate regression analysis

showed that higher lesion level (p < 0.05) and head circumference ≥ 90th percentile

(p = 0.01) were independent predictors of demise. None of the ultrasound features

were predictors of motor or mental functioning. Ultrasound predicted anatomical le-

sion level within 1 level of the postnatal fi ndings in 50% of the fi rst cohort and 89% of

the second cohort (p < 0.01).

Conclusion

Lesion level and head circumference on prenatal ultrasound are predictive for survival

in children with spina bifi da. No predictors were found for mental or motor function

at the age of fi ve years.

| Th

e ro

le o

f p

ren

atal

ult

raso

un

d in

pre

dic

tin

g su

rviv

al a

nd

men

tal a

nd

mot

or f

un

ctio

nin

g in

ch

ildre

n w

ith

sp

ina

bif

ida

113

INTRODUCTION

Spina bifi da is one of the most common and severe structural malformations in hu-

mans. The birth prevalence is between 0.3 and 0.6/1000 in Europe and Northern

America (1) but may be decreasing due to increased use of folic acid (2) or prenatal

diagnosis and termination of affected pregnancies (3).

Routine antenatal ultrasound is an effi cient means of screening for spina bifi da (4)

and a substantial proportion of parents opt for pregnancy termination when faced

with this diagnosis (5;6). Some parents might have chosen this option due to counsel-

ling which places emphasis on severe disability associated with spina bifi da (7). The

ability to reliably predict future functioning of children with spina bifi da, based on

prenatal ultrasound fi ndings, would be valuable to enable parents to make informed

decisions.

Previous studies have explored the role of prenatal imaging in predicting the progno-

sis of children with spina bifi da. These studies showed that lesion level, macrocephaly

and ventriculomegaly were associated with mental and motor outcome in children

with spina bifi da (8-12). However, these studies were based on relatively short follow-

up periods and are of varying methodological quality. We therefore decided to evalu-

ate the ability of prenatal ultrasound fi ndings to predict survival and future mental

and motor functioning in a retrospective cohort of children prenatally diagnosed with

spina bifi da.

METHODS

Patients

The study group consisted of all live born children who were prenatally diagnosed

with spina bifi da from 1997 through 2002 in the University Medical Centre Utrecht, a

tertiary care clinic in the Netherlands (First cohort). Exclusion criteria were insuffi cient

prenatal information, associated autosomal chromosomal disorders or additional

anomalies which could infl uence neurodevelopmental outcome. The gestational age

at diagnosis and the ability of prenatal ultrasound to determine lesion level were also

assessed in all fetuses diagnosed with spina bifi da from 2006-2007 in our institution

(second cohort).

| Ch

apte

r 7

114

Ultrasound

Ultrasound examinations were performed by obstetricians with special training in

antenatal ultrasound examination and with state-of-the art equipment. Ultrasound

data were retrieved from text reports, medical notes and if necessary from reviews

of stored still or moving images by one of the authors (LP) while being unaware of

postnatal data. Items to be scored were the lesion level (the lowest intact vertebral

body) (13), head circumference (HC) and HC percentile for gestational age, width of

the atrium of the lateral ventricle (14) and the ventricle-hemisphere ratio (VHR), sco-

liosis and talipes. The lesion level was defi ned as a high lesion when the lowest intact

vertebral body was at L3 or above and a low lesion if the lowest intact vertebral body

was at L4 or below (12;15). Ventriculomegaly was defi ned as atrial width ≥ 15 mm.

Both unilateral and bilateral involvement was considered as presence of talipes.

If multiple ultrasound examinations were available for one fetus, the examination

closest to the term of 20 weeks of gestation was used.

Follow-up

All children were postnatally evaluated and followed by a multi-disciplinary spina bi-

fi da team. The postnatal course and lesion level as recorded in the clinic notes were

noted for all patients. In case of survival, mental and motor functioning were classi-

fi ed according to the fi ndings during follow-up visits at the age of fi ve years.

Motor functioning was dichotomised to discriminate good motor function (commu-

nity or household ambulators) from poor motor function (non-functional ambulators

or non-ambulators) (16).

Good mental outcome was defi ned as following regular education or an IQ score ≥

80. Poor mental functioning was defi ned as a combination of following special educa-

tion and an IQ-score < 80. Mental functioning was also assessed using the Cognitive

Health Score (CHS) of a Dutch version (17) of the Hydrocephalus Outcome Question-

naire (HOQ) (18). The need for shunt placement or revision for hydrocephalus was

noted.

Statistical analysis

We assessed the ability of prenatal ultrasound to predict survival and the height of the

lesion level in the whole study population, and motor and mental functioning in chil-

dren who survived until the age of fi ve years. We also assessed the ability to predict

motor and mental functioning by means of the postnatally determined anatomical

| Th

e ro

le o

f p

ren

atal

ult

raso

un

d in

pre

dic

tin

g su

rviv

al a

nd

men

tal a

nd

mot

or f

un

ctio

nin

g in

ch

ildre

n w

ith

sp

ina

bif

ida

115

lesion level.

Descriptive analyses were performed using Fisher’s exact test on dichotomised pa-

rameters. To relate survival, mental and motor prognosis to the different parameters

univariate logistic regression analysis was fi rst performed to identify the most im-

portant predictive factors. Items with a p-value < 0.10 in univariate analysis were in-

cluded in a multivariate logistic regression analysis, in which p < 0.05 was considered

signifi cant. The results are expressed as odds ratios (OR) and 95% confi dence intervals

(95% CI). The appropriateness of the cut-off values for lesion level, ventriculomegaly

and ventricle-hemisphere ratio were further investigated with the use of a receiver

operator characteristic curve (ROC) to determine the cut-off point with optimal sensi-

tivity and specifi city.

To relate mental outcome as assessed by the CHS (a continuous variable) to ultra-

sound items, linear regression was used.

The agreement of prenatal and postnatal lesion level assignation was assessed by

computing the difference in prenatal level assignment by ultrasound and postnatal

level assignment by MRI or X-ray imaging.

Data management and statistical analysis were performed with SPSS version 12.0 for

Windows (SPSS Inc. Chicago, Illinois).

RESULTS

Study population

Seventy fetuses were diagnosed with spina bifi da in the study period at a median

gestational age of 29 weeks (range 13-42 weeks). Twenty two pregnancies were ter-

minated. Six children were excluded because of the following: associated syndromal

or chromosomal disorders (n = 3), additional anomalies which infl uenced neurodevel-

opmental outcome (n = 1), insuffi cient prenatal information (n = 1) and parents de-

clining consent for participation in this project (n = 1). One child was lost to follow-up

before fi ve years of age. This led to the fi nal study population of 41 children. Nineteen

of these children died (46%), eighteen of them during the neonatal period and one

at the age of two years due to a symptomatic Chiari II malformation. Information re-

garding motor functioning and type of education could be retrieved for all 22 children

who survived until fi ve years. In three cases the HOQ could not be completed due to a

language barrier or the parents declining to complete the HOQ.

| Ch

apte

r 7

116

Of the fi nal study population 25 were male (61%). The median gestational age at ul-

trasound examination was 35 (range 20-42) weeks in the main cohort, and 20 (range

18–38) weeks in the second cohort. The second cohort comprised 23 children, of

whom the postnatal lesion height was known in 18.

Survival

In univariate analysis a lesion at L3 or higher, a HC ≥ 90th percentile, VHR>0.5 and the

presence of scoliosis were negatively correlated with survival with p<0.10 (Table 1).

In multivariate analysis only HC ≥ 90th percentile (OR 12.6, 95% CI 1.7-91.5, p = 0.01)

and high lesion level (OR 5.8, 95% CI 1.0-32.1, p < 0.05) appeared to be independently

related to demise before 5 years.

Employing a ROC-curve, a cut-off point between T12 and L1 gave the best combina-

tion of sensitivity and specifi city. Using this cut-off point resulted in a stronger rela-

tion between lesion level, HC ≥ 90th percentile and survival. Presence of these two

ultrasound items demonstrated odds ratios of 16.5 (95% CI 2.6-104.0, p = 0.003) and

15.3 (95% CI 1.8-130.8, p = 0.01) for demise for a high lesion level and increased HC

respectively. The chances of survival according to lesion level and HC are depicted in

Table 2.

Motor outcome

Fifteen children (68%) were household or community ambulators and seven were

non-ambulators. Neither the ultrasound markers nor the postnatally determined ana-

tomical lesion level was related to motor outcome at the age of fi ve years in univariate

analysis, regardless of the cut-off point for lesion level (Table 1). As a result multivari-

ate analysis was not performed for motor outcome.

Mental outcome

Thirteen children (59%) were able to follow regular education. Nine children, of whom

six with an IQ score below 80, attended special education. In univariate analysis none

of the ultrasound markers showed a signifi cant relationship with poor mental out-

come (Table 1).

The median CHS in this patient sample was 0.58 (range 0.25-1.00). In univariate analy-

sis only HC ≤ 10th percentile reached a p-value < 0.10 (Table 1). HC ≤ 10th percentile

showed a trend in predicting higher CHS (p = 0.06).

Postnatal anatomical lesion level was not related to mental outcome. However, the

postnatal anatomical lesion level at T12 or higher predicted a lower CHS (p = 0.04). In

fi

| Th

e ro

le o

f p

ren

atal

ult

raso

un

d in

pre

dic

tin

g su

rviv

al a

nd

men

tal a

nd

mot

or f

un

ctio

nin

g in

ch

ildre

n w

ith

sp

ina

bif

ida

117

mu

ltiv

aria

te a

nal

ysis

, nei

ther

th

e p

ostn

atal

an

atom

ical

leve

l nor

a H

C ≤

10t

h p

erce

nti

le c

orre

late

d s

ign

ifi ca

ntl

y w

ith

th

e C

HS.

Tab

le 1

. Res

ult

s of

un

ivar

iate

an

alys

is, r

elat

ing

ult

raso

un

d it

ems

to s

urv

ival

, men

tal a

nd

mot

or o

utc

ome

Dem

ise

bef

ore

5 yr

Poor

mot

or f

un

ctio

nin

gM

enta

l ou

tcom

eSp

ecia

l ed

uca

tion

+ IQ

Men

tal o

utc

ome

CH

S

OR

CI

pO

RC

Ip

OR

CI

pB

CI

p

HC

≥ 9

0th p

erce

nti

le8.

01.

4;44

.90.

022.

30.

1;43

.80.

570.

00.

00.

99-0

.11

-0.4

9;0.

270.

55

HC

≤ 1

0th p

erce

nti

le0.

80.

2;3.

30.

720.

30.

03;3

.60.

360.

40.

04;4

.80.

500.

23-0

.01;

0.48

0.06

VM

3.4

0.6;

19.5

0.17

1.0

0.1;

7.5

1.00

2.5

0.23

;27.

60.

450.

04-0

.25;

0.32

0.79

VH

R ≥

0.50

4.6

1.2;

17.5

0.03

0.6

0.1;

4.2

0.61

6.0

0.78

;46.

10.

09-0

.11

-0.3

5;0.

120.

33

Scol

iosi

s8.

30.

8;83

.20.

07-*

--

-*-

--*

--

Talip

es6.

00.

5;68

.70.

150.

00.

01.

000.

00.

01.

00-*

--

US

lesi

on L

3 or

hig

her

3.1

0.8;

11.8

0.09

0.9

0.1;

5.2

0.87

1.3

0.2;

8.4

0.79

0.02

-0.2

2;0.

250.

89

US

lesi

on T

12 o

r h

igh

er9.

02.

1;38

.80.

003

2.6

0.3;

23.8

0.40

3.5

0.37

;33.

30.

280.

18-0

.09;

0.46

0.18

PN le

sion

L3

or h

igh

er-

--

2.9

0.42

;19.

60.

292.

00.

28;1

4.2

0.49

-0.1

5-0

.37;

0.08

0.20

PN le

sion

T12

or

hig

her

--

-3.

70.

56;2

4.1

0.18

6.0

0.78

;46.

10.

090.

230.

01;0

.45

0.04

HC

: Hea

d c

ircu

mfe

ren

ce; V

M: v

entr

icu

lom

egal

y; V

HR:

Ven

tric

le h

emis

ph

ere

rati

o; U

S: u

ltra

sou

nd

; PN

: pos

tnat

al; C

HS:

cog

nit

ive

hea

lth

sco

re; O

R: o

dd

s ra

tio;

CI:

95%

con

fid

ence

inte

rval

; p: p

-val

ue

* Log

isti

c re

gres

sion

an

alys

is c

ould

not

be

calc

ula

ted

du

e to

sm

all s

amp

le s

ize.

| Ch

apte

r 7

118

Table 2. Survival according to head circumference and lesion height categories.

Head circumference ≥p90 Head circumference <p90

Lesion ≥L3 1/6 (17%) 9/17 (53%)

Lesion ≤L4 1/4 (25%) 11/12 (92%)

Head circumference ≥p90 Head circumference <p90

Lesion ≥T12 1/6 (17%) 5/12 (42%)

Lesion ≤L1 1/4 (25%) 15/17 (88%)

Survivors / total n (%)

Shunting for hydrocephalus

Twelve patients with good (80%) and six with poor motor function (86%) required

shunting for hydrocephalus. Four with good (33%) and two with poor motor function

(33%) required more than one shunt revision. Thirteen patients with good (85%) and

seven patients with poor mental function (78%) required shunting (p=1.0). Five with

good (39%) and two with poor mental function (28%) required more than one shunt

revision (p= 0.32).

<0: prenatal level assignment lower than postnatal level assignment >0: prenatal level assignment higher than postnatal level assignment

Figure 1. Difference between antenatal and postnatal lesion level assignment.

| Th

e ro

le o

f p

ren

atal

ult

raso

un

d in

pre

dic

tin

g su

rviv

al a

nd

men

tal a

nd

mot

or f

un

ctio

nin

g in

ch

ildre

n w

ith

sp

ina

bif

ida

119

Prenatal assignation of lesion level

Both the prenatal and postnatal assessment of lesion level was available in 40 pa-

tients in the main cohort and in 18 patients in the second cohort. The prenatally as-

signed lesion level underestimated the postnatal anatomical lesion level with a mean

of 1.3 (SD 2.4) vertebral levels in the main cohort and with a mean of 0.3 (SD 0.9)

vertebral levels in the second cohort. In 20 patients (50%) of the fi rst cohort and in

in 16 patients (89%) of the second cohort the prenatally assigned level was within 1

level (p<0.01) (Figure 1). Six patients of the second cohort were assessed with three

dimensional (3D) ultrasound, and the prenatally assigned level was within 1 level of

the postnatal anatomical level in all six. Examples of level determination with 2D and

3D equipment is shown in fi gure 2.

Two dimensional ultrasound Three dimensional ultrasound

Figure 2. Ultrasound diagnosis of anatomical level of spina bifi da.

Prenatally assigned lesion level underestimated the postnatal functional lesion level

as assigned by neurological examination with a mean of 0.8 vertebral levels (SD 2.9).

The postnatal anatomical level was a mean of 0.4 levels higher than postnatal func-

tional level (SD 2.3, range four levels lower to six levels higher) (Figure 3).

| Ch

apte

r 7

120

<0: anatomical level lower than functional level >0: anatomical level higher than functional level

Figure 3. Difference between anatomical and functional level assignment.

DISCUSSION

Our study group was chosen to allow for fi ve-year follow-up in all infants. The disad-

vantage is that the ultrasound examinations were done with equipment with lower

resolution than nowadays and without a 3D capability, and at a time when screening

for neural tube defects in the Netherlands was offered to high-risk patients rather

than as routine screening (6). Indeed, there was a distinct improvement in our ability

to predict lesion level in the second cohort using modern ultrasound (and sometimes

3D (19)), equipment and the gestational age at diagnosis has decreased dramatically

since the introduction of routine ultrasound screening in pregnancy.

It is interesting that we found a strong correlation between survival and prenatal pa-

rameters such as lesion height and macrocephaly which have been found to correlate

with motor and mental function in other studies (8-12). One reason for this might be

the difference in management in our unit, where all infants born with spina bifi da are

not routinely scheduled for surgery, but are fi rst evaluated by a multidisciplinary team.

If it is the opinion of the multidisciplinary team that a very poor quality of life is antici-

| Th

e ro

le o

f p

ren

atal

ult

raso

un

d in

pre

dic

tin

g su

rviv

al a

nd

men

tal a

nd

mot

or f

un

ctio

nin

g in

ch

ildre

n w

ith

sp

ina

bif

ida

121

pated despite surgery, and the parents opt against surgery, only palliative care is given

(28). It is conceivable that an infant who otherwise would have survived with poor

motor and mental function after surgery would not survive, with the consequence

that the predictors of poor function (such as high lesion level, ventriculomegaly or

macrocephaly) found in other studies correlated to (neonatal) demise in our group.

Other possible explanations for the differences between our and other studies may

lie in the longer follow-up period, differences in determining or categorizing motor

and mental outcome and our more rigorous statistical analyses. We categorized com-

munity and household ambulation as good motor outcome, as the possibility to walk

even indoors adds to autonomy and quality of life.

We categorized a combination of inability to attend normal education and IQ below

80 (“mild mental retardation”) as poor mental function, because several parents could

not indicate whether their child was in special education due to the physical handicap

alone or because of an additional developmental delay. Furthermore, the CHS was

chosen to assess mental functioning more reliably. Previous studies have shown that

hydrocephalus is an important determinant of cognitive functioning in children with

spina bifi da (21-28). The Dutch translation of this questionnaire has been shown to

have good applicability for determining outcome in hydrocephalic children (17).

We found a trend between the presence of a HC at or below the 10th percentile and

higher CHS scores. The sample size of this study is too small to conclude that a small

HC shows a protective effect on cognitive function, but that the presence of a small

HC in a fetus with spina bifi da is not necessarily a negative fi nding. Also, postnatally

determined high lesion level (T12 or above) showed a relationship to higher CHS but

not to type of education and IQ. Lesion level might possibly be related to mental func-

tioning, but this relation is not unequivocal and we cannot claim to predict future

mental outcome by means of prenatally determined lesion levels.

Suboptimal determination of the lesion level may also play a part in the differences

between our and other studies. However, anatomical lesion level as determined post-

natally also differs from the functional level with as much as six vertebral levels, and

does not predict motor and mental function, either.

The available literature varies regarding the accuracy of antenatal ultrasound to dem-

onstrate the lesion level varies. Six (8;10;12;29-31) out of eight (8;10;12;29-33) studies

concluded that prenatal ultrasound revealed good agreement with postnatal fi ndings.

However, only three studies correlated prenatal lesion level assignation to postnatal

functional lesion level, also with varying accuracy (10;30;31). Intervening factors such

| Ch

apte

r 7

122

as central nervous system infections and shunt complications may also have infl u-

enced neuromotor development (34).

It is also possible that some parameters may indeed be better predictors than shown

by our study. Retrieving ultrasound data from ten-year old video tapes or suboptimal

still images cannot be as accurate as real-time scanning. Lesion level, unless explicitly

stated, as well as the presence or absence of scoliosis or talipes was often diffi cult to

ascertain. Although we tried to minimise the effect of gestational age by using per-

centiles and ratios for analysis, we cannot rule out an effect of gestational age on spe-

cifi c ultrasound parameters (35;36). This may restrict the applicability of the results.

On the other hand, the lesion level would not change during gestation, although it

might be easier to ascertain it at twenty weeks than in the third trimester. To our best

knowledge, this is the fi rst study to investigate the possibility of predicting survival

with prenatal ultrasound fi ndings in children with spina bifi da. We found a lesion level

at L3 or above and the presence of HC ≥ 90th percentile to be independent prognostic

items of demise with high odds ratios, with an even stronger association if a cut-off

level at T12 is used. The wide confi dence intervals might be due to the small sample

size. Since the cut-off point at T12 was determined retrospectively with ROC analysis,

it needs to be evaluated prospectively.

In conclusion, we found prenatal ultrasound predictive of survival, but not of men-

tal or motor function at the age of fi ve years. Prediction of future mental or motor

function from ultrasound parameters during antenatal counselling seems therefore

impossible.

| Th

e ro

le o

f p

ren

atal

ult

raso

un

d in

pre

dic

tin

g su

rviv

al a

nd

men

tal a

nd

mot

or f

un

ctio

nin

g in

ch

ildre

n w

ith

sp

ina

bif

ida

123

REFERENCES

(1) Mitchell LE, Adzick NS, Melchionne J, Pasquariello PS, Sutton LN, Whitehead AS. Spina bifi da. Lancet 2004 Nov

20;364(9448):1885-95.

(2) van der Pal-De Bruin KM, Buitendijk SE, Hirasing RA, Den Ouden AL. Geboorteprevalentie van neuralebuisdefecten

voor en na campagne voor periconceptioneel foliumzuurgebruik.[Birth prevalence of neural tube defects before

and after campaign for periconceptional use of folic acid.]. Ned Tijdschr Geneesk 2000;144(36):1732-6.

(3) Chan A, Robertson EF, Haan EA, Keane RJ, Ranieri E, Carney A. Prevalence of neural tube defects in South Australia,

1966-91: effectiveness and impact of prenatal diagnosis. BMJ 1993 Sep 18;307(6906):703-6.

(4) Norem CT, Schoen EJ, Walton DL, Krieger RC, O’Keefe J, To TT, et al. Routine ultrasonography compared with mater-

nal serum alpha-fetoprotein for neural tube defect screening. Obstet Gynecol 2005 Oct;106(4):747-52.

(5) Garne E, Loane M, Dolk H, De Vigan C, Scarano G, Tucker D, et al. Prenatal diagnosis of severe structural congenital

malformations in Europe. Ultrasound Obstet Gynecol 2005 Jan;25(1):6-11.

(6) Olde Scholtenhuis MA, Cohen-Overbeek TE, Offringa M, Barth PG, Stoutenbeek P, Gooskens RH, et al. Audit of

prenatal and postnatal diagnosis of isolated open spina bifi da in three university hospitals in The Netherlands.

Ultrasound Obstet Gynecol 2003 Jan;21(1):48-52.

(7) Bruner JP, Tulipan N. Tell the truth about spina bifi da. Ultrasound Obstet Gynecol 2004 Nov;24(6):595-6.

(8) Biggio JR, Jr., Owen J, Wenstrom KD, Oakes WJ. Can prenatal ultrasound fi ndings predict ambulatory status in

fetuses with open spina bifi da? Am J Obstet Gynecol 2001 Nov;185(5):1016-20.

(9) Brumfi eld CG, Aronin PA, Cloud GA, Davis RO. Fetal myelomeningocele. Is antenatal ultrasound useful in predicting

neonatal outcome? J Reprod Med 1995 Jan;40(1):26-30.

(10) Coniglio SJ, Anderson SM, Ferguson JE. Functional motor outcome in children with myelomeningocele: correlation

with anatomic level on prenatal ultrasound. Dev Med Child Neurol 1996 Aug;38(8):675-80.

(11) Coniglio SJ, Anderson SM, Ferguson JE. Developmental outcomes of children with myelomeningocele: prenatal

predictors. Am J Obstet Gynecol 1997 Aug;177(2):319-24.

(12) Peralta CF, Bunduki V, Plese JP, Figueiredo EG, Miguelez J, Zugaib M. Association between prenatal sonographic

fi ndings and post-natal outcomes in 30 cases of isolated spina bifi da aperta. Prenat Diagn 2003 Apr;23(4):311-4.

(13) American Spinal Injury Association. International Standards for Neurological Classifi cations of Spinal Cord Injury

(revised). Chicago: American Spinal Injury Association; 2000.

(14) Cardoza JD, Goldstein RB, Filly RA. Exclusion of fetal ventriculomegaly with a single measurement: the width of the

lateral ventricular atrium. Radiology 1988 Dec;169(3):711-4.

(15) Schoenmakers MA, Uiterwaal CS, Gulmans VA, Gooskens RH, Helders PJ. Determinants of functional indepen-

dence and quality of life in children with spina bifi da. Clin Rehabil 2005 Sep;19(6):677-85.

(16) Hoffer MM, Feiwell E, Perry R, Perry J, Bonnett C. Functional ambulation in patients with myelomeningocele. J Bone

Joint Surg Am 1973 Jan;55(1):137-48.

| Ch

apte

r 7

124

(17) Platenkamp M, Hanlo PW, Fischer K, Gooskens RH. Outcome in pediatric hydrocephalus: a comparison between

previously used outcome measures and the hydrocephalus outcome questionnaire. J Neurosurg 2007 Jul;107(1

Suppl):26-31.

(18) Kulkarni A, Rabin D, Drake J. An instrument to measure the health status in children with hydrocephalus: the Hy-

drocephalus Outcome Questionnaire. J Neurosurg: Pediatrics 2004;101(November):134-40.

(19) Pilu G, Ghi T, Carletti A, Segata M, Perolo A, Rizzo N. Three-dimensional ultrasound examination of the fetal central

nervous system. Ultrasound Obstet Gynecol 2007 Aug;30(2):233-45.

(20) Verhagen E, Sauer PJ. The Groningen protocol--euthanasia in severely ill newborns. N Engl J Med 2005 Mar

10;352(10):959-62.

(21) Barf HA, Verhoef M, Jennekens-Schinkel A, Post MW, Gooskens RH, Prevo AJ. Cognitive status of young adults with

spina bifi da. Dev Med Child Neurol 2003 Dec;45(12):813-20.

(22) Barf HA, Verhoef M, Post MW, Jennekens-Schinkel A, Gooskens RH, Mullaart RA, et al. Educational career and pre-

dictors of type of education in young adults with spina bifi da. Int J Rehabil Res 2004 Mar;27(1):45-52.

(23) Brookshire BL, Fletcher JM, Bohan TP, Landry SH, Davidson KC, Francis DJ. Verbal and nonverbal skill discrepancies

in children with hydrocephalus: a fi ve-year longitudinal follow-up. J Pediatr Psychol 1995 Dec;20(6):785-800.

(24) Casari EF, Fantino AG. A longitudinal study of cognitive abilities and achievement status of children with myelome-

ningocele and their relationship with clinical types. Eur J Pediatr Surg 1998 Dec;8 Suppl 1:52-4.

(25) Iddon JL, Morgan DJ, Loveday C, Sahakian BJ, Pickard JD. Neuropsychological profi le of young adults with spina

bifi da with or without hydrocephalus. J Neurol Neurosurg Psychiatry 2004 Aug;75(8):1112-8.

(26) Snow JH, Prince M, Souheaver G, Ashcraft E, Stefans V, Edmonds J. Neuropsychological patterns of adolescents and

young adults with spina bifi da. Arch Clin Neuropsychol 1994 May;9(3):277-87.

(27) Yeates KO, Enrile BG, Loss N, Blumenstein E, Delis DC. Verbal learning and memory in children with myelomenin-

gocele. J Pediatr Psychol 1995 Dec;20(6):801-15.

(28) Wills K. Neuropsychological functioning in children with spina bifi da and/or hydrocephalus. J Clinical Child Psych

1993;22:247-65.

(29) Aaronson OS, Hernanz-Schulman M, Bruner JP, Reed GW, Tulipan NB. Myelomeningocele: prenatal evaluation--

comparison between transabdominal US and MR imaging. Radiology 2003 Jun;227(3):839-43.

(30) Cochrane DD, Wilson RD, Steinbok P, Farquharson DF, Irwin B, Irvine B, et al. Prenatal spinal evaluation and func-

tional outcome of patients born with myelomeningocele: information for improved prenatal counselling and out-

come prediction. Fetal Diagn Ther 1996 May;11(3):159-68.

(31) Kollias SS, Goldstein RB, Cogen PH, Filly RA. Prenatally detected myelomeningoceles: sonographic accuracy in esti-

mation of the spinal level. Radiology 1992 Oct;185(1):109-12.

(32) Appasamy M, Roberts D, Pilling D, Buxton N. Antenatal ultrasound and magnetic resonance imaging in local-

izing the level of lesion in spina bifi da and correlation with postnatal outcome. Ultrasound Obstet Gynecol 2006

May;27(5):530-6.

| Th

e ro

le o

f p

ren

atal

ult

raso

un

d in

pre

dic

tin

g su

rviv

al a

nd

men

tal a

nd

mot

or f

un

ctio

nin

g in

ch

ildre

n w

ith

sp

ina

bif

ida

125

(33) Bruner JP, Tulipan N, Dabrowiak ME, Luker KS, Walters K, Burns P, et al. Upper level of the spina bifi da defect: how

good are we? Ultrasound Obstet Gynecol 2004 Nov;24(6):612-7.

(34) Hunt GM. Open spina bifi da: outcome for a complete cohort treated unselectively and followed into adulthood.

Dev Med Child Neurol 1990 Feb;32(2):108-18.

(35) Babcook CJ, Goldstein RB, Barth RA, Damato NM, Callen PW, Filly RA. Prevalence of ventriculomegaly in association

with myelomeningocele: correlation with gestational age and severity of posterior fossa deformity. Radiology

1994 Mar;190(3):703-7.

(36) Biggio JR, Jr., Wenstrom KD, Owen J. Fetal open spina bifi da: a natural history of disease progression in utero. Pre-

nat Diagn 2004 Apr;24(4):287-9.

Chapter 8MIDDLE CEREBRAL ARTERY PULSATILITY INDEX IN

FETAL CENTRAL NERVOUS SYSTEM ABNORMALITIES

L.R. Pistorius1, R.H.J.M. Gooskens2, Ph. Stoutenbeek1, E.J.H. Mulder1, G.H.A. Visser1

1Department Obstetrics, Division of Perinatology and Gynecology, University Medical Centre Utrecht, The Netherlands

2Department of Neurology and Neurosurgery, Rudolf Magnus Institute of Neuroscience, Utrecht, The Netherlands

Ultrasound Obstet Gynecol (under revision)

| Ch

apte

r 8

128

ABSTRACT

Objectives

A retrospective cohort study was undertaken to evaluate the prognostic value of the

middle cerebral artery pulsatility index (mca PI) in fetuses with central nervous system

abnormalities.

Methods

Antenatal mca PI and ultrasound fi ndings of fetuses with central nervous system

abnormalities evaluated from 2000 through 2006 in a tertiary unit were related to

subsequent neurodevelopmental outcome.

Results

The mca PI was available for 141 fetuses. Median follow up of 21 months was available

for 135 infants (96%). There was a signifi cantly higher risk of neurodevelopmental delay

in infants with an antenatal fi nding of an increased mca PI. This was true for the whole

group, and especially the subgroup of infants with isolated central nervous system

pathology including aqueductal stenosis, but excluding infants with spina bifi da

or multiple abnormalities. An interesting pattern of initially high and subsequently

decreasing mca PI was seen in infants with malformations of cortical development,

microcephaly, agenesis of the corpus callosum and West’s syndrome. A rapid rise of

mca PI was associated with a high risk of demise.

Conclusions

In fetuses with abnormalities of the central nervous system an antenatal fi nding of

increase in mca PI is associated with a increased risk of adverse neurodevelopmental

outcome. This does not apply to fetuses with spina bifi da or multiple abnormalities.

An increased mca PI which subsequently decreased was seen in fetuses that subsequently

were diagnosed with malformations of cortical development, microcephaly, agenesis

of the corpus callosum and West’s syndrome. A rapid rise in mca was seen in fetuses

with a high risk of demise.

| Mid

dle

cere

bra

l art

ery

pu

lsat

ility

ind

ex in

fet

al c

entr

al n

ervo

us

syst

em a

bn

orm

alit

ies

129

INTRODUCTION

Doppler indices of cerebral blood fl ow are used postnatally to diagnose raised

intracranial pressure in infants with hydrocephalus (1). The value of these indices is

less certain in the prenatal period (2;3), and standard fetal neurosonography does

not include these measurements (4). This raises the question whether measuring

middle cerebral artery fl ow is of value in a fetus with pathology of the central nervous

system (CNS). We therefore undertook a retrospective cohort study to determine the

prognostic value of the middle cerebral artery pulsatility index (mca PI) in fetuses with

ultrasound evidence of CNS pathology. To test the hypothesis that an enlargement

of cerebral ventricles above a certain critical size might lead to a decrease in cerebral

blood fl ow, the relationship between ventricular atrial diameter, head circumference

and mca PI was also examined.

METHODS

A complete fetal ultrasound examination including measurement of the mca PI was

performed in patients who were referred to a single tertiary unit with an ultrasound

diagnosis of a fetal central nervous system abnormality. Antenatal and intrapartum

management (including repetition of the examination) and postnatal follow-up were

performed as clinically indicated. Examinations from 2000 until 2006 were included in

the analysis. Abnormalities were classifi ed under spina bifi da, multiple abnormalities

and isolated central nervous system abnormalities. For the analysis, a pulsatility

index value more than two standard deviations above the mean for gestational

age was considered abnormal (5). In fetuses with repeated mca PI measurements,

the pattern was analyzed (with the expected fi ndings either a repeatedly normal

mca PI or an increased mca PI developing in a fetus with initially a normal mca PI).

Major neurodevelopmental delay was defi ned as the presence of cerebral palsy, a

Bayley neurodevelopment score less than two standard deviations below the mean,

deafness or blindness (6). Minor neurodevelopmental delay was defi ned as any other

neurological, psychological or behavioral abnormality of suffi cient severity to warrant

repeated medical referral. Statistical analysis was done with SPSS (version 12.0.1, SPSS

Inc, Chicago). Descriptive statistics of continuous data which are normally distributed

are given as mean (standard deviation), and of continuous data which are not normally

| Ch

apte

r 8

130

distributed as median (range). Doppler fi ndings were correlated with the presence of

neurodevelopmental delay in live born fetuses in a two by two table, and analyzed with

Fisher’s exact test. Pearson’s R was calculated for the correlation between the multiple

of the median (MoM) of the mca PI and the MoM of the head circumference (HC), and

the MoM of the mca PI and the maximum ventricular atrial width, respectively.

RESULTS

A total of 385 mca PI measurements were obtained from 141 fetuses. The gestational

age at diagnosis and delivery, and interval between the last mca PI measurement

and delivery are given in Table 1. Forty seven infants (33%) had a fi nal diagnosis of

a spina bifi da, in 35 fetuses (25%) the fi nal diagnosis confi rmed CNS fi ndings as part

of a complex with multiple abnormalities (including chromosomal abnormalities), 50

infants (26) had a CNS abnormality not related to a spina bifi da or other abnormalities,

and in 9 infants (26) no CNS abnormality or physiological variations of the normal could

be confi rmed postnatally and these were excluded from further analysis. Follow-up

was available in 126 infants (95%). Of these, sixty eight infants survived the neonatal

period and were followed up for a median of 21 months (range 1-84 months).

a.”Increasing” pattern b. “Decreasing” pattern

Figure 1. Repeated mca PI measurements

ff

| Mid

dle

cere

bra

l art

ery

pu

lsat

ility

ind

ex in

fet

al c

entr

al n

ervo

us

syst

em a

bn

orm

alit

ies

131

Tab

le 1

. Ges

tati

onal

age

at

dia

gnos

is, t

ime

inte

rval

fro

m la

st D

opp

ler

mea

sure

men

t u

nti

l del

iver

y, m

ode

of d

eliv

ery

and

su

rviv

al

Tota

l:n

Ges

tati

onal

age

at

dia

gnos

is:

Med

ian

wee

ks(r

ange

)

Ges

tati

onal

age

at

del

iver

y:

Med

ian

wee

ks(r

ange

)

Tim

e fr

om

last

Dop

ple

r m

easu

rem

ent

un

til

del

iver

y:M

edia

n d

ays

(ran

ge)

Mod

e of

del

iver

y:ca

esar

ean

sec

tion

/

vagi

nal

del

iver

yn

Surv

ival

aft

er

neo

nat

al p

erio

d

(los

t to

fol

low

-u

p):

n

Spin

a b

ifi d

a47

32(1

6-40

)39

(21-

42)

10(0

-48)

18 /

29

28

Mu

ltip

le a

bn

orm

alit

ies

3529

(17-

37)

35(1

5-42

)21

(0-1

23)

5 /

307

CN

S ab

nor

mal

itie

s w

ith

out

spin

a b

ifi d

a or

oth

er a

bn

orm

alit

ies

5028

(14-

40)

38(2

1-41

)13

(0-1

16)

18 /

32

27 (3

)

C

orp

us

callo

sum

age

nes

is6

2 /

46

(1)

A

qu

edu

ct s

ten

osis

2812

/ 1

612

M

icro

cep

hal

y3

1 /

23

(2)

M

alf

of c

orti

cal d

evel

opm

ent

31

/ 2

2

H

olop

rose

nce

ph

aly

21

/ 1

2

C

ereb

ellu

m a

nd

pos

t fo

ssa

20

/ 2

0

O

ther

61

/ 5

5

Un

confi

rmed

927

(21-

33)

39(3

6-40

)53

(12-

110)

1 /

89

(3)

Tota

l14

129

(14-

40)

38(1

5-42

)14

(0-1

23)

42 /

99

74 (6

)

| Ch

apte

r 8

132

The median gestational age at diagnosis decreased from a median of 31 weeks (range

16-40) in 2000-2003 to 27 weeks (14-40) in 2005-2006 (Mann-Whitney U test 2-tailed

p=0.011).

A single mca PI measurement was available in 62 fetuses, two measurements were

done in 26 fetuses, and three or more measurements were done in 44 fetuses. The

time interval between the last mca PI measurement and delivery was a median of 14

days (range 0 to 123 days).

In 105 fetuses the mca PI was (and remained) normal, and in 27 the mca PI was

increased to more than 2SD above the mean (Table 2).

Table 2. Middle cerebral artery Doppler measurements and fi ndings

Middle cerebral artery pulsatility index

Normal once

Normal by repetition

Increasing pattern

Decreasing pattern

High once

Total

Spina bifi da 15 30 0 2 0 47

Multiple abnormalities 19 9 1 1 5 35

Isolated CNS abnormalities 18 14 4 9 5 50

Cca 4 1 0 1 0 6

Aqueduct stenosis 11 7 4 3 4 29

Microcephaly 0 1 0 2 0 3

Mcd 1 0 0 1 1 3

Holoprosencephaly 2 0 0 0 0 2

Cerebellum 0 2 0 0 0 2

Other 0 3 0 2 0 5

Total 52 53 5 12 10 132

Percentage of total 39% 40% 4% 9% 8% 100%

CNS = central nervous system; cca=corpus callosum agenesis; mcd = malformation of cortical development

In case of an increased mca PI , two main patterns could be distinguished: a pattern

of increasing PI, and a pattern of decreasing PI. The “increasing” pattern (Figure 1 a.)

was seen in 5 patients (5/70 or 7% of fetuses with repeated Doppler measurements),

and occurred signifi cantly more frequently in fetuses with aqueduct stenosis (4/14 of

repeated Doppler measurements in fetuses with aqueduct stenosis; 1/56 of all other

fetuses; p=0.0048). The median gestational age at delivery in infants with repeatedly

normal Doppler measurements was 38 weeks (21-42 weeks) and in infants with an

“increasing” pattern was 36 weeks 35-40 weeks) (difference not signifi cant; Mann-

Whitney U test p=0.32). The time until delivery in patients with aqueduct stenosis and

| Mid

dle

cere

bra

l art

ery

pu

lsat

ility

ind

ex in

fet

al c

entr

al n

ervo

us

syst

em a

bn

orm

alit

ies

133

f

an “increasing” pattern was a median of 9.5 days (range 4-16 days); and in the other

patients with repeated Doppler measurements it was 14.5 (1-52) days (Mann-Whitney

U test p=0.24). Three out of fi ve infants with an “increasing” pattern died intrapartum

or postnatally (Table 3).

Table 3. Abnormalities and outcome of foetuses with abnormal mca PI

Pattern Patient Findings and outcome

Increasing pattern (Figure 1a)

1 Microcephaly, akinesia, early neonatal death

2 Severe hydrocephalus, intrapartum demise after cephalocentesis

3 Severe progressive hydrocephalus, early neonatal death after abstaining from treatment

4 Hydrocephalus, delayed speech and motoric development at 18 months

5 Hydrocephalus, hemiplegia, syndrome of West at 18 months

Decreasing pattern (Figure 1b)

1 Lumbosacral spina bifi da, hydrocephalus, motoric impairment lower limbs at 12 months

2 Lumbosacral spina bifi da, motoric impairment lower limbs at 30 months

3 Congenital CMV infection, moderate neurodevelopmental impairment at 13 months

4 Corpus callosum agenesis; delayed motoric development at 15 months

5 Unilateral ventriculomegaly; normal development at 14 months

6 Ventriculomegaly, syndrome of West at 30 months

7 Hydrocephalus, lissencephaly, probably Walker-Warburg syndrome, intrapartum demise

8 Microcephaly, lost to follow-up after age 4 months

9 Microcephaly chorioretinal dysplasia at 16 months

10 Lissencephaly and corpus callosum agenesis, global neurological delay at 13 months

11 Encephalocoele, corpus callosum atrophy, nodular heterotopia, epilepsy at 12 months

12 Arachnoid cyst, syndrome of West at 6 months

The “decreasing” pattern (Figure 1 b.) of the PI mca was seen in twelve infants, eight

of whom subsequently manifested abnormalities related to malformations of cortical

development, microcephaly, corpus callosum agenesis or syndrome of West (Table 3).

No signifi cant correlation was found between the MoM mca PI and maximum

ventricular atrial diameter (Pearson R=0.011, P=0.92, n=76). A weak positive correlation

was found between the MoM mca PI and MoM head circumference (Pearson R=0.22,

P=0.010, n=134).

| Ch

apte

r 8

134

The distribution of Doppler fi ndings associated with different neurological diagnoses

is displayed in Table 2. The relationship between Doppler fi ndings and outcome is

displayed in Table 4.

Table 4. Pattern of middle cerebral artery Doppler abnormalities and outcome at follow-up

Outcome Middle cerebral artery pulsatility index

Normal once

Normal by repetition

Increasing pattern

Decreasing pattern

High once

Total Percentage of total

Unknown 3 2 0 1 0 6 5%

TOP / IUD 22 9 0 0 6 37 28%

Neonatal / infant death 12 10 3 1 1 27 20%

Alive & well 5 5 0 1 0 11 8%

Mild neuro 5 17 0 2 1 25 19%

Severe neuro 5 10 2 7 2 26 20%

Middle cerebral artery pulsatility index

Always normal Ever abnormal Total Percentage of total

Unknown 5 1 6 5%

TOP / IUD 31 6 37 28%

Neonatal / infant death 22 5 27 20%

Alive & well 10 1 11 8%

Mild neuro 22 3 25 19%

Severe neuro 15 11 26 20%

TOP=termination of pregnancy; IUD=intra-uterine or intrapartum demise; neuro=neurodevelopmental delay

In the whole group there was a signifi cant association between adverse mca PI fi ndings

antenatally and subsequent adverse neurodevelopmental outcome (p<0.05). In fetuses

with a spina bifi da or multiple anomalies, there was no association between the PI mca

and outcome. In fetuses with CNS abnormalities not associated with a spina bifi da or

other abnormalities, there was also a signifi cantly higher risk of adverse outcome if

an increased mca PI was found (p<0.05). In fetuses with a fi nal diagnosis of aqueduct

stenosis, the increase in adverse outcome in case of Doppler abnormalities was on the

borderline of statistical signifi cance (Fisher’s exact p=0.072) (Table 5).

No correlation could be found between head circumference (expressed as multiples

of the median) and outcome in the study group or any subgroup. No signifi cant

association could be detected between the maximum diameter of the ventricular

atrium and outcome either.

| Mid

dle

cere

bra

l art

ery

pu

lsat

ility

ind

ex in

fet

al c

entr

al n

ervo

us

syst

em a

bn

orm

alit

ies

135

f

Table 5. Correlation between middle cerebral artery pulsatility index and neurodevelopmental outcome

Doppler always normal Doppler ever highA B

Outcome normal moderate severe normal moderate severe

Study 11 20 16 1 3 11 0.26 0.007**

Spina 2 19 5 0 2 0 1.0 1.0

Multiple 0 0 5 0 0 2 # #

Isolated 9 1 6 1 1 9 0.04* 0.047*

Aqueduct 5 0 3 0 1 3 0.26 0.6

Study=study group (all known follow-up, confi rmed central nervous system abnormalities, n=62)Spina = spina bifi daMultiple = multiple abnormalitiesIsolated = isolated CNS abnormalitiesAqueduct = aqueduct stenosisNormal = normal outcomeModerate = moderate neurological delaySevere = Severe neurodevelopmental delayA=Fisher’s exact test (normal versus moderate or severe)B= Fisher’s exact test (normal or moderate versus severe) * = signifi cant (p<0.05)** = signifi cant (p<0.01)# = impossible to calculate

DISCUSSION

The results from this study demonstrate an association between increased pulsatility

index of the middle cerebral artery and neurodevelopmental outcome in fetuses with

central nervous system abnormalities. Our fi ndings are consistent with neonatal

fi ndings (1;7), as well as previous fi ndings of a progressive increase in pulsatility index of

the internal carotid artery in four fetuses with increasing hydrocephalus (8). However,

these studies did not try to determine an association between Doppler fi ndings and

neurodevelopmental outcome.

The increase in mca PI might refl ect an increase in intracranial pressure, with a resulting

decrease in cerebral perfusion pressure and decrease in diastolic fl ow (9). This was,

indeed, the working hypothesis, and an increase in mca PI acted as an indication for

earlier delivery. This was confi rmed by the shorter (although not statistically signifi cant)

interval from the last Doppler measurement until delivery in these patients. Despite

the earlier delivery, the outcome was still worse than in the group of patients with

normal mca PI measurements. It might be that the interval to delivery was still too long,

exposing the fetal cortex to increased pressure for too long. The increase in pulsatility

| Ch

apte

r 8

136

index might also be too late a fi nding, and another index such as the transsystolic

time (9), or maximum systolic fl ow velocity (vmax) might be more sensitive to detect

impaired cerebral fl ow. It might be also be more appropriate to assess cerebral venous

fl ow patterns (10).

Previously, no correlation between the middle cerebral artery resistance index and the

lateral ventricular width could be demonstrated in 430 normal fetuses, three fetuses

with ventriculomegaly and three with hydrocephalus (11). In our large group of fetuses

with central nervous system pathology, there was also no correlation between mca

PI and ventricular diameter. Only a very weak correlation between mca PI and head

circumference could be found in cases with progressive hydrocephalus. This absence

of a strong correlation might be related to the capability of the fetal skull to expand

with increasing ventricular diameter without an increase in intracranial pressure great

enough to infl uence cerebral blood fl ow, or be related to the development of ex vacuo

hydrocephalus where the intraventricular and intracranial pressure would be expected

to be normal.

The fact that no association could be demonstrated between antenatal mca PI

measurements and outcome in fetuses with spina bifi da might be related to the

fact that the ventricular diameter is of less importance to the neurodevelopmental

development than the spinal abnormality. It might also be that increased intracranial

pressure would not develop to the same extent, as cerebrospinal fl uid could fl ow into

a meningocoele in case of increased intracranial pressure.

An unexpected fi nding was the “decreasing” pattern. The initially increased mca PI

in the fetus with asymmetrical ventriculomegaly and normal neurodevelopmental

outcome at 14 months might refl ect a false positive fi nding, or it might be too early

to detect more subtle neurodevelopmental delay. It could also have been related to

an intraventricular bleeding, as this can be diffi cult to detect on ultrasound (12). An

increase of the mca PI has previously been described in fetuses with intraventricular

hemorrhage (13;14). It is also interesting that the highest mca PI values were seen in

the fetus with congenital CMV infection, as cerebral vasculitis has been associated

with congenital CMV infection (15). The other eight cases with a “decreasing” pattern

were associated with malformations of cortical development (n=4 , two of which were

also associated with corpus callosum agenesis), microcephaly (n=2), West syndrome

(n=2) or corpus callosum agenesis with impaired neurological development (n=1).

Conversely, all infants with syndrome of West or malformations of cortical development

demonstrated an increased mca PI at some time antenatally.

| Mid

dle

cere

bra

l art

ery

pu

lsat

ility

ind

ex in

fet

al c

entr

al n

ervo

us

syst

em a

bn

orm

alit

ies

137

If the pressure hypothesis holds true for the patients with a normalizing pattern,

the law of Laplace (with a pronounced decrease in pressure associated with a small

increase in diameter) might play a part to allow a decrease in intracranial pressure

with an increase in head circumference. Another explanation would be a postulated

transient phenomenon which might lead to subsequent structural and functional

neurological abnormalities. If brain development is faulty, it is also conceivable that

the development of small vasculature is slower, with a cerebral vascular resistance

that is still high at relatively early gestational age, which would be associated with a

raised mca PI. Delayed development of the medullar venous system has been noted

in infants with malformations of cortical development (16), and it is conceivable that

this can be associated with a delayed development on the arterial side of the cerebral

circulation as well.

The relatively advanced gestational age at diagnosis is one of the limitations of this

study. The age at diagnosis did decrease somewhat during the course of the study,

which is probably related to the introduction of a structural ultrasound examination

at 20 gestational weeks for all pregnant women in the Netherlands.

During future prospective evaluation it would be useful to assess the value of other

indices such as the transsystolic time and maximum fl ow velocity in addition to the

pulsatility index of the middle cerebral artery, starting earlier in pregnancy. The use of

venous Doppler measurements, and the use of MRI, especially to assess the presence

or timing of hemorrhage or ischemia (17) and normality of oxygenation (18) should

also be assessed.

CONCLUSIONS

In fetuses with isolated abnormalities of the central nervous system (including fetuses

with aqueduct stenosis), the antenatal fi nding of increase in mca PI is associated with

a subsequently increased risk of demise or adverse neurodevelopmental outcome.

An increased mca PI which subsequently normalized was seen in fetuses that

subsequently were diagnosed with malformations of cortical development,

microcephaly and syndrome of West.

In fetuses with neural tube defects or multiple abnormalities, an increased mca PI was

not demonstrated with an increased risk of adverse neurodevelopmental outcome.

Future research should focus on elucidating mechanisms which could explain this

| Ch

apte

r 8

138

fi nding, and on determining the clinical applicability of measurement of cerebral fl ow

in fetuses with aqueduct stenosis or other central nervous system abnormalities.

| Mid

dle

cere

bra

l art

ery

pu

lsat

ility

ind

ex in

fet

al c

entr

al n

ervo

us

syst

em a

bn

orm

alit

ies

139

REFERENCES

(1) Hanlo PW, Gooskens RH, Nijhuis IJ, Faber JA, Peters RJ, van Huffelen AC et al. Value of transcranial Doppler

indices in predicting raised ICP in infantile hydrocephalus. A study with review of the literature. Childs Nerv Syst

1995; 11(10):595-603.

(2) Kirkinen P, Muller R, Baumann H, Briner J, Lang W, Huch R et al. Cerebral blood fl ow velocity waveforms in

hydrocephalic fetuses. J Clin Ultrasound 1988; 16(7):493-498.

(3) Mai R, Rempen A, Kristen P. Color fl ow mapping of the middle cerebral artery in 23 hydrocephalic fetuses. Arch

Gynecol Obstet 1995; 256(3):155-158.

(4) Malinger G, Monteagudo A, Pilu G, Timor-Tritsch IE, Toi A. Sonographic examination of the fetal central nervous

system: guidelines for performing the ‚basic examination‘ and the ‚fetal neurosonogram‘. Ultrasound Obstet

Gynecol 2007; 29(1):109-116.

(5) Wladimiroff JW. Fetal cerebral blood fl ow. Clin Obstet Gynecol 1989; 32(4):710-718.

(6) Washburn LK, Dillard RG, Goldstein DJ, Klinepeter KL, deRegnier RA, O‘Shea TM. Survival and major

neurodevelopmental impairment in extremely low gestational age newborns born 1990-2000: a retrospective

cohort study. BMC Pediatr 2007; 7:20.

(7) Hill A, Volpe JJ. Decrease in pulsatile fl ow in the anterior cerebral arteries in infantile hydrocephalus. Pediatrics

1982; 69(1):4-7.

(8) Degani S, Lewinsky R, Shapiro I, Sharf M. Decrease in pulsatile fl ow in the internal carotid artery in fetal

hydrocephalus. Br J Obstet Gynaecol 1988; 95(2):138-141.

(9) Hanlo PW, Peters RJ, Gooskens RH, Heethaar RM, Keunen RW, van Huffelen AC et al. Monitoring intracranial

dynamics by transcranial Doppler--a new Doppler index: trans systolic time. Ultrasound Med Biol 1995;

21(5):613-621.

(10) Pooh RK, Pooh KH, Nakagawa Y, Maeda K, Fukui R, Aono T. Transvaginal Doppler assessment of fetal intracranial

venous fl ow. Obstet Gynecol 1999; 93(5 Pt 1):697-701.

(11) Zalel Y, Almog B, Seidman DS, Achiron R, Lidor A, Gamzu R. The resistance index in the fetal middle cerebral

artery by gestational age and ventricle size in a normal population. Obstet Gynecol 2002; 100(6):1203-1207.

(12) Levine D, Barnes PD, Robertson RR, Wong G, Mehta TS. Fast MR imaging of fetal central nervous system

abnormalities. Radiology 2003; 229(1):51-61.

(13) Achiron R, Pinchas OH, Reichman B, Heyman Z, Schimmel M, Eidelman A et al. Fetal intracranial haemorrhage:

clinical signifi cance of in utero ultrasonographic diagnosis. Br J Obstet Gynaecol 1993; 100(11):995-999.

(14) Hadi HA, Finley J, Mallette JQ, Strickland D. Prenatal diagnosis of cerebellar hemorrhage: medicolegal

implications. Am J Obstet Gynecol 1994; 170(5 Pt 1):1392-1395.

(15) Booss J, Dann PR, Winkler SR, Griffi th BP, Kim JH. Mechanisms of injury to the central nervous system following

experimental cytomegalovirus infection. Am J Otolaryngol 1990; 11(5):313-317.

| Ch

apte

r 8

140

(16) Pooh RK, Pooh KH. Medullary venous system of fetal brain; high-frequency transvaginal bidirectional power

Doppler study. Ultrasound Obstet Gynecol 2007; 30(4):416.

(17) Brugger PC, Stuhr F, Lindner C, Prayer D. Methods of fetal MR: beyond T2-weighted imaging. Eur J Radiol 2006;

57(2):172-181.

(18) Levine D. Science to practice: can MR oxygenation imaging be used to assess at-risk pregnancies? Radiology

2006; 238(3):765-766.

Chapter 9DISTURBANCE OF CEREBRAL MIGRATION

FOLLOWING CONGENITAL PARVOVIRUS

B19 INFECTION

L.R. Pistorius1, J. Smal2, T.R. de Haan3, G.C.M.L. Page-Christiaens1, M. Verboon-Maciolek2, D. Oepkes4, L.S. de Vries2

1Department of Obstetrics, University Medical Center, Utrecht, The Netherlands, 2Department of Neonatology, University Medical Center Utrecht, The Netherlands

3Department of Neonatology, Academic Medical Center, Amsterdam, The Netherlands4Department of Obstetrics, Leiden University Medical Center, Leiden, The Netherlands

Fetal Diagn Ther 2008 (in press)

| Ch

apte

r 9

142

ABSTRACT

Objective

We describe the clinical course of an infant who presented with severe fetal anemia and

fetal hydrops following congenital Parvovirus B19 infection before sixteen gestational

weeks. The fetus was treated by cordocentesis and intra-uterine transfusion at

eighteen weeks.

Results

The infant demonstrated mild unilateral ventriculomegaly on antenatal magnetic

resonance imaging, and polymicrogyria and heterotopia on postnatal magnetic

resonance imaging.

Conclusion

This adds to the evidence in recent literature of central nervous system damage

associated with congenital Parvovirus B19 infection.

| Dis

turb

ance

of

cere

bra

l mig

rati

on f

ollo

win

g co

nge

nit

al P

arvo

viru

s B

19 in

fect

ion

143

INTRODUCTION

Intra-uterine infection with Parvovirus B19 can cause severe fetal anemia, fetal hydrops

and intrauterine fetal death. The anemia is secondary to a transient aplastic crisis (1)

and can be effectively treated by intra-uterine transfusion. More than 80% of hydropic

infants survive after intra-uterine transfusion (2-5), whereas untreated fetal hydrops

may lead to fetal death in 30 – 50% (2;3;5;6) of cases or a maternal pre-eclampsia-like

or “mirror syndrome” (7).

Previous studies reported a normal neurodevelopmental prognosis following intra-

uterine Parvovirus B19 infection (8). More recently, mild to severe neurodevelopmental

delay was reported in fi ve out of sixteen Parvovirus B19 hydrops survivors (9).

We describe the course of an infant who developed intrauterine hydrops and severe

anemia due to Parvovirus B19 infection and in whom malformations of cortical

development were seen on cerebral MRI.

CASE REPORT

A genetic amniocentesis was performed for advanced maternal age at a gestational

age of 16 weeks in a 38-year-old G3P2 Caucasian female in the University Medical

Centre Utrecht (UMCU). At that time the patient mentioned to have had a Parvovirus

B19 contact within the family a couple of weeks before. The presence of parvovirus B19

specifi c IgM antibodies suggested recent infection. A Parvovirus B19 PCR performed on

an amniotic fl uid sample proved positive. The infant had a normal female karyotype.

The pregnancy was followed up with weekly ultrasound examinations for hydrops

and middle cerebral artery peak systolic velocity (MCA PSV) measurements were

performed to detect fetal anemia (10). At a gestational age of 17 weeks, the MCA PSV

of the middle cerebral artery was 32cm/s, equivalent to 1.4 multiples of the median

(MoM). Seven days later, fetal ascites was visible on ultrasound, and the MCA PSV

of the middle cerebral artery was 55cm/s, equivalent to 2.3 MoM. Because of the

high probability of fetal anemia, she was referred to the Leiden University Medical

Centre (LUMC) for a diagnostic cordocentesis and intra-uterine transfusion. This

was performed the next day, at a gestational age of eighteen weeks. Cordocentesis

demonstrated a fetal hemoglobin level of 1.4 g/dL (0.9 mmol/L) with a hematocrit of

5%. The platelet count was 39x109/L. Intrauterine erythrocyte and platelet transfusions

| Ch

apte

r 9

144

were given intravenously. The post-transfusion fetal hemoglobin levels were 12.3 g/dL

(7.5 mmol/L), the hematocrit 35%, and the platelet count 122 x109/L. Fetal Parvovirus

B19 infection was confi rmed by detection of viral DNA in fetal blood, with a viral

load of log 8.6 (4.0 x 108) copies/mL. The ascites resolved within two weeks after the

transfusion. The MCA PSV remained normal for the duration of the pregnancy. At 22

weeks, ultrasound examination did not demonstrate any intracranial abnormalities,

but an antenatal MRI examination demonstrated mild unilateral dilatation of the

frontal and occipital horn of the left lateral ventricle (Figure 1) with the ventriclular

atrium measuring 10mm (11).

Figure 1. Prenatal MRI at 22 weeks of gestational age demonstrating mild unilateral dilatation of the occipital and frontal horn of the left lateral ventricle (arrows)

A female infant was delivered vaginally after an induction of labor at a gestational age

of 41 weeks. The Apgar scores were 8 and 10 at 1 and 5 minutes respectively and the

physical examination was normal. The birth weight was 3475g (P50). The hemoglobin

level was 20.5g/dL (12.4mmol/L), the hematocrit 63% and the platelet count 190x109/L.

The liver function tests were normal. The Parvovirus B19 DNA load in blood was

negative. Parvovirus B19 specifi c IgG antibodies were detected in serum of the infant

whereas parvovirus B19 specifi c IgM antibodies were negative. Postnatal cranial

ultrasound demonstrated a slight enlargement of the anterior horn of the left lateral

ventricle, and MRI examination demonstrated a mildly enlarged left frontal horn and

marked ipsilateral frontal polymicrogyria and heterotopia. There was loss of volume of

| Dis

turb

ance

of

cere

bra

l mig

rati

on f

ollo

win

g co

nge

nit

al P

arvo

viru

s B

19 in

fect

ion

145

the left posterior region without evidence of polymicrogyria (Figure 2) .

At the age of twelve months the child showed a normal early development, with

growth and head circumference on the 50th centile.

Figure 2. Postnatal MRI demonstrating mildly enlarged left frontal horn (white line arrow), marked ipsilateral frontal polymicrogyria (solid arrow) and heterotopia (black line arrow) and loss of volume of the left posterior region

DISCUSSION

We describe the outcome of an infant who recovered from non-immune fetal hydrops

due to intrauterine parvovirus B19 infection after an intrauterine transfusion (IUT)

for severe fetal anemia and thrombocytopenia. Neonatal imaging demonstrated

cerebral migratory abnormalities, namely cortical heterotopia and polymicrogyria.

Cortical heterotopia are clusters of neurons in an abnormal position as the result of

arrested migration of neurons which normally occurs from the germinal matrix in the

lateral ventricular walls towards the cortex, and is seen as an abnormally situated T2

hypointese area on MRI. Polymicrogyria is seen as an area of cortex with abnormally

small convolutions (11). Cerebral migratory abnormalities can be associated with

multiple congenital abnormalities, mental retardation syndromes and epilepsy. It has

| Ch

apte

r 9

146

been described in association with ischaemia, Cytomegaloviral (CMV) infection and

complications of monochorionic twin pregnancies occuring before 20 gestational

weeks (12).

The primary target of parvovirus B19 infection is the erythroid cell line but

thrombocytopenia is also often observed in infected patients. Furthermore an

association has been reported between human parvovirus B19 infection and

encephalitis or even fetal stroke (13-17). Myocarditis due to human parvovirus B19 has

been described in the human fetus, infants, older children and even adults (18;19).

Until recently, the available published literature of infants treated with IUT for

anemia related to maternal parvovirus B19 infection was reassuring with regards to

neurodevelopmental outcome. Four publications on a total of 24 survivors of IUT for

parvovirus B19 infection reported a normal neurodevelopmental outcome (6;8;20;21).

However, the upper 95% confi dence limit of absence of neurological abnormalities in

a series of 24 is 3/24 = 12.5% (22) and indeed, rare cases of fetal neurological injury

associated with maternal parvovirus B19 infection have been reported. These include

hydrocephalus (16), cortical dysplasia (16), perivascular white matter calcifi cations

(15), fetal stroke (13;14), cerebellar haemorrhage (29) and encephalitis (17). A very

recent report mentioned for the fi rst time neurodevelopmental delay in fi ve out of

sixteen survivors of IUT for Parvovirus B19 induced fetal anemia. Two of these had

severe neurodevelopmental delay. MRI was not performed in any of these children (9).

The infant in our report is still young, and although she is doing well at twelve months,

she is still at risk of later development of epilepsy in view of the cortical dysplasia (24).

It is diffi cult to quantify the risk of adverse neurorological outcome, as most cases of

cortical dysplasia are found in children where brain imaging is performed because of

intractable epilepsy. The outcome does tend to correlate with the location and extent

of the lesions, and possibly with the underlying cause. Most cases of severe epilepsy

tend to start early, and 38% of mild epilepsy also starts before the age of twelve

months (25).

Although the infant presented with a very low hemoglobin level at the time of

intra-uterine transfusion, no relation between the minimum fetal hemoglobin and

subsequent neurodevelopmental status has been demonstrated so far (8;9). This

might refl ect absence of evidence rather than evidence of absence of an effect, as

the number of reported cases is still only a handful. Detailed follow up of transfused

children is paramount to answer this question.

Can we learn something from other intrauterine viral infections? Congenital

| Dis

turb

ance

of

cere

bra

l mig

rati

on f

ollo

win

g co

nge

nit

al P

arvo

viru

s B

19 in

fect

ion

147

cytomegalovirus infections in the early second trimester can be associated with

cerebral abnormalities such as polymicrogyria in the absence of extreme fetal anemia.

This supports the hypothesis that viral infections as such may have negative effects

on fetal neuronal migration (26-29). The pathogenesis might be related to a disturbed

blood supply during a crucial time of cerebral development (28), through direct

cytotoxic effects of viral infection (15) or by a chronic hypoxic state caused by fetal

vasculitis, severe anemia or myocarditis (30). Parvovirus B19 genome DNA has been

detected in the nuclei of multinucleated giant cells in the white matter of a fetal brain

at post-mortem investigation subsequent to second trimester Parvovirus B19 infection

(15). Neuronal migration has also been shown to be delayed by intermittent or chronic

hypoxic episodes in the infant (30). The previous case of cortical dysplasia associated

with intrauterine Parvovirus B19 infection was also unconfi rmed by histology or PCR

evidence of viral DNA (16) Whether the injury in our patient was caused by a direct

viral effect, by hemodynamic changes caused by fetal myocarditis or anemia, by

the maternal immune response, or by complications of the cordocentesis, remains

uncertain.

If a patient is scheduled for an intrauterine transfusion for Parvovirus B19 induced

fetal anemia, she should be counseled on the risk of a possible association with

neurodevelopmental delay due to the congenital Parvovirus B19 infection despite

successful treatment of fetal anemia and regardless of the severity of the anemia (9).

We therefore propose routine postnatal brain imaging with ultrasound and MRI as

well as subsequent neurodevelopmental follow-up (15) in survivors of parvovirus B19-

associated hydrops, similar to the recommendation in preterm infants (24) and infants

exposed to cytomegalovirus in utero (31;32). This proposal stands regardless of the

initial fetal hemoglobin levels or subsequent duration of hydrops.

| Ch

apte

r 9

148

REFERENCES

(1) Anderson MJ, Higgins PG, Davis LR, Willman JS, Jones SE, Kidd IM et al. Experimental parvoviral infection in

humans. J Infect Dis 1985; 152(2):257-265.

(2) Enders M, Weidner A, Zoellner I, Searle K, Enders G. Fetal morbidity and mortality after acute human parvovirus

B19 infection in pregnancy: prospective evaluation of 1018 cases. Prenat Diagn 2004; 24(7):513-518.

(3) Fairley CK, Smoleniec JS, Caul OE, Miller E. Observational study of effect of intrauterine transfusions on outcome

of fetal hydrops after parvovirus B19 infection. Lancet 1995; 346(8986):1335-1337.

(4) Rodis JF, Borgida AF, Wilson M, Egan JF, Leo MV, Odibo AO et al. Management of parvovirus infection in pregnancy

and outcomes of hydrops: a survey of members of the Society of Perinatal Obstetricians. Am J Obstet Gynecol

1998; 179(4):985-988.

(5) Schild RL, Bald R, Plath H, Eis-Hubinger AM, Enders G, Hansmann M. Intrauterine management of fetal parvovirus

B19 infection. Ultrasound Obstet Gynecol 1999; 13(3):161-166.

(6) Rodis JF, Rodner C, Hansen AA, Borgida AF, Deoliveira I, Shulman RS. Long-term outcome of children following

maternal human parvovirus B19 infection. Obstet Gynecol 1998; 91(1):125-128.

(7) de Jong EP, de Haan TR, Kroes AC, Beersma MF, Oepkes D, Walther FJ. Parvovirus B19 infection in pregnancy. J Clin

Virol 2006; 36(1):1-7.

(8) Dembinski J, Haverkamp F, Maara H, Hansmann M, Eis-Hubinger AM, Bartmann P. Neurodevelopmental outcome

after intrauterine red cell transfusion for parvovirus B19-induced fetal hydrops. BJOG 2002; 109(11):1232-

1234.

(9) Nagel HT, de Haan TR, Vandenbussche FP, Oepkes D, Walther FJ. Long-term outcome after fetal transfusion for

hydrops associated with parvovirus B19 infection. Obstet Gynecol 2007; 109(1):42-47.

(10) Mari G, Deter RL, Carpenter RL, Rahman F, Zimmerman R, Moise KJ, Jr. et al. Noninvasive diagnosis by Doppler

ultrasonography of fetal anemia due to maternal red-cell alloimmunization. Collaborative Group for Doppler

Assessment of the Blood Velocity in Anemic Fetuses. N Engl J Med 2000; 342(1):9-14.

(11) Garel C. MRI of the fetal brain. Berlin: Springer, 2004.

(12) Barkovich AJ, Kuzniecky RI, Jackson GD, Guerrini R, Dobyns WB. A developmental and genetic classifi cation for

malformations of cortical development. Neurology 2005; 65(12):1873-1887.

(13) Craze JL, Salisbury AJ, Pike MG. Prenatal stroke associated with maternal parvovirus infection. Dev Med Child

Neurol 1996; 38(1):84-85.

(14) de Haan TR, Wezel-Meijler G, Beersma MF, Von Lindern JS, Van Duinen SG, Walther FJ. Fetal stroke and congenital

parvovirus B19 infection complicated by activated protein C resistance. Acta Paediatr 2006; 95(7):863-867.

(15) Isumi H, Nunoue T, Nishida A, Takashima S. Fetal brain infection with human parvovirus B19. Pediatr Neurol

1999; 21(3):661-663.

(16) Katz VL, McCoy MC, Kuller JA, Hansen WF. An association between fetal parvovirus B19 infection and fetal

anomalies: a report of two cases. Am J Perinatol 1996; 13(1):43-45.

| Dis

turb

ance

of

cere

bra

l mig

rati

on f

ollo

win

g co

nge

nit

al P

arvo

viru

s B

19 in

fect

ion

149

(17) Kerr JR, Barah F, Chiswick ML, McDonnell GV, Smith J, Chapman MD et al. Evidence for the role of demyelination,

HLA-DR alleles, and cytokines in the pathogenesis of parvovirus B19 meningoencephalitis and its sequelae. J

Neurol Neurosurg Psychiatry 2002; 73(6):739-746.

(18) Ergaz Z, Ornoy A. Parvovirus B19 in pregnancy. Reprod Toxicol 2006; 21(4):421-435.

(19) O’Malley A, Barry-Kinsella C, Hughes C, Kelehan P, Devaney D, Mooney E et al. Parvovirus infects cardiac myocytes

in hydrops fetalis. Pediatr Dev Pathol 2003; 6(5):414-420.

(20) Cameron AD, Swain S, Patrick WJ. Human parvovirus B19 infection associated with hydrops fetalis. Aust N Z J

Obstet Gynaecol 1997; 37(3):316-319.

(21) Miller E, Fairley CK, Cohen BJ, Seng C. Immediate and long term outcome of human parvovirus B19 infection in

pregnancy. Br J Obstet Gynaecol 1998; 105(2):174-178.

(22) Hanley JA, Lippman-Hand A. If nothing goes wrong, is everything all right? Interpreting zero numerators. JAMA

1983; 249(13):1743-1745.

(23) Glenn OA, Bianco K, Barkovich AJ, Callen PW, Parer JT. Fetal cerebellar hemorrhage in parvovirus-associated non-

immune hydrops fetalis. J Matern Fetal Neonatal Med 2007; 20(10):769-772.

(24) Rademaker KJ, Uiterwaal CS, Beek FJ, van Haastert IC, Lieftink AF, Groenendaal F et al. Neonatal cranial

ultrasound versus MRI and neurodevelopmental outcome at school age in children born preterm. Arch Dis Child

Fetal Neonatal Ed 2005; 90(6):F489-F493.

(25) Whiting S, Duchowny M. Clinical spectrum of cortical dysplasia in childhood: diagnosis and treatment issues. J

Child Neurol 1999; 14(12):759-771.

(26) Hayward JC, Titelbaum DS, Clancy RR, Zimmerman RA. Lissencephaly-pachygyria associated with congenital

cytomegalovirus infection. J Child Neurol 1991; 6(2):109-114.

(27) Iannetti P, Nigro G, Spalice A, Faiella A, Boncinelli E. Cytomegalovirus infection and schizencephaly: case reports.

Ann Neurol 1998; 43(1):123-127.

(28) Marques Dias MJ, Harmant-van Rijckevorsel G, Landrieu P, Lyon G. Prenatal cytomegalovirus disease and

cerebral microgyria: evidence for perfusion failure, not disturbance of histogenesis, as the major cause of fetal

cytomegalovirus encephalopathy. Neuropediatrics 1984; 15(1):18-24.

(29) Soussotte C, Maugey-Laulom B, Carles D, Diard F. Contribution of transvaginal ultrasonography and fetal

cerebral MRI in a case of congenital cytomegalovirus infection. Fetal Diagn Ther 2000; 15(4):219-223.

(30) Rees S, Stringer M, Just Y, Hooper SB, Harding R. The vulnerability of the fetal sheep brain to hypoxemia at mid-

gestation. Brain Res Dev Brain Res 1997; 103(2):103-118.

(31) de Vries LS, Gunardi H, Barth PG, Bok LA, Verboon-Maciolek MA, Groenendaal F. The spectrum of cranial ultrasound

and magnetic resonance imaging abnormalities in congenital cytomegalovirus infection. Neuropediatrics 2004;

35(2):113-119.

(32) Barkovich AJ, Lindan CE. Congenital cytomegalovirus infection of the brain: imaging analysis and embryologic

considerations. AJNR Am J Neuroradiol 1994; 15(4):703-715.

Chapter 10FETAL CENTRAL NERVOUS SYSTEM

ABNORMALITIES: ULTRASOUND, MRI OR

MULTIDISCIPLINARY DISCUSSIONS?

L.R. Pistorius1, R.H.J.M. Gooskens2, F. Groenendaal3, L.S. de Vries3, Ph. Stoutenbeek1, G.H.A. Visser1

1Department Obstetrics, Division of Perinatology and Gynecology, University Medical Centre Utrecht, The Netherlands

2Department of Neurology and Neurosurgery, Rudolf Magnus Institute of Neuroscience, Utrecht3Department Neonatology, Division Perinatology and Gynaecology, University Medical Centre Utrecht,

Utrecht, The Netherlands

Submitted

| Ch

apte

r 10

152

ABSTRACT

Introduction

There are many articles comparing the ability of ultrasound and MRI to diagnose fetal

central nervous system (CNS) abnormalities. The quality of ultrasound and MRI and

a multidisciplinary discussion might all improve the diagnostic accuracy. We decided

to evaluate our own experience with the diagnostic accuracy of ultrasound, MRI and

multidisciplinary discussions of fetuses with CNS lesions.

Methods

A retrospective cohort of patients referred for MRI of the fetal CNS from 2000 to 2008

was evaluated for the diagnostic accuracy of ultrasound, MRI, and multidisciplinary

discussions before and after MRI as compared to the fi nal postnatal diagnosis. The

primary endpoint was the change in accuracy of the multidisciplinary discussion before

and after the MRI. We also evaluated the accuracy of basic ultrasound examinations

and neurosonograms separately.

Results

Forty six patients were referred for MRI. A postnatal diagnosis could be made in 38,

of whom 28 were discussed in a multidisciplinary team before and after MRI. The

diagnosis made during the multidisciplinary discussion was accurate in 62% (95%CI 22-

77%) before, and 73% (57-85%) of patients after MRI (p=0.17). The diagnosis remained

unchanged in 23/28 patients, changed from inaccurate to accurate in 4, and from

accurate to inaccurate in 1 patient. Basic ultrasounds were accurate in 42% (23-64%)

of patients, and neurosonograms in 63% (41-81%) of patients (p=0.097).

Conclusions

In this cohort, the highest accuracy of prenatal diagnosis of fetal CNS lesions was

obtained with a multidisciplinary discussion after neurosonography and MRI.

| Fet

al c

entr

al n

ervo

us

syst

em a

bn

orm

alit

ies:

ult

raso

un

d, M

RI o

r m

ult

idis

cip

linar

y d

iscu

ssio

ns?

153

INTRODUCTION

Prenatal ultrasound examination is widely used to screen for fetal abnormalities.

When a suspicion of a fetal abnormality arises, ultrasound, MRI, or both, can be used

as the main imaging modality to arrive at a fi nal diagnosis. At the same time, the value

of a thorough history and additional tests such as infectious serology, karyotyping and

tests for coagulopathy should not be forgotten.

Although there are many articles comparing the diagnostic capabilities of US and MRI,

many are hampered by an obvious bias by comparing a diagnosis made in primary

care on US with a tertiary diagnosis by MRI, with only few trying to compare like

with like (1;2). It is also diffi cult to distinguish between the effect of MRI and that of

a multidisciplinary discussion (3). The latter has also been shown to be important in

improving the accuracy of diagnosis of fetal central nervous system abnormalities (4).

We have previously found MRI to be of additional value in the diagnosis of fetal central

nervous system (CNS) abnormalities where ultrasound examination yielded uncertain

or limited results (5). We therefore decided to analyze our subsequent results to

compare the accuracy of ultrasound, MRI and the additional value of a multidisciplinary

discussion with the postnatal diagnosis.

METHODS

All patients with an ultrasound diagnosis or suspicion of a fetal CNS abnormality and who

were referred for antenatal MRI from 2000-2008, were included in this retrospective

cohort from the University Medical Centre Utrecht, one of the ten tertiary perinatal

centres in the Netherlands. The type of ultrasound was noted as a “basic examination”

if only axial views or mainly axial and limited coronal and sagittal views were recorded,

or as a “neurosonogram” when multiple axial, sagittal and coronal views had been

recorded (6). High resolution equipment was used, such as the Toshiba Aplio or Aplio

XG (Toshiba Medical, Tokyo, Japan) or the General Electric Voluson 730 Expert (General

Electric Healthcare, London, United Kingdom) with as a high resolution as possible;

typically 4-8MHz transabdominally or 5-9MHz transvaginally. The fi nal diagnosis (of

the central nervous and other systems) by the maternal and fetal medicine specialist

who performed the last ultrasound examination prior to MRI was noted, as was the

diagnosis made on MRI by the attending radiologist who was aware of the ultrasound

| Ch

apte

r 10

154

fi ndings. MRI was performed on a 1,5T Philips system (Philips Medical Systems,

Best, Netherlands) with T2-weighted, T1-weighted and diffusion-weighted imaging

protocols (7), usually after oral sedation with 10mg diazepam 3 hours prior to the

examination. The settings were adjusted to the gestational age. The T2-weighted

sequence was a single shot turbo-spin echo (SSH-TSE) with fi eld of view (FOV) 170-

300mm, slice thickness 4mm, gap 0-0.4mm, matrix 169-256x256, repetition time (TR)

shortest, echo time (TE) 140-250ms, fl ip angle 90º, 18-25 slices and scan time 15-25

seconds. For the T1-weighted sequence we used two-dimensional gradient echo (2D-

GRE) with FOV 300 – 325 mm, slice thickness 5mm and gap 0.5mm, matrix 166 x 256,

shortest TR, TE 3.8-4.6ms, fl ip angle 21-80º, 14-15 slices and scan time 14-15 seconds,

and for the DWI a FOV 240mm, slice thickness 5mm and gap 0.1mm, matrix 108 x 256,

TR 1,470, TE 125ms, fl ip angle 90º, 16 slices and scan time 19 seconds (5;8;9).

The diagnoses made during a multidisciplinary discussion prior and subsequent to the

MRI were also recorded. The multidisciplinary discussions were regularly attended

by members of the following specialities: maternal and fetal medicine, neonatology,

pediatric neurology, clinical genetics, pathology and pediatric cardiology. The MRI

images were reviewed by a neonatologist / paediatric neurologist (LdV, FG or RG) prior

to the multidisciplinary discussion. The fi nal diagnosis made postnatally by neuro-

imaging or autopsy was used as the gold standard, and the accuracy (true positives

& true negatives divided by the total) was compared with the chi square test. The

diagnoses were classifi ed according to a postnatal neuro-imaging classifi cation of

congenital abnormalities of the CNS (10). The primary outcome measure was the

difference in accuracy of the multidisciplinary meetings’ diagnoses before and after

the MRI examination. Other outcome measures were the difference in accuracy of

basic ultrasound examinations and neurosonograms and the difference in accuracy

between ultrasound and MRI examinations.

Data management and statistical analysis were performed with SPSS version 15.0 for

Windows (SPSS Inc. Chicago, Illinois). Descriptive data are given as mean (standard

deviation) if normally distributed and as median (range) if not normally distributed.

95% confi dence intervals for proportions were calculated with the modifi ed Wald

method (11).

| Fet

al c

entr

al n

ervo

us

syst

em a

bn

orm

alit

ies:

ult

raso

un

d, M

RI o

r m

ult

idis

cip

linar

y d

iscu

ssio

ns?

155

Table 1. Accuracy of ultrasound, multidisciplinary and MRI diagnoses Table 1.1 According to ultrasound diagnosis

Ultrasound diagnosis

Ultrasound Multidisciplinary before

MRI Multidisciplinary after

n* % n % n % n %

Neural tube defects

4/4 100 4/4 100 4/4 100 4/4 100

Posterior fossa lesions

2/2 100 2/2 100 1/2 50 2/2 100

Ventriculo-megaly

4/14 29 3/10 30 4/13 31 7/13 54

Corpus callosum agenesis

3/4 75 2/2 100 3/4 75 3/4 75

1/1 100 1/1 100 1/1 100 1/1 100

Cysts and tumors

2/4 50 3/3 100 3/4 75 4/4 100

Acquired lesions

0/1 0 1/1 100 1/1 100

Other lesions 3/4 73 2/3 67 2/4 50 6/6 100

Normal fi ndings

1/4 25 1/4 25 1/4 25 1/5 20

Table 1.2 According to fi nal diagnosis

Final diagnosis

Ultrasound Multidisciplinary before

MRI Multidisciplinary after

n % n % n % n %

Neural tube defects

4/4 100 4/4 100 4/4 100 4/4 100

Posterior fossa lesions

2/2 100 2/2 100 1/2 50 2/2 100

Aqueduct stenosis

4/4 100 2/2 100 1/3 33 3/3 100

Corpus callosum agenesis

3/5 60 2/2 100 5/5 100 5/5 100

2/9 22 3/9 33 2/9 22 4/9 44

Cysts and tumors

2/2 100 2/2 100 2/2 100 2/2 100

Acquired lesions

0/3 0 0/1 0 0/3 0 0/3 0

Other lesions 2/4 50 2/4 50 2/4 50 2/4 50

Normal fi ndings

1/5 20 1/3 33 3/5 60 5/5 100

*correct diagnoses / total

| Ch

apte

r 10

156

RESULTS

Forty six patients were referred for MRI because of an ultrasound diagnosis or suspicion

of a fetal CNS abnormality during the study period. From 2000 to 2005, the majority of

ultrasound examinations (20/26, or 71%) were basic examinations. From 2006 onwards,

the majority of ultrasound examinations (15/20, or 75%) were neurosonograms. The

median gestational age of the last ultrasound examination before MRI was 31 weeks

(range 20 – 38 weeks) and the median gestational age at MRI was 32 weeks (21 – 38

weeks). The median interval between the ultrasound and MRI examinations was 5 days

(range 0 – 30 days). The MRI examination failed in one patient due to claustrophobia.

In eight patients, a fi nal CNS diagnosis could not be established because no postnatal

imaging or autopsy was performed, or because severe maceration made an accurate

diagnosis impossible. The accuracy of the ultrasound, multidisciplinary meeting before

and after MRI and MRI diagnoses are given in Table 1 according to the main diagnostic

groups.

Table 2. Accuracy of antenatal diagnosis according to type of ultrasound examinationTable 2.1 Basic ultrasound

Ultrasound Multidisciplinary before

MRI Multidisciplinary after

n % n % n % n %

true diagnosis 8 42 7 54 11 58 13 68

false negative 2 11 2 15 2 11 2 11

false positive 1 5 0 0 0 0 0 0

different diagnosis 8 42 4 31 6 32 4 21

total 19 100 13 100 19 100 19 68

Table 2.2 Neurosonogram

Ultrasound Multidisciplinary before

MRI Multidisciplinary after

n % n % n % n %

true diagnosis 12 63 11 69 9 50 14 78

false negative 1 5 1 6 1 6 1 6

false positive 3 16 2 13 2 11 0 0

different diagnosis 3 16 2 13 6 33 3 17

total 19 100 16 100 18 100 18 100

| Fet

al c

entr

al n

ervo

us

syst

em a

bn

orm

alit

ies:

ult

raso

un

d, M

RI o

r m

ult

idis

cip

linar

y d

iscu

ssio

ns?

157

f

The accuracy of the multidisciplinary meetings before the MRI examinations was

62% (95% confi dence intervals 44-77%) and after the MRI examinations 73% (57-85%)

(Single tailed chi square p=0.17) A multidisciplinary meeting was held before and after

MRI in 28 patients. In 23 (82%) the diagnosis remained unchanged, in 1 (4%) a correct

diagnosis was changed to an incorrect diagnosis after MRI, and in 4 (16%) an incorrect

diagnosis was changed to a correct diagnosis after MRI.

The accuracy of diagnoses according to the type of ultrasound is shown in Table 2. The

lowest accuracy of 42% (23-64%) was seen with basic ultrasound examination and the

highest accuracy of 78% (54-92%) was seen with multidisciplinary meetings after a

neurosonogram (p=0.01). If the patients with neural tube defects where all diagnoses

were 100% accurate are excluded, the accuracy of 61% (39-80%) of neurosonograms

was signifi cantly higher than the accuracy of 31% (14-56%) of basic ultrasound

examinations (p=0.04, single-tailed chi square test).

DISCUSSION

Although the absolute numbers are small, some defi nite trends emerge from this

analysis. Firstly, the spectrum of disorders refl ects the fact that we used MRI in cases

where it would be likely to add to the diagnostic value, rather than as a routine in

patients with (suspicion of) fetal CNS abnormalities on ultrasound. Secondly,

MRI might add to the diagnostic accuracy as can be seen by the (not statistically

signifi cantly) higher accuracy of the diagnoses of the multidisciplinary discussion

after MRI compared to before the MRI. An MRI examination can to some extent

compensate for an insuffi cient ultrasound examination. However, the accuracy of the

specifi c MRI diagnosis as made by the attending radiologist was hardly or no better

than the accuracy of the ultrasound as made by the attending maternal fetal medicine

specialist. The real improvement in diagnostic accuracy was made by discussing the

patient in a multidisciplinary setting. This has been suggested previously (3) and has

also been demonstrated in a study where a panel retrospectively reviewed the image

material and case management of fetuses where ultrasound and brain MRI had been

done (12). Finally, the greatest improvement was achieved by performing a full fetal

neurosonogram, which entails transvaginal ultrasonography if the fetus is in a cephalic

presentation, and manipulating the fetus to visualize the intracranial content in axial,

coronal and sagittal planes (6).

| Ch

apte

r 10

158

Many studies have found that MRI changed the diagnosis by ultrasound for the better

in 27-29% of fetuses (14-17). If we compare ultrasound and MRI, we would have

changed the diagnosis for the better in 13% of patients, and for the worse in 11% of

patients. The diagnosis of the multidisciplinary meeting improved from 62% before to

73% after the MRI.

Most of this improvement can probably be ascribed to the MRI, although the

ultrasound was often repeated at the same time. It is possible that the execution and

interpretation of MRI in our study were not as good as in the other studies, but it is

also possible that the shorter interval between ultrasound and MRI in our study, and

the fact that we did not compare referral diagnoses with MRI diagnoses led to these

results. Similar results have been reported in other studies where the initial ultrasound

examination had been more detailed (2;18).

We found little additional value in MRI of fetuses with neural tube defects. Most

infants with corpus callosum agenesis were detected by ultrasound, even though the

examinations had been mostly basic, but with indirect signs such as colpocephaly and

an absent cavum septum pellucidum easily visible. With fetal neurosonography it is

even easier to detect corpus callosum agenesis or hypoplasia, as the corpus callosum

and pericallosal vessels can be visualized easily, either directly, or with the aid of three

dimensional ultrasound (19).

Patients with malformations of cortical development mostly presented with

ventriculomegaly. The fi nal diagnosis was made correctly at the fi nal multidisciplinary

discussion in four of the eight patients: two with schizencephaly, both of whom were

correctly diagnosed on MRI and one of whom was correctly diagnosed on ultrasound,

and two with delayed cortical maturation, which was identifi ed in both on ultrasound

and in neither on MRI on the initial evaluation, although it was confi rmed on

revision of the MRI images. Additional pointers to possible malformations of cortical

development which mostly did not lead to the correct diagnosis in our cases were

associated cerebellar or callosal abnormalities (28).

Although MRI has been found superior to ultrasound in the diagnosis of posterior fossa

lesions in many studies (15;17;21-24), we found ultrasound to be equally accurate in

the diagnosis of posterior fossa lesions, whether vermian abnormalities or cysts.

Several of the false positive diagnoses in our study might in fact have been due to

transient fi ndings, such as mild ventriculomegaly or asymmetrical ventriculomegaly

due to intraventricular haemorrhage. The development of the fetal brain remains

a dynamic process, and this could also explain “false negative”diagnoses such as

malformations of cortical development not yet visible at 22 weeks, or peripartum

| Fet

al c

entr

al n

ervo

us

syst

em a

bn

orm

alit

ies:

ult

raso

un

d, M

RI o

r m

ult

idis

cip

linar

y d

iscu

ssio

ns?

159

intracranial haemorrhage which was not present at the time of antenatal imaging.

A change in diagnosis leading to a change in policy has been described in as much as

46% of patients (1;15;17;25;26). We did not evaluate the effect of the MRI on the change

of policy, as policy decisions were usually deferred until all results were available.

So what do our data add to the literature? Obviously our small numbers cannot compete

with the excellent work by experts in the fi eld (1;17;23;27;28). reporting series of even

more than 3000 MRI’s (29). On the other hand, that is precisely the point: although

excellent diagnostic accuracy can be obtained with fetal MRI, both obtaining and

interpreting MRI images are operator dependent (8). There are conditions where the

different physical properties of image generation of MRI can lead to greater accuracy,

such as the early detection of ischemia (8) or estimating the age of a haemorrhage (30).

In units with excellent MRI service and evaluation, the MRI diagnosis is also likely to be

superior to the ultrasound diagnosis (1). On the other hand, if a fetal central nervous

system abnormality is found or suspected on ultrasound, in our setting we would

expect greater gain from performing a neurosonogram and discussing the patient in a

multidisciplinary setting, rather than referring the patient for MRI.

| Ch

apte

r 10

160

REFERENCES

(1) Levine D, Barnes PD, Robertson RR, Wong G, Mehta TS. Fast MR imaging of fetal central nervous system

abnormalities. Radiology 2003; 229(1):51-61.

(2) Malinger G, Ben Sira L, Lev D, Ben Aroya Z, Kidron D, Lerman-Sagie T. Fetal brain imaging: a comparison between

magnetic resonance imaging and dedicated neurosonography. Ultrasound Obstet Gynecol 2004; 23(4):333-

340.

(3) Malinger G, Lev D, Lerman-Sagie T. Is fetal magnetic resonance imaging superior to neurosonography for

detection of brain anomalies? Ultrasound Obstet Gynecol 2002; 20(4):317-321.

(4) Hagmann CF, Robertson NJ, Leung WC, Chong KW, Chitty LS. Foetal brain imaging: ultrasound or MRI. A

comparison between magnetic resonance imaging and a dedicated multidisciplinary neurosonographic opinion.

Acta Paediatr 2008; 97(4):414-419.

(5) Gerards FA, Stoutenbeek P, Gooskens RH, Beek FJ, Groenendaal F. [Diagnostic value of prenatal MRI in fetus with

intracranial anomalies diagnosed by ultrasonography]. Ned Tijdschr Geneeskd 2001; 145(4):179-184.

(6) Malinger G, Monteagudo A, Pilu G, Timor-Tritsch IE, Toi A. Sonographic examination of the fetal central nervous

system: guidelines for performing the ‘basic examination’ and the ‘fetal neurosonogram’. Ultrasound Obstet

Gynecol 2007; 29(1):109-116.

(7) Prayer D, Brugger PC, Prayer L. Fetal MRI: techniques and protocols. Pediatr Radiol 2004; 34(9):685-693.

(8) Brugger PC, Stuhr F, Lindner C, Prayer D. Methods of fetal MR: beyond T2-weighted imaging. Eur J Radiol 2006;

57(2):172-181.

(9) Prayer D, Brugger PC, Prayer L. Fetal MRI: techniques and protocols. Pediatr Radiol 2004; 34(9):685-693.

(10) van der Knaap MS, Valk J. Classifi cation of congenital abnormalities of the CNS. AJNR Am J Neuroradiol 1988;

9(2):315-326.

(11) Agresti A, Coull BA. Approximate is better than “Exact” for interval estimation of binomial proportions. American

Statistician 1998; 52:119-126.

(12) Hagmann CF, Robertson NJ, Leung WC, Chong KW, Chitty LS. Foetal brain imaging: ultrasound or MRI. A

comparison between magnetic resonance imaging and a dedicated multidisciplinary neurosonographic opinion.

Acta Paediatr 2008; 97(4):414-419.

(13) Hagmann CF, Robertson NJ, Leung WC, Chong KW, Chitty LS. Foetal brain imaging: ultrasound or MRI. A

comparison between magnetic resonance imaging and a dedicated multidisciplinary neurosonographic opinion.

Acta Paediatr 2008; 97(4):414-419.

(14) Ismail KM, Ashworth JR, Martin WL, Chapman S, McHugo J, Whittle MJ et al. Fetal magnetic resonance imaging

in prenatal diagnosis of central nervous system abnormalities: 3-year experience. J Matern Fetal Neonatal Med

2002; 12(3):185-190.

| Fet

al c

entr

al n

ervo

us

syst

em a

bn

orm

alit

ies:

ult

raso

un

d, M

RI o

r m

ult

idis

cip

linar

y d

iscu

ssio

ns?

161

(15) Twickler DM, Magee KP, Caire J, Zaretsky M, Fleckenstein JL, Ramus RM. Second-opinion magnetic resonance

imaging for suspected fetal central nervous system abnormalities. Am J Obstet Gynecol 2003; 188(2):492-496.

(16) Wang GB, Shan RQ, Ma YX, Shi H, Chen LG, Liu W et al. Fetal central nervous system anomalies: comparison of

magnetic resonance imaging and ultrasonography for diagnosis. Chin Med J (Engl ) 2006; 119(15):1272-1277.

(17) Whitby EH, Paley MN, Sprigg A, Rutter S, Davies NP, Wilkinson ID et al. Comparison of ultrasound and magnetic

resonance imaging in 100 singleton pregnancies with suspected brain abnormalities. BJOG 2004; 111(8):784-

792.

(18) Kubik-Huch RA, Huisman TA, Wisser J, Gottstein-Aalame N, Debatin JF, Seifert B et al. Ultrafast MR imaging of

the fetus. AJR Am J Roentgenol 2000; 174(6):1599-1606.

(19) Pilu G, Segata M, Ghi T, Carletti A, Perolo A, Santini D et al. Diagnosis of midline anomalies of the fetal brain with

the three-dimensional median view. Ultrasound Obstet Gynecol 2006; 27(5):522-529.

(20) Malinger G, Kidron D, Schreiber L, Ben Sira L, Hoffmann C, Lev D et al. Prenatal diagnosis of malformations of

cortical development by dedicated neurosonography. Ultrasound Obstet Gynecol 2007; 29(2):178-191.

(21) Breysem L, Bosmans H, Dymarkowski S, Schoubroeck DV, Witters I, Deprest J et al. The value of fast MR imaging

as an adjunct to ultrasound in prenatal diagnosis. Eur Radiol 2003; 13(7):1538-1548.

(22) Frates MC, Kumar AJ, Benson CB, Ward VL, Tempany CM. Fetal anomalies: comparison of MR imaging and US for

diagnosis. Radiology 2004; 232(2):398-404.

(23) Sonigo PC, Rypens FF, Carteret M, Delezoide AL, Brunelle FO. MR imaging of fetal cerebral anomalies. Pediatr

Radiol 1998; 28(4):212-222.

(24) Limperopoulos C, Robertson RL, Jr., Khwaja OS, Robson CD, Estroff JA, Barnewolt C et al. How accurately does

current fetal imaging identify posterior fossa anomalies? AJR Am J Roentgenol 2008; 190(6):1637-1643.

(25) Simon EM, Goldstein RB, Coakley FV, Filly RA, Broderick KC, Musci TJ et al. Fast MR imaging of fetal CNS anomalies

in utero. AJNR Am J Neuroradiol 2000; 21(9):1688-1698.

(26) Vimercati A, Greco P, Vera L, Resta M, Selvaggi L. The diagnostic role of “in utero” magnetic resonance imaging.

J Perinat Med 1999; 27(4):303-308.

(27) Girard N, Raybaud C, Gambarelli D, Figarella-Branger D. Fetal brain MR imaging. Magn Reson Imaging Clin N Am

2001; 9(1):19-56, vii.

(28) Raybaud C, Levrier O, Brunel H, Girard N, Farnarier P. MR imaging of fetal brain malformations. Childs Nerv Syst

2003; 19(7-8):455-470.

(29) Zimmerman RA, Bilaniuk LT. Magnetic resonance evaluation of fetal ventriculomegaly-associated congenital

malformations and lesions. Semin Fetal Neonatal Med 2005; 10(5):429-443.

(30) Prayer D, Brugger PC, Kasprian G, Witzani L, Helmer H, Dietrich W et al. MRI of fetal acquired brain lesions. Eur J

Radiol 2006; 57(2):233-249.

Chapter 11LACTATE TO CREATININE RATIO IN AMNIOTIC

FLUID: A PILOT STUDY

H.L. Torrance MD1, L. Pistorius MD1, H.A.M. Voorbij MD PhD2, G.H.A. Visser MD PhD1

1 Perinatal Center, Wilhelmina Children’s Hospital, University Medical Center Utrecht, the Netherlands; 2 Department of Clinical Chemistry, University Medical Center Utrecht, the Netherlands

Submitted

| Ch

apte

r 11

164

ABSTRACT

Background

Measurement of amniotic fl uid lactate concentration in complicated pregnancies

may provide information on the extent of fetal academia. However, normalization

for amniotic fl uid volume may be necessary by calculating the lactate-creatinine (L:

C) ratio.

Methods and principal fi ndings

Amniotic fl uid lactate and creatinine concentrations were obtained at caesarean

section and were compared to lactate concentration in simultaneously collected

arterial cord blood of 28 term and 10 preterm pregnancies. Cord blood lactate was not

correlated to amniotic fl uid lactate, but was correlated to the L:C ratio in the complete

study population (R=0.54, p=0.001) and the subgroups. Correlation was strongest in a

preterm intrauterine growth restricted subgroup (n=7, R=0.83, p=0.02).

Conclusion

The L:C ratio is more accurate in estimating fetal lacticemia than the amniotic

fl uid lactate concentration. When transabdominal amniocentesis is performed for

determination of fetal lung maturity, the L:C ratio can be determined simultaneously

in merely 2 ml of amniotic fl uid. In future, simultaneous non-invasive assessment of

fetal acidosis and fetal lung maturity may become possible by measurement of the L:

C and L/S ratios via MRS and/or infrared spectroscopy.

| Lac

tate

to

Cre

atin

ine

Rati

o in

Am

nio

tic

Flu

id: a

Pilo

t St

ud

y

165

INTRODUCTION

Fetal acidemia may develop in intrauterine growth restricted (IUGR) fetuses as a

result of diminished transport of nutrients and oxygen to the fetus due to placental

insuffi ciency. With increasing placental insuffi ciency, a point may be reached where the

disadvantages of a detrimental intrauterine environment outweigh the disadvantages

of the delivery of an immature infant. In each individual case, clinicians must weigh the

advantages and disadvantages of prolonged intrauterine life and come to a decision

on the optimal time of delivery for the fetus (1).

Determining fetal lung maturity can aid in this decision. This can be assessed in amniotic

fl uid obtained via transabdominal amniocentesis. An earlier study by our group, showed

that the lecithin to sphingomyelin (L/S) ratio is >2 (indicating fetal lung maturity) in

half of the IUGR fetuses between 30 and 32 weeks of gestation (2). Simultaneous

measurement of amniotic fl uid lactate concentration may provide insight into the

extent of fetal lacticaemia. However, this concentration may be infl uenced by changes

in amniotic fl uid volume which is often diminished in IUGR. One approach to ensure

that differences in amniotic fl uid metabolite levels are not due to a different rate of

fetal urine production is to calculate a ratio of the amniotic fl uid metabolite of interest

to creatinine (3). The lactate to creatinine (L:C) ratio may therefore be more accurate

to determine the fetal condition in complicated pregnancies. This ratio has not been

studied before as an indicator of fetal wellbeing in IUGR. However, in postnatal life

the urinary L:C ratio has been shown to be predictive for the development of hypoxic-

ischemic encephalopathy in asphyxiated newborns (4).

The aim of the present study was to measure the L:C ratio and lactate concentration

in amniotic fl uid obtained during caesarean section and to correlate these parameters

with fetal lactate concentrations measured simultaneously in arterial cord blood.

METHODS

In this pilot study, amniotic fl uid samples were collected prospectively from pregnant

women undergoing caesarean section. After obtaining written informed consent,

women with uncomplicated pregnancies undergoing elective caesarean section at

term (for instance for breech presentation or repeat caesarean section) and women

undergoing caesarean section prematurely with an estimated fetal weight <p10 were

| Ch

apte

r 11

166

included in the study. Fetuses with chromosomal and syndromal abnormalities were

excluded.

Amniotic fl uid samples (2 ml) were collected during caesarean section. Samples that

were contaminated with blood or meconium were discarded. Arterial umbilical cord

blood samples were collected simultaneously. In all materials, creatinine and lactate

concentrations were measured immediately (SI units).

Metabolite concentrations were analyzed in two gestational age (GA) groups: term

fetuses (GA>37 weeks) (1) and preterm fetuses (GA≤34 weeks) (2). The latter group

was divided into fetuses with intrauterine growth restriction (IUGR) associated with

placental insuffi ciency (if the pulsatility index (PI) of the umbilical artery was repeatedly

more than 2 standard deviations above the median) and fetuses that were small for

gestational age (SGA) and demonstrated normal umbilical artery PI.

Statistical analysis was performed using the Statistical Package for the Social Sciences

(SPSS version 14.0). Pearson’s correlations were determined or non-parametric

Spearman’s correlations where appropriate. Statistical signifi cance was accepted

when p<0.05.

The study was approved by the Medical Ethics Committee of the University Medical

Center Utrecht, the Netherlands.

RESULTS

After obtaining written informed consent, 54 pregnant women were included in the

study and 56 amniotic fl uid samples were collected (two twin pregnancies). In 18 cases,

the arterial cord blood lactate could not be measured (because too little arterial cord

blood was available or due to errors in blood sampling). These cases were therefore

excluded from analysis (including both twin pregnancies).

All 28 infants delivered electively at term were healthy. Median GA at birth in this

group was 274 days (range 272-278) and median birth weight 3480 grams (range 2855-

4630). Term infants from mothers with insulin treatment for diabetes were admitted

to the neonatal medium care unit for monitoring of blood glucose levels (n=3). Group

2 consisted of 10 infants with a median GA of 215 days (range 199-231) and median

birth weight of 1100 grams (range 760-1380). In this group, seven infants were IUGR,

fi ve infants were delivered for nonreassuring fetal monitoring and nine infants were

delivered for maternal reasons (preeclampsia or HELLP syndrome). All IUGR infants

| Lac

tate

to

Cre

atin

ine

Rati

o in

Am

nio

tic

Flu

id: a

Pilo

t St

ud

y

167

were admitted to the neonatal intensive care unit.

The L:C ratio decreased with increasing gestational age (correlation coeffi cient:

-0.72, p value 0.000). No correlation was found between arterial cord blood lactate

and amniotic fl uid lactate concentration in the complete study population (R=0.25, p

value 0.13), group 1 (R=0.19, p value 0.34) or group 2 (R=0.46, p value 0.18). In contrast,

arterial cord blood lactate concentration was signifi cantly correlated to the L:C ratio in

the complete study population (R=0.54, p value 0.001), group 1 (R=0.59, p value 0.001)

and group 2 (R=0.72, p value 0.02) (see Figure 1). Correlation was strongest in the IUGR

subgroup (R=0.83, p value 0.02).

Figure 1. Correlation between arterial cord blood lactate concentration and amniotic fl uid lactate-creatinine ratio of preterm fetuses

DISCUSSION

The present study indicates that the amniotic fl uid L:C ratio decreases with increasing

gestational age and that this ratio is signifi cantly correlated with arterial umbilical cord

lacticaemia. In contrast, the amniotic fl uid lactate concentration was not correlated

with fetal lacticaemia.

This fi nding may be explained by the fact that the amniotic fl uid lactate concentration

| Ch

apte

r 11

168

is infl uenced by the amniotic fl uid volume, similar to the lecithin concentration (5,6).

Normalization for amniotic fl uid volume was performed by calculating the amniotic

fl uid lactate to creatinine ratio (3).

A good deal of research into normal biochemical composition of amniotic fl uid has been

performed many years ago and it has shown that metabolite concentrations change

during normal pregnancy. With increasing gestational age, lactate concentration falls

whereas creatinine has been shown to increase (7). This means that the L:C ratio can be

expected to fall with increasing gestational age, as was shown in the present study.

To our knowledge, this study is the fi rst to report data on the correlation between

simultaneously measured arterial umbilical cord lactate concentrations and amniotic

fl uid lactate and creatinine concentrations after caesarean section delivery of both term

and SGA preterm infants. The present study shows that the L:C ratio is of signifi cance

in both populations. Unfortunately, this pilot study was not designed to correlate fetal

lacticaemia results with neonatal outcome, which obviously is of great interest. The

urinary L:C ratio, however, has been shown to be predictive for the development of

hypoxic-ischemic encephalopathy in asphyxiated newborns (4). Future studies should

focus on measuring the amniotic fl uid L:C ratio and on correlating this ratio with

neonatal outcome.

From a practical point of view, the L:C ratio may become important in future, as

recent studies have demonstrated the capability to measure various amniotic fl uid

metabolites (including lecithin, lactate and creatinine) non-invasively via magnetic

resonance spectroscopy (MRS) (8-10) or infra red spectroscopy (11). These noninvasive

diagnostic tools could become valuable alternatives to invasive amniocentesis for the

simultaneous assessment of fetal asphyxia and fetal lung maturity.

ACKNOWLEDGMENTS

The authors wish to thank E. van Alderen for her participation in the conception and

design of the study.

| Lac

tate

to

Cre

atin

ine

Rati

o in

Am

nio

tic

Flu

id: a

Pilo

t St

ud

y

169

REFERENCES

(1) Thornton JG, Hornbuckle J, Vail A, Spiegelhalter DJ, Levene M. Infant wellbeing at 2 years of age in the Growth

Restriction Intervention Trial (GRIT): multicentred randomised controlled trial. Lancet 2004; 364: 513-520.

(2) Wijnberger LD, de Kleine M, Voorbij HA, Arabin B, van de Leur JJ et al. The effect of clinical characteristics on

the lecithin/sphingomyelin ratio and lamellar body count: a cross-sectional study. J Matern Fetal Neonatal

Med 2003; 14: 373-382.

(3) di Iorio R, Marinoni E, Coacci F, La Torre R, Cosmi EV. Amniotic fluid nitric oxide and uteroplacental blood flow

in pregnancy complicated by intrauterine growth retardation. Br J Obstet Gynaecol 1997; 104: 1134-1139.

(4) Huang CC, Wang ST, Chang YC, Lin KP, Wu PL. Measurement of the urinary lactate:creatinine ratio for the early

identification of newborn infants at risk for hypoxic-ischemic encephalopathy. N Engl J Med 1999; 341: 328-

335.

(5) Roux JF, Nakamura J, Brown E, Sweet AY, Gluck L. The lecithin-sphingomyelin ratio of amniotic fluid: an index of

fetal lung maturity? Pediatrics 1972; 49: 464-467.

(6) Gluck L, Kulovich MV, Borer RC, Jr., Brenner PH, Anderson GG et al. Diagnosis of the respiratory distress

syndrome by amniocentesis. Am J Obstet Gynecol 1999; 109: 440-445.

(7) Lind T. The biochemistry of amniotic fluid. In: Sandler M editor. Amniotic fluid and its clinical significance. 1st

Edition.New York: Marcel Dekker Inc, 1981.

(8) Bock JL. Metabolic profi ling of amniotic fl uid by proton nuclear magnetic resonance spectroscopy: correlation

with fetal maturation and other clinical variables. Clin Chem 1994; 40: 56-61.

(9) McGowan PE, Reglinski J, Wilson R, Walker JJ, Wisdoms S et al. Quantitative 1H-NMR analysis of amniotic fluid.

J Pharm Biomed Anal 1993; 11: 629-632.

(10) Sims CJ, Fujito DT, Burholt DR, Dadok J, Giles HR et al. Quantification of human amniotic fluid constituents by

high resolution proton nuclear magnetic resonance (NMR) spectroscopy. Prenat Diagn 1993; 13: 473-480.

(11) Liu KZ, Mantsch HH. Simultaneous quantitation from infrared spectra of glucose concentrations, lactate

concentrations, and lecithin/sphingomyelin ratios in amniotic fluid. Am J Obstet Gynecol 1999; 180: 696-702.

SUMMARY

Chapter 12GENERAL DISCUSSION

AND RECOMMENDATIONS

| Ch

apte

r 12

174

INTRODUCTION AND LITERATURE REVIEW

The intricacy of the development of the fetal brain is matched by the complexity of

the literature dealing with the respective merits and demerits of ultrasound and MRI

imaging of the fetal central nervous system. Since the 1960’s, ultrasound has been

used for brain imaging: fi rstly, with A-mode ultrasound (1), with which it was possible

to measure the biparietal diameter and visualize the midline. With B-mode ultrasound

it became feasible to visualize fetal structures in two dimensional (2D) cross-sectional

planes. With the arrival of three dimensional (3D) ultrasound, three orthogonal

planes could be displayed simultaneously, planes which might otherwise have been

inaccessible could be visualized, and volumes could be measured more accurately (2).

During the same time, magnetic resonance imaging (MRI) developed from a technique

which could provide “tissue characterization information that complements the

superior anatomic detail of ultrasound” (3) in the mid eighties, to arguably “the

optimal method for depicting the specifi c abnormalities that characterize each type of

malformation of the brain in the fetus” (4). And therein lies the rub: is MRI the optimal

method, or is “dedicated neurosonology … equal to MRI in the diagnosis of fetal brain

abnormalities” (5)?

Theoretically, the two modalities should be synergistic, as the physics behind image

generation is completely different. Ultrasound generates images by means of the

ultrasound waves refl ected at the interface between areas with different ultrasound

conductivity, and MRI contrast depends on the relative fat, water and proton content

of tissues. In practice, it is diffi cult to determine the relative merit of either technique

from the literature which is often clouded by an implicit or explicit bias for either

modality. Also, imaging does not equate diagnosis, and several publications confi rm

the value of a multidisciplinary approach to the diagnosis of fetal central nervous

system abnormalities (5;6). Both techniques are operator-dependent (7), and both

techniques are deemed to be safe in pregnancy (8;9).

Having said this, ultrasound imaging of the central nervous system seems preferable

(Chapter 2) for:

• screening examinations,

• repeated examinations,

• examinations before 20 weeks of gestation,

• the assessment of fetal movements before the third trimester,

• evaluation of cerebral blood fl ow,

| Gen

eral

dis

cuss

ion

an

d r

ecom

men

dat

ion

s

175

• evaluation of associated (extracranial) abnormalities and

• where MRI is contra-indicated or has failed.

MRI is preferred for:

• diffi cult ultrasound examination (such as oligohydramnios, an engaged fetal

head and ruptured membranes),

• the assessment of:

o posterior fossa abnormalities,

o schizencephaly,

o acute fetal asphyxia or

o severe microcephaly,

• evaluating fetal movements in the late third trimester (albeit at a somewhat

low frame rate),

• detecting and determining the age of intracranial bleeding,

• detecting intracranial tuberous sclerosis and

• postmortem brain imaging.

Both ultrasound and MRI should be combined in:

• cerebellar telangiectasis,

• cytomegaloviral infection,

• intracranial tumors or trauma,

• vein of Galen abnormalities,

• germinal matrix and intraventricular bleeding,

• hemimegalencephaly and

• septo-optic dysplasia.

Either would be suitable after 20 weeks for:

• the assessment of holoprosencephaly

• abnormalities of the corpus callosum,

• evaluation of ventriculomegaly or

• the diagnosis of craniosynostosis.

The literature review in chapter 2 also demonstrated some relatively uncharted

frontiers of the knowledge of the development of normal and abnormal fetal central

nervous system development, which we explored in the sections on physiology and

pathology.

| Ch

apte

r 12

176

PHYSIOLOGY

With the high resolution images of commercially available high frequency transvaginal

transducers and software developed to chart ovarian follicle development, we could

obtain 3D images of the embryonic ventricular system at eight weeks of gestational

age that compared very favorably with images which were previously only obtainable

in a research setting (10;11). In addition, the single three-dimensional sweep used to

obtain these images as well as images which could be used for crown rump length

measurement and further anatomical evaluation could also minimize the exposure of

the rapidly developing embryo to ultrasound energy (Chapter 3). The applicability for

screening remains to be evaluated in greater detail.

The feasibility to limit the time of ultrasound exposure by using 3D ultrasound was

further explored in a longitudinal study in which standard neurosonographical planes

(12) were obtained in real time with two dimensional ultrasound and subsequently

from saved volume blocks obtained with 3D ultrasound. This demonstrated that it was

possible to visualize fetal cortical development (Chapter 4) and measure fetal ventricles

(Chapter 5) as well or better with 3D ultrasound as compared with two dimensional

ultrasound, while limiting the time of fetal exposure to ultrasound, and limiting the

total examination time (scanning and evaluation time) (Chapter 4).

Several publications have demonstrated clearly that fetal cortical development can

be demonstrated well with ultrasound (13-16). However, it is still diffi cult to translate

these fi ndings into practice, as the only the fi rst appearance of sulci and gyri were

typically described, rather than the development over time. Another limitation of

the available literature was the lack of a systematical description of physiological

asymmetry, as only the distal hemisphere is typically seen clearly using transabdominal

ultrasound. In our prospective study, a good intra- and inter-observer agreement was

achieved with a simple scoring system to describe cortical development. Using this

scoring system, and systematically endeavoring to visualize both hemispheres, we

demonstrated asymmetrical cortical development in a third of fetuses between 24

and 28 gestational weeks (Chapter 4). Longitudinal growth curves based on individual

developmental trajectories were generated using multilevel analysis. This scoring

system now needs to be evaluated prospectively to determine whether it is useful to

detect fetuses with malformations of cortical development.

Information on the prevalence of physiological asymmetry of the lateral cerebral

ventricles is also lacking, while the normal size of fetal ventricle is well described in the

| Gen

eral

dis

cuss

ion

an

d r

ecom

men

dat

ion

s

177

literature. A simple cut-off of 10mm for the ventricular atrium as measured on an axial

view has been used as a screening test for ventriculomegaly throughout pregnancy

(12;17). We confi rmed this cut-off value (which had originally been developed from

cross-sectional data) in our longitudinal study. Obtaining the midcoronal view on 2D

ultrasound and using volume contrast imaging (a modality of 3D ultrasound) facilitated

the evaluation of both left and right lateral cerebral ventricles. We demonstrated

asymmetry of the lateral ventricles in 4% of examinations. This is more than the 0.2

– 0.4% that has previously been reported (18). However, it is still much less than the

postnatal incidence of more than 40% (19) (Chapter 5). It would be interesting to

examine this prospectively a group of infants before and after delivery.

Longitudinal growth curves for the cerebellar volume were also generated from

individual developmental trajectories. The left fetal cerebellar hemisphere was found

to be signifi cantly larger than the right, especially before 32 weeks’ gestation. We

tested two ways of measuring the cerebellar volume, namely the multiplanar and

VOCALTM techniques. Both had a good intra- and inter-observer variability, and yielded

similar results. In the late third trimester, our data demonstrated larger fetal cerebellar

volumes than previously reported in Taiwanese (20;21) and Brazilian populations (22).

This might refl ect a difference between the different population groups, or a difference

in technique (Chapter 6). As is the case with the cortical scoring system, it remains

to be evaluated whether measurement of the cerebellar volume, rather than a one-

dimensional measurement such as the transverse cerebellar diameter, is more sensitive

and specifi c to detect syndromic (23) or acquired (24) cerebellar abnormalities.

PATHOLOGY

To determine whether the parameters visible on prenatal imaging could predict future

prognosis, infants who were antenatally diagnosed with spina bifi da were followed up

until demise or fi ve years of age. Multivariate regression analysis showed that a higher

lesion level and head circumference at or above the 90th percentile were independent

predictors of demise. None of the ultrasound features were associated with motor

or mental functioning at fi ve years of age. Prenatal ultrasound performed between

1997 and 2002 predicted the anatomical lesion level within one level of the postnatal

fi ndings in 50%, and in 89% between 2006 and 2007 (p < 0.01). This improvement

in the accuracy of ultrasound is offset by a difference between the functional level

| Ch

apte

r 12

178

of neurological defi cit and the anatomical lesion level which still limits the ability of

ultrasound to predict motor function (Chapter 7).

We also investigated the prognostic value of Doppler ultrasound measurement of

the middle cerebral artery pulsatility index (mca PI) in fetuses with central nervous

system abnormalities. In fetuses with abnormalities of the central nervous system

an antenatal fi nding of increase in mca PI was associated with an increased risk of

adverse neurodevelopmental outcome. This was not found in fetuses with spina

bifi da or multiple abnormalities. An unexpected fi nding was than an increased mca

PI which subsequently decreased was seen in fetuses with malformations of cortical

development, microcephaly and West’s syndrome. Whether this refl ects an underlying

etiology (such as hemorrhage, vasculitis or delayed development of the arterial blood

supply) or is an incidental fi nding remains to be seen. A rapid rise in mca PI was seen

in fetuses with a high risk of demise, especially in fetuses with hydrocephalus. The

advice that a rise in mca PI should be an indication for delivery was not born out by our

fi ndings. One possible explanation is that arterial blood fl ow would only be impaired

at relatively high levels of intracranial pressure and thus refl ect a very advanced stage

of disease progression. Another arterial index (such as the trans systolic time) (25) or

venous fl ow characteristics (26) might be more appropriate, but would need to be

tested prospectively (Chapter 8).

As introduction to the evaluation of our own experience with the relative value of

ultrasound and MRI and multidisciplinary discussions, we present a case of a fetus

that developed malformations of cortical development after requiring intra-uterine

transfusion for a severe Parvovirus B19 infection. Antenatal MRI demonstrated mild

unilateral ventriculomegaly. Polymicrogyria and heterotopia were confi rmed on

postnatal MRI (Chapter 9).

For this evaluation we determined the diagnostic accuracy of ultrasound, MRI, and

multidisciplinary discussions before and after MRI. We investigated a retrospective

cohort of patients where an MRI of the fetal central nervous system was performed

between 2000 and 2008. We also compared the accuracy of standard ultrasound

examinations with that of neurosonograms, and found that the diagnosis made during

the multidisciplinary discussion was accurate in 62% before and 73% of patients after

MRI. The diagnosis remained unchanged in 23 out of 28 patients. The accuracy of the

diagnosis improved in four patients and decreased in one patient after MRI. Standard

ultrasound examinations were accurate in 42% of patients, and neurosonograms in

63% of patients. The highest accuracy of prenatal diagnosis of fetal CNS lesions (78%)

| Gen

eral

dis

cuss

ion

an

d r

ecom

men

dat

ion

s

179

was obtained with a multidisciplinary discussion after neurosonography and MRI

(Chapter 10). No improvement in accuracy after MRI was seen in patients with neural

tube lesions, and the biggest improvement was seen in patients with an ultrasound

diagnosis of ventriculomegaly.

Of interest for future research into using MR spectroscopy to diagnose the degree of

fetal asphyxia non-invasively is the fi nding that the amniotic fl uid lactate-creatinine-

ratio is a good predictor of fetal lacticemia (Chapter 11).

RECOMMENDATIONS AND FUTURE DEVELOPMENTS

• It is possible to obtain clear pictures of the developing cerebral ventricle system

at eight gestational weeks from a single 3D ultrasound volume and automatic

volume calculation software. The value of this technique to diagnose intracranial

and extracranial pathology should be evaluated further.

• It is feasible to grade fetal cortical development using a simple scoring system.

This scoring system should be evaluated for its usefulness to diagnose of fetal

malformations of cortical development.

• It is possible to limit fetal exposure and total examination time by using 3D

ultrasound.

• Both lateral ventricles can usually be visualized with transabdominal ultrasound

by obtaining the mid coronal plane with 2D ultrasound, and by using (3D) volume

contrast imaging.

• Asymmetry is a feature of normal fetal brain development. It is most pronounced

in cortical development and to a lesser extent in ventricle development. A group

of infants should be followed up longitudinally during pre- and postnatally to

fi nd an explanation for the differences in prevalence of asymmetry of the brain

before and after birth.

• Cerebellar volumes can be measured equally well with VOCALTM or the multiplanar

technique. The use of the cerebellar volume to detect abnormalities in cerebellar

growth a should be evaluated prospectively.

• The fetus with a suspected fetal central nervous system lesion should be

discussed in a multidisciplinary meeting. If a fetal neurosonogram does not

provide suffi cient information, an MRI should be performed.

| Ch

apte

r 12

180

• In fetuses with spina bifi da, survival can be predicted accurately with antenatal

ultrasound, but future mental or motor function cannot be predicted.

• A raised mca PI in a fetus with a CNS abnormality is associated with an increased

risk of death or adverse neurodevelopmental outcome. The evaluation of the

fetal cerebral circulation deserves more study; both in terms of predicting

malformations of cortical development, and in terms of timing the delivery of the

fetus with hydrocephalus. 3D ultrasound techniques such as the vascularization

and fl ow indices (27) and MRI techniques to identify hypo- or hyperperfused

areas of the fetal brain or to determine oxygen and pH levels (28;29) might be

useful in this regard.

• Fetal lacticemia (as measure of asphyxia) can be predicted by the ratio of lactate

to creatinine in amniotic fl uid. It should be evaluate how best to assess this non-

invasively with MR spectroscopy (30;31).

• The prenatal use of MRI with higher fi eld strengths than the currently accepted

1.5 Tesla equipment needs further evaluation to confi rm its safety before it

should be used in pregnancy.

• It is also conceivable that the evaluation of fetal movements on ultrasound

and MRI will be included in clinical practice to move from an anatomical and

structural diagnosis to a full fetal neurological examination (32).

• There is still a need for a prospective, unbiased evaluation to determine the

relative value of ultrasound and MRI in the diagnosis of fetal central nervous

system abnormalities.

| Gen

eral

dis

cuss

ion

an

d r

ecom

men

dat

ion

s

181

REFERENCES

(1) Pasto ME, Kurtz AB. Ultrasonography of the normal fetal brain. Neuroradiology 1986; 28(5-6):380-385.

(2) Kalache KD, Eder K, Esser T, Proquitte H, Stoltenburg-Didinger G, Hartung JP et al. Three-dimensional

ultrasonographic reslicing of the fetal brain to assist prenatal diagnosis of central nervous system anomalies. J

Ultrasound Med 2006; 25(4):509-514.

(3) McCarthy SM, Filly RA, Stark DD, Hricak H, Brant-Zawadzki MN, Callen PW et al. Obstetrical magnetic resonance

imaging: fetal anatomy. Radiology 1985; 154(2):427-432.

(4) Raybaud C, Levrier O, Brunel H, Girard N, Farnarier P. MR imaging of fetal brain malformations. Childs Nerv Syst

2003; 19(7-8):455-470.

(5) Malinger G, Ben Sira L, Lev D, Ben Aroya Z, Kidron D, Lerman-Sagie T. Fetal brain imaging: a comparison between

magnetic resonance imaging and dedicated neurosonography. Ultrasound Obstet Gynecol 2004; 23(4):333-

340.

(6) Brugger PC, Stuhr F, Lindner C, Prayer D. Methods of fetal MR: beyond T2-weighted imaging. Eur J Radiol 2006;

57(2):172-181.

(7) Abramowicz JS. Prenatal exposure to ultrasound waves: is there a risk? Ultrasound Obstet Gynecol 2007;

29(4):363-367.

(8) Kanal E. Pregnancy and the safety of magnetic resonance imaging. Magn Reson Imaging Clin N Am 1994;

2(2):309-317.

(9) Hagmann CF, Robertson NJ, Leung WC, Chong KW, Chitty LS. Foetal brain imaging: ultrasound or MRI. A

comparison between magnetic resonance imaging and a dedicated multidisciplinary neurosonographic opinion.

Acta Paediatr 2008; 97(4):414-419.

(10) Blaas HG, Eik-Nes SH, Berg S, Torp H. In-vivo three-dimensional ultrasound reconstructions of embryos and early

fetuses. Lancet 1998; 352(9135):1182-1186.

(11) Tanaka H, Senoh D, Yanagihara T, Hata T. Intrauterine sonographic measurement of embryonic brain vesicle.

Hum Reprod 2000; 15(6):1407-1412.

(12) Malinger G, Monteagudo A, Pilu G, Timor-Tritsch IE, Toi A. Sonographic examination of the fetal central nervous

system: guidelines for performing the ‘basic examination’ and the ‘fetal neurosonogram’. Ultrasound Obstet

Gynecol 2007; 29(1):109-116.

(13) Bernard C, Droulle P, Didier F, Gerard H, Larroche JC, Plenat F et al. [Echographic aspects of cerebral sulci in the

ante- and perinatal period]. J Radiol 1988; 69(8-9):521-532.

(14) Cohen-Sacher B, Lerman-Sagie T, Lev D, Malinger G. Sonographic developmental milestones of the fetal cerebral

cortex: a longitudinal study. Ultrasound Obstet Gynecol 2006; 27(5):494-502.

(15) Monteagudo A, Timor-Tritsch IE. Development of fetal gyri, sulci and fi ssures: a transvaginal sonographic study.

Ultrasound Obstet Gynecol 1997; 9(4):222-228.

| Ch

apte

r 12

182

(16) Toi A, Lister WS, Fong KW. How early are fetal cerebral sulci visible at prenatal ultrasound and what is the normal

pattern of early fetal sulcal development? Ultrasound Obstet Gynecol 2004; 24(7):706-715.

(17) Cardoza JD, Goldstein RB, Filly RA. Exclusion of fetal ventriculomegaly with a single measurement: the width of

the lateral ventricular atrium. Radiology 1988; 169(3):711-714.

(18) Achiron R, Yagel S, Rotstein Z, Inbar O, Mashiach S, Lipitz S. Cerebral lateral ventricular asymmetry: is this a

normal ultrasonographic fi nding in the fetal brain? Obstet Gynecol 1997; 89(2):233-237.

(19) Shen EY, Huang FY. Sonographic fi nding of ventricular asymmetry in neonatal brain. Arch Dis Child 1989;

64(5):730-732.

(20) Chang CH, Chang FM, Yu CH, Ko HC, Chen HY. Assessment of fetal cerebellar volume using three-dimensional

ultrasound. Ultrasound Med Biol 2000; 26(6):981-988.

(21) Hata T, Kuno A, Dai S-Y, Inubashiri E, Hanaoka U, Kanenishi K et al. Three-dimensional sonographic volume

measurement of the fetal cerebellum. J Med Ultrasonics 2007; 34(1):17-21.

(22) Araujo JE, Guimaraes Filho HA, Pires CR, Nardozza LM, Moron AF, Mattar R. Validation of fetal cerebellar volume

by three-dimensional ultrasonography in Brazilian population. Arch Gynecol Obstet 2007; 275(1):5-11.

(23) McCann E, Pilling D, Hesseling M, Roberts D, Subhedar N, Sweeney E. Pontomedullary disconnection: fetal and

neonatal considerations. Pediatr Radiol 2005; 35(8):812-814.

(24) Johnsen SD, Bodensteiner JB, Lotze TE. Frequency and nature of cerebellar injury in the extremely premature

survivor with cerebral palsy. J Child Neurol 2005; 20(1):60-64.

(25) Hanlo PW, Gooskens RH, Nijhuis IJ, Faber JA, Peters RJ, van Huffelen AC et al. Value of transcranial Doppler

indices in predicting raised ICP in infantile hydrocephalus. A study with review of the literature. Childs Nerv Syst

1995; 11(10):595-603.

(26) Pooh RK, Pooh KH, Nakagawa Y, Maeda K, Fukui R, Aono T. Transvaginal Doppler assessment of fetal intracranial

venous fl ow. Obstet Gynecol 1999; 93(5 Pt 1):697-701.

(27) Chang CH, Yu CH, Ko HC, Chen CL, Chang FM. Three-dimensional power Doppler ultrasound for the assessment

of the fetal brain blood fl ow in normal gestation. Ultrasound Med Biol 2003; 29(9):1273-1279.

(28) Levine D. Science to practice: can MR oxygenation imaging be used to assess at-risk pregnancies? Radiology

2006; 238(3):765-766.

(29) Wedegartner U, Tchirikov M, Schafer S, Priest AN, Kooijman H, Adam G et al. Functional MR imaging: comparison

of BOLD signal intensity changes in fetal organs with fetal and maternal oxyhemoglobin saturation during

hypoxia in sheep. Radiology 2006; 238(3):872-880.

(30) Borowska-Matwiejczuk K, Lemancewicz A, Tarasow E, Urban J, Urban R, Walecki J et al. Assessment of fetal

distress based on magnetic resonance examinations: preliminary report. Acad Radiol 2003; 10(11):1274-1282.

(31) Roelants-van Rijn AM, Groenendaal F, Stoutenbeek P, van der GJ. Lactate in the foetal brain: detection and

implications. Acta Paediatr 2004; 93(7):937-940.

(32) de Vries JI, Fong BF. Changes in fetal motility as a result of congenital disorders: an overview. Ultrasound Obstet

Gynecol 2007; 29(5):590-599.

Appendix

NEDERLANDSE SAMENVATTING

AFRIKAANSE OPSOMMING

ACKNOWLEDGMENTS

CURRICULUM VITAE

LIST OF PUBLICATIONS

| Ap

pen

dix

184

NEDERLANDSE SAMENVATTING

Vanaf 1960 kunnen de hersenen van de foetus met echografi e worden afgebeeld.

Met A-mode echografi e (1) kon de bipariëtale diameter gemeten en de falx cerebri

zichtbaar gemaakt worden. B-mode echografi e maakte het mogelijk foetale

structuren in tweedimensionale (2D) snijvlakken zichtbaar te maken. Met de komst

van driedimensionale (3D) echografi e, kunnen drie orthogonale snijvlakken gelijktijdig

worden getoond, en kunnen volumes nauwkeuriger worden gemeten (2). Gelijktijdig

ontwikkelde de Magnetic Resonance Imaging (MRI) zich van een techniek die “informatie

over weefsels kon verstrekken als aanvulling op het superieure anatomische detail van

echografi e” (3), tot een techniek die door sommigen beschouwd wordt als “de optimale

methode om afwijkingen af te beelden die specifi ek zijn voor elke afwijking van de

hersenen van de foetus” (4). Nu rijst dus de vraag of MRI superieur is aan specifi eke

neurosono(=echo)grafi e om foetale hersenenafwijkingen te diagnosticeren (5).

Beide technieken zijn afhankelijk van degene die ze hanteert (6) en beide technieken

worden geacht veilig te zijn in de zwangerschap (7;8). Theoretisch zouden de twee

methoden elkaar kunnen aanvullen, aangezien de fysica achter de beeldvorming

geheel verschillend is. Echografi e produceert een beeld door middel van ultrasone

geluidsgolven die worden teruggekaatst op de overgang van structuren met een

verschillend geluidsgeleidingsvermogen en bij de MRI ontstaat het beeld doordat

weefsels een verschillende samenstelling hebben van onder meer vet en water.

Beeldvorming is geen diagnose. Uit literatuur gegevens is het niet goed mogelijk de

waarde van beide technieken in te schatten. In een aantal publicaties wordt daarom de

nadruk gelegd op een multidisciplinaire benadering om tot een diagnose van foetale

centrale zenuwstelselafwijkingen te komen (5;9).

Uit onderzoek in Hoofdstuk 2 beschrijven blijkt dat echografi e te verkiezen is voor de

volgende onderzoeken:

• screenings onderzoek,

• follow-up onderzoek,

• onderzoek vóór de 20ste zwangerschapsweek,

• beoordeling van foetale bewegingen vóór het derde trimester,

• evaluatie van doorbloeding van de hersenen,

• onderzoek naar afwijkingen anders dan aan het centrale zenuwstelsel en

• indien er een contra-indicatie is voor MRI, of waar MRI geen diagnose

oplevert.

| Ned

erla

nd

se s

amen

vatt

ing

185

MRI heeft de voorkeur bij:

• onvoldoende echografi sche beeldvorming,

• beoordeling van foetale hersenen in geval van:

o afwijkingen in de fossa posterior,

o schizencephaly,

o acute foetale asfyxie en

o ernstige microcefalie,

• evaluatie van foetale bewegingen aan het eind van de zwangerschap,

• evaluatie van intracraniële bloedingen,

• onderzoek naar intracraniële tubereuse sclerose en

• postmortaal onderzoek van de hersenen.

Echografi e en MRI vullen elkaar aan bij onderzoek naar:

• cerebellaire telangiëctasie,

• cytomegalovirusinfectie,

• intracraniële tumoren of trauma,

• afwijkingen van de vena van Galen,

• intraventriculaire en germinale matrix bloedingen,

• hemimegalencephaly en

• septo-optische dysplasie.

Beide technieken zijn na 20 weken in gelijke mate geschikt voor:

• beoordeling van holoprosencephalie,

• afwijkingen aan het corpus callosum,

• evaluatie van ventriculomegalie of

• diagnose van craniosynostosis.

Het literatuuroverzicht, beschreven in Hoofdstuk 2, toonde nog een aantal lacunes

met betrekking tot de kennis omtrent normale en afwijkende ontwikkeling van

de foetale centrale zenuwstelsel. Deze hadden vooral betrekking op de eventuele

toegevoegde waarde van de nieuwe generatie 3D echo apparaten, zowel in beperking

van expositietijd, visualisatie van cerebrale structuren vroeg in de zwangerschap

en volumemetingen. Ontwikkeling van gyri en sulci bleek nog onvoldoende

bestudeerd, alsmede het optreden van een asymmetrie eventueel tijdens de cerebrale

ontwikkeling. Deze aspecten vormden het doel van de onderzoeken beschreven in de

sectie “fysiologie”. In de sectie “pathologie” bestuderen wij de voorspellende waarde

van 2D en 3D echografi e (en MRI) om die kans op overleving en het psychische en

motorische vermogen van kinderen met spina bifi da te bepalen. Ook werd de

| Ap

pen

dix

186

prognostische betekenis van cerebrale arteriële bloedstroomprofi elen op de uitkomst

van de zwangerschap bestudeerd, bij een heel scala van foetale cerebrale afwijkingen.

De diagnostische betrouwbaarheid van echografi e en MRI en de betekenis van

multidisciplinaire team besprekingen werk bestudeerd. Ten slotte werd een eerste

onderzoek gedaan naar de mogelijkheid van MR spectroscopie om de graad van foetale

bedreiging (lactaat) vast te stellen.

FYSIOLOGIE

Met behulp van software speciaal ontwikkeld om ovariële follikel groei te

vervolgen, hebben wij 3D beelden van het ventrikelsysteem van het embryo bij een

zwangerschapstermijn van acht weken verkregen. Deze beelden waren uitstekend

vergelijkbaar met beelden die eerder alleen in een specifi eke onderzoekssetting

mogelijk waren (10; 11). Bovendien kon deze techniek ook gebruikt worden om de

kruin-stuit-lengte te meten en de verdere anatomie te evalueren. Met deze techniek

werd het zich ontwikkelende embryo korter blootgesteld aan ultrageluidenergie dan

bij conventionele echografi e (Hoofdstuk 3). Toepassing van deze techniek bij screening

van foetale afwijkingen moet nog nader worden geëvalueerd.

Om te onderzoeken of met 3D echografi e ook in ander onderzoek de

ultrageluidsblootstelling beperkt werd, is een longitudinaal onderzoek verricht

waarbij de neurosonografi sche snijvlakken (12) in real-time met 2D echografi e en

van opgeslagen 3D volumeblokken werden vergeleken. Dit onderzoek toonde aan

dat de beoordeling van de foetale ventrikels (Hoofdstuk 5) en de ontwikkeling van de

cortex (Hoofdstuk 4) met 3D even betrouwbaar was als met 2D echografi e, terwijl de

tijdsduur van blootstelling van de foetus aan ultrageluid en de totale onderzoekstijd

met 3D echografi e korter was (Hoofdstuk 4).

In verschillende publicaties is aangetoond dat de foetale cortikale ontwikkeling goed

met echografi e (13-16) kan worden bekeken. Eerste verschijning van sulci en gyri is

beschreven, maar onderzoek naar de verdere ontwikkeling van de windingen in tijd

is beperkt. Ook is nauwelijks onderzoek gedaan naar het optreden van fysiologische

cortikale asymmetrie, aangezien met transabdominale echografi e slechts de distale

hemisfeer goed zichtbaar kan worden gemaakt. In ons prospectieve onderzoek werd

een eenvoudig scoresysteem ontwikkeld om cortikale ontwikkeling te beschrijven

met een score van 0 tot 5 voor verschillende stadia van ontplooiing. Met dit systeem

| Ned

erla

nd

se s

amen

vatt

ing

187

toonden wij een asymmetrische cortikale ontwikkeling aan bij een derde van de

foetus tussen 24 en 28 zwangerschapsweken (Hoofdstuk 4). De intra- en interobserver

overeenkomst van dit telsysteem was goed. Met behulp van multi-level analyse werden

longitudinale groeicurven van de verschillende sulci en gyri samengesteld. Het is de

bedoeling dit systeem prospectief te evalueren om de eventuele waarde aan te tonen

bij de identifi catie van afwijkingen van de ontwikkeling van de cortex van de foetus.

Ook over de prevalentie van fysiologische asymmetrie van de laterale hersenventrikels

zijn weinig gegevens bekend, ondanks het feit dat de normale grootte van de foetale

ventrikels uitvoerig beschreven is. De afkapwaarde van 10mm voor het axiaal

gemeten ventriculaire atrium, is vanuit cross-sectionele gegevens ontwikkeld en wordt

gedurende de gehele zwangerschap gebruikt als grenswaarde voor ventriculomegalie

(12; 17). Wij bevestigden deze afkapwaarde in onze longitudinale studie. Het verkrijgen

van het midcoronale vlak met 2D echografi e en het gebruiken van volume contrast

imaging (een modaliteit van 3D echografi e) vereenvoudigden de evaluatie van zowel

linker als rechter laterale hersenventrikels. Wij vonden asymmetrie van de laterale

ventrikels in 4%. Dit is meer dan de 0.2 - 0.4% die in eerdere publicaties gevonden zijn,

(18) maar aanmerkelijk minder dan de 40% die bij postnataal onderzoek is gevonden

(19) (Hoofdstuk 5). Een verklaring voor de verschillen in asymmetrie pre- en postnataal

is niet eenvoudig te geven. Het is van belang om een groep kinderen vóór en na de

bevalling prospectief te vervolgen.

Longitudinale groeicurven van het volume van het cerebellum zijn ook door ons met

multi-level analyse van individuele groeicurven ontwikkeld. De linker cerebellaire

hemisfeer bleek beduidend groter dan de rechter, vooral vóór 32 zwangerschapsweken.

Wij testten twee manieren om het volume van de kleine hersenen te meten, namelijk

de multiplanaire en de VOCALTM techniek. Beide technieken hadden een goede intra-

en interobserver correlatie en leverden gelijkwaardige resultaten op. Aan het einde

van het derde trimester, vonden wij een groter foetaal volume van het cerebellum dan

eerder gerapporteerd in Taiwanese (20; 21) en Braziliaanse bevolkingsgroepen (22). Dit

kan op een populatieverschil of op een verschil in techniek berusten (Hoofdstuk 6).

Onderzocht moet nog worden of meting van het cerebellaire volume, in plaats van een

unidimensionale meting zoals de transcerebellaire diameter, van grotere waarde is om

syndromale (23) of verworven (24) afwijkingen van het cerebellum te diagnosticeren.

| Ap

pen

dix

188

PATHOLOGIE

Om te onderzoeken of prenatale beeldvorming van een foetus met een spina bifi da

prognostische waarde heeft, werden kinderen met een antenataal vastgestelde spina

bifi da gevolgd tot de leeftijd van vijf jaar of tot hun overlijden. De plaats van de spina

in de wervelkolom en de grootte van het hoofd bleken onafhankelijk gecorreleerd

met de kans op overlijden. De motorische en mentale ontwikkeling op 5-jarige leeftijd

waren niet gerelateerd met echografi sche bevindingen. De echografi sche bepaling

van het anatomische letselniveau bleek accuraat binnen één wervelniveau in 50% van

de onderzoeken die tussen 1997 en 2002 plaatsvonden en in 89% van de onderzoeken

die tussen 2006 en 2007 uitgevoerd werden (p < 0.01). Hoewel de nauwkeurigheid van

antenataal echografi sch onderzoek dus toegenomen is, heeft dit geen voorspellende

waarde voor de uiteindelijke motorische functie. Dit komt omdat het functionele

niveau van neurologische uitval niet correleert met het anatomische letselniveau

(Hoofdstuk 7).

Wij onderzochten ook de voorspellende waarde van Doppler bloedstroomprofi elen in

de middelste cerebrale arterie (pulsatiliteitsindex; mca PI) bij foetus met een centrale

zenuwstelselafwijking (Hoofdstuk 8). Bij foetus met een afwijking van het centrale

zenuwstelsel was de aanwezigheid van een verhoogde mca PI, indicatief voor een

verhoogde foetale vaatweerstand, geassocieerd met een verhoogd risico op een

ongunstige neurologische ontwikkeling. Dit werd niet gevonden bij de foetus met een

spina bifi da of complexe hersenafwijkingen. Een onverwachte bevinding was dat als

een initieel verhoogde mca PI later in de zwangerschap afnam, dit geassocieerd bleek

met cortikale ontwikkelingsstoornissen, microcefalie of het syndroom van West. Of de

onderliggende etiologie hiervan een bloeding, vasculitis of vertraagde ontwikkeling

van de arteriële bloedaanvoer of een toevalsbevinding is, dient nader te worden

onderzocht.

Een snelle stijging van de mca PI werd gezien bij de foetus met een hoog risico op

overlijden, vooral bij foetus met een hydrocefalus. Het advies om de zwangerschap

bij een snelle stijging van de mca PI voortijdig te beëindigen wordt daarom niet

door ons bevestigd omdat men dan eigenlijk al te laat is. Een mogelijke verklaring is

dat de arteriële fl ow pas bij een vrij hoge intracraniële druk afneemt en dus op een

vergevorderd stadium van ziekte zou kunnen wijzen. Andere arteriële indices zoals

transsystolische tijd (25) of karakteristieken van de veneuze fl ow (26) zouden mogelijk

een betere voorspellende waarde hebben, maar ook dit moet prospectief worden

| Ned

erla

nd

se s

amen

vatt

ing

189

geëvalueerd.

De aanleiding om onze bevindingen bij echografi e, MRI en multidisciplinaire

besprekingen te evalueren was een casus van een foetus met malformaties van

cortikale ontwikkeling na een intra-uterine transfusie voor een ernstige anemie na

een Parvovirus B19 besmetting (Hoofdstuk 9). Voor deze evaluatie bepaalden wij de

diagnostische nauwkeurigheid van echografi e, MRI en multidisciplinaire besprekingen

voor en na MRI. Wij onderzochten een retrospectief cohort van patiënten bij wie

tussen 2000 tot 2008 een MRI van het foetale centrale zenuwstelsel was verricht. Wij

vergeleken ook de nauwkeurigheid van het standaard echografi sch onderzoek met

dat van neurosonografi e. De diagnose die tijdens de multidisciplinaire bespreking

werd gesteld was accuraat bij 62% vóór en bij 73% van de patiënten na de MRI.

Neurosonografi e was aanmerkelijk beter dan standaard echografi e (juiste diagnose

in respectievelijk 63% en 42% van die gevallen). De hoogste nauwkeurige prenatale

diagnose van foetale centrale zenuwstelselletsels (78%) werd verkregen door een

multidisciplinaire bespreking na neurosonografi e en MRI (Hoofdstuk 10). Bij patiënten

met een echografi sche diagnose van ventriculomegalie gaf MRI de grootste verbetering

in nauwkeurigheid van de diagnose. Bij patiënten met neuraalbuisafwijkingen was

MRI niet van aanvullende betekenis.

Van belang voor toekomstig onderzoek naar de mogelijkheid van MR spectroscopie

om de graad van foetale bedreiging vast te stellen is de bevinding dat het vruchtwater

lactate-creatinine ratio een goede voorspeller is voor foetale lacticemie.

CONCLUSIES EN AANBEVELINGEN

• Het is mogelijk om bij acht weken zwangerschap duidelijke beelden van het

ontwikkelende hersenventrikelsysteem uit één enkel 3D echografi sch volume

te verkregen met behulp van automatische volumen calculatie software. De

waarde van deze techniek voor zowel intracraniële als extracraniële pathologie

dient nader bepaald te worden.

• Een eenvoudig scoringsysteem blijkt goed te functioneren om de foetale

cortikale ontwikkeling te graderen. De waarde hiervan om malformaties van

cortikale ontwikkeling te bepalen dient nader bestudeerd te worden.

• 3D echografi e verkort de tijd noodzaaklijk voor echo onderzoek; dit betreft

zowel de foetale blootstellingtijd als de totale onderzoekstijd.

| Ap

pen

dix

190

• Verkrijging van het midcoronale vlak met 2D echografi e en het gebruik van (3D)

volume contrast imaging bij transabdominale echografi e faciliteert in de meeste

gevallen beoordeling van beide laterale ventrikels.

• Asymmetrie is een kenmerk van normale foetale hersenontwikkeling; dat is

het meest uitgesproken bij de cortikale ontwikkeling en in mindere mate bij

de ventrikelontwikkeling. Het is van belang om een groep kinderen vóór en

na de bevalling prospectief te vervolgen om een verklaring te vinden voor de

verschillen in prevalentie van asymmetrie van de hersenventrikels voor en na

geboorte.

• Cerebellaire volumina kunnen even goed met VOCALTM als met een multiplanaire

techniek gemeten worden. De waarde van volumemetingen om afwijkingen

van cerebellaire ontwikkeling te ontdekken dient prospectief te worden

geëvalueerd.

• Foetus met een mogelijk centraal zenuwstelselletsel moeten multidisciplinair

worden besproken. Indien een foetaal neurosonogram onvoldoende informatie

geeft, dan dient het onderzoek uitgebreid te worden met een MRI.

• Bij de foetus met een spina bifi da kan de overleving met prenatale echografi e

worden voorspeld, maar het latere motorisch en mentaal functioneren van het

kind kan niet worden voorspeld.

• Een verhoogde mca PI bij foetus met een centrale zenuwstelselafwijking duidt

op een groter risico op overlijden of abnormale neurologische ontwikkeling.

Verder onderzoek dient zicht te richten op een eventueel verband met cortikale

ontwikkelingstoornissen en op de waarde ten aanzien van het vaststellen

van het optimale moment van geboorte van een kind met hydrocephalus. 3D

echografi sche technieken zoals vascularisatie en fl ow- Dopplerindexen (27) en

MRI technieken om hypo- of hypergeperfundeerde gebieden van de foetale

hersenen te identifi ceren en daarin hypoxie en pH te bepalen (28;29) zouden

hierbij nuttig kunnen zijn.

• Foetale lacticemie (als marker van asfyxie) kan door de ratio van lactaat tot

creatinine in vruchtwater voorspeld worden. Het dient onderzocht te worden

hoe dit het beste non-invasief met MR spectroscopie bepaald kan worden

(30;31).

• De veiligheid in de zwangerschap van MRI-apparaten met een hogere veldsterkte

dan de momenteel toegelaten 1.5 Tesla apparaten vergt nadere evaluatie,

alvorens dit prenataal gebruikt kan worden.

| Ned

erla

nd

se s

amen

vatt

ing

191

• Mogelijk zal evaluatie van foetale bewegingen met echografi e en MRI een

plaats krijgen in de klinische praktijk niet alleen om een anatomische diagnose

te stellen, maar ook om de neurologische status van de foetus te bepalen (32).

• Er blijft behoefte aan verder prospectief, goed opgezet onderzoek om de

waarde te bepalen van echografi e en MRI bij diagnostiek van foetale centrale

zenuwstelselafwijkingen.

| Ap

pen

dix

192

AFRIKAANSE OPSOMMING

Inleiding en literatuuroorsig

Sedert die 1960’s is dit al moontlik om die fetale brein met sonar te ondersoek. Met A-

mode sonar was dit moontlik om die bipariëtale diameter te meet en die falx af te beeld

(1). Met die koms van B-mode sonar kon fetale strukture in twee-dimensionele (2D)

snitvlakke gesien word. Sedert die beskikbaarheid van drie-dimensionele (3D) sonar

is dit moontlik om drie ortogonale vlakke (vlakke loodreg op mekaar) terselfdertyd te

vertoon, en om vlakke te sien wat andersins nie sigbaar is nie, en om volumes meer

akkuraat te meet (2).

Gedurende dieselfde tyd het MRI ook ontwikkel van ‘n tegniek wat in die mid

tagtigerjare gebruik is vir die “vertoning van weefselkenmerke as aanvulling tot

die beter anatomiese detail van sonar” (3) tot ’n tegniek wat deur sommige beskou

word as “die optimale metode om die spesifi eke abnormaliteite af te beeld wat

elke tipe malformasie van die fetale brein kenmerk” (4). Daar lê die vraag: is MRI die

optimale metode, of is gedetailleerde neurosonografi e net so goed as MRI om fetale

breinabnormaliteite te diagnoseer (5)?

Beide tegnieke is operateur-afhanklik (6), en volgens huidige kennis veilig in swangerskap

(7;8). Teoreties behoort die twee metodes sinergisties te wees, omdat die fi sika agter die

beeldvorming heeltemal verskil. Sonarbeelde word opgewek deur ultraklankgolwe wat

weerkaats word vanaf skeidingsvlakke tussen areas met verskillende klankgeleiding,

terwyl MRI kontras afhanklik is van die relatiewe vet-, water- en protoninhoud van

weefsels. In die praktyk word die literatuur dikwels verduister deur ‘n implisiete of

eksplisiete sydigheid vir een van die twee tegnieke. Beeldvorming is ook nie gelyk

aan diagnose nie, en verskeie publikasies bevestig die waarde van ‘n multidissiplinêre

benadering tot die diagnose van fetale sentrale senustelsel abnormaliteite (5;9).

Dit lyk wel asof sonar verkieslik is vir (Hoofstuk 2):

• siftingsondersoeke,

• herhaalde ondersoeke,

• ondersoeke voor 20 swangerskapsweke,

• die beoordeling van fetale bewegings voor die derde trimester,

• evaluasie van serebrale bloedvloei,

• om vir geassosieerde (ekstrakraniële) afwykings te soek en

• waar daar ‘n kontra-indikasie vir MRI is of die MRI nie geslaag het nie.

193

MRI is weer verkieslik in:

• gevalle van moeilike sonarondersoeke (byvoorbeeld by oligohidramnion, ‘n

ingedaalde fetale skedel en gebroke vliese),

• die beoordeling van

o posterior fossa abnormaliteite,

o skisensefalie,

o akute fetale asfi ksie of

o erge mikrosefalie,

• evaluasie van fetale bewegings laat in die derde trimester,

• evaluasie van intrakraniële bloeding,

• opsporing van intrakraniële tubereuse sklerose en

• vir postmortem beeldvorming.

Sonar en MRI behoort in kombinasie gebruik te word by:

• serebellêre telangiëktase,

• sitomegalovirusinfeksie,

• intrakraniële tumore of trouma,

• malformasies van die vena van Galenus,

• germinale matriks en intraventrikulêre bloeding,

• hemimegalensefalie en

• septo-optiese displasie.

Enige van die twee is na 20 weke geskik vir die beoordeling van:

• holoprosensefalie,

• abnormaliteite van die corpus callosum,

• evaluasie van ventrikulomegalie of

• die diagnose van kraniosinostose.

Die literatuuroorsig in hoofstuk 2 het ook ‘n aantal ongekaarte terreine op die gebied

van die normale en abnormale ontwikkeling van fetale sentrale senustelsel aangedui,

wat verder in die afdelings oor fi siologie en patologie verken word.

FISIOLOGIE

Met behulp van sagteware wat ontwikkel is om ovariële follikelontwikkeling te volg,

was dit moontlik om al teen agt weke 3D beelde van die embrionale ventrikelsisteem

te verkry. Die beelde het uitstekend vergelyk met beelde wat vantevore net in ‘n

| Afr

ikaa

nse

op

som

min

g

| Ap

pen

dix

194

navorsingsopset moontlik was (10;11). Uit die enkele 3D-volume blok waaruit hierdie

beelde verkry is, was dit ook moontlik om de standaardsnitte vir anatomiese evaluasie

en meting van die kruin-romp-lengte te verkry, en terselfdertyd die blootstelling van

die snel-ontwikkelende embrio aan ultraklankenergie te beperk (Hoofstuk 3). Die

toepasbaarheid as siftingsondersoek by hierdie vroeë swangerskapsduur moet nog

verder in detail ondersoek word.

Hoe uitvoerbaar dit is om die blootstelling aan ultraklankenergie te beperk deur

middel van 3D sonar is verder verken in ‘n longitudinale studie waarin standaard

neurosonografi ese vlakke (12) tydens die sonarondersoek met 2D, en daarna vanuit

opgebergde 3D sonar volumes verkry is. Hieruit het dit duidelik geword dat dit moontlik

is om fetale ventrikels (Hoofstuk 5) en kortikale ontwikkeling (Hoofstuk 4) met 3D

sonar met dieselfde akkuraatheid as met 2D te beoordeel, terwyl die tyd waarin die

fetus blootgestel word aan ultraklank, asook die totale ondersoektyd, beperk word

(Hoofstuk 4)

Verskeie publikasies het al duidelik aangetoon dat fetale kortikale ontwikkeling duidelik

met sonar afgebeeld kan word (13-16). Dit is egter nog moeilik om in praktyk toe te pas,

omdat die eerste verskyning van sulci and gyri beskryf word, eerder as die ontwikkeling

met die verloop van tyd. Nog ‘n beperking van die beskikbare literatuur is die gebrek

aan ‘n sistematiese beskrywing van fi siologiese asimmetrie, omdat meestal net die

distale hemisfeer goed sigbaar is met transabdominale sonar. ‘n Eenvoudige telsisteem

is in ons longitudinale studie ontwikkel om kortikale ontwikkeling te beskryf. Ons het

ook probeer om beide hemisfere af te beeld, en asimmetriese kortikale ontwikkeling

is in ‘n derde van fetusse tussen 24 en 28 weke aangetoon (Hoofstuk 4). Die intra- en

inter-waarnemer ooreenkoms van die telsisteem was goed. Longitudinale groeikurwes

is met mutivlak-analise vanuit die individuele groeikurwes bereken. Die volgende stap

is om die telsisteem prospektief te toets om te bepaal of dit nuttig is om fetusse met

malformasies van kortikale ontwikkeling te identifi seer.

Daar is ook min inligting oor die prevalensie van fi siologiese asimmetrie van die

laterale ventrikels beskikbaar, ondanks die feit dat die normale grootte van die fetale

ventrikels goed beskryf is. Die afkapwaarde van 10 mm van die atrium van die laterale

ventrikel op ‘n aksiale vlak is aanvanklik uit data van ‘n dwarssnit-studie bepaal en

word gebruik as afsnypunt vir ventrikulomegalie deur die hele swangerskap (12;17).

Ons het hierdie afsnypunt in ons longitudinale studie bevestig. Deur die midkoronale

vlak op 2D sonar te verkry en die gebruik van volume-kontras beeldvorming (‘n 3D

sonar modaliteit) van aksiale vlakke het die evaluasie van beide regter- en linker-

195

laterale ventrikels vergemaklik. Ons het asimmetrie van die laterale ventrikels in tot

4% gevind. Dis is enersyds meer as wat vroeër in die literatuur by 0,2 tot 0,4% beskryf

word (18). Andersyds is dit aanmerklik minder as die postnatale insidensie van meer as

40% (19) (Hoofstuk 5). Dit sal interessant wees om ‘n groep kinders prospektief voor

en na geboorte te vervolg .

Longitudinale groeikurwes van die serebellêre volume is ook met mutivlak-analise

vanuit individuele fetale groeikurwes bereken. Die linker serebellêre hemisfeer was

betekenisvol groter as regs, veral voor 32 weke. Ons het twee metodes getoets om die

volume te bereken, naamlik die multiplanêre en VOCALTM tegniek. Albei het ’n goeie

intra- en inter-waarnemer betroubaarheid gehad, en soortgelyke waardes opgelewer.

Veral laat in die derde trimester het ons serebellêre volumes gevind wat groter was

as wat vantevore in Taiwanese (20;21) en Braziliese bevolkings beskryf is (22). Dit

weerspieël miskien ‘n verskil tussen die bevolkingsgroepe, of miskien ook ‘n verskil in

tegniek (Hoofstuk 6). Dit moet ook nog bepaal word of metings van die serebellêre

volume, eerder as ‘n een-dimensionele meting soos die transserebellêre diameter, van

groter waarde is om sindromiese (23) of verworwe (24) afwykings van die serebellum

te diagnoseer.

PATOLOGIE

Om te bepaal of prenatale beeldvorming by fetusse met spina bifi da van prognostiese

waarde is, het ons kinders met ‘n prenatale diagnose van spina bifi da tot vyf jaar

ouderdom (of sterfte) opgevolg. Die hoogte van die letsel en die grootte van die

hoofomtrek was volgens multivariant analise onafhanklik met sterfte geassosieerd.

Motoriese of verstandelike funksionering by vyf jaar leeftyd was nie verwant aan

enige sonografi ese bevinding nie. Die bepaling van die anatomiese letselhoogte was

akkuraat binne een werwelvlak in 50% van sonarondersoeke tussen 1997 en 2002, en

in 89% tussen 2006 en 2007 (p < 0.01). Alhoewel die akkuraatheid van sonarondersoek

dus verbeter het, kan motoriese funksie nie voorspel word nie omdat die funksionele

vlak van neurologiese uitval verskil van die anatomiese vlak (Hoofstuk 7).

Ons het ook die prognostiese waarde van Dopplermetings van die middel serebrale

arterie pulsatiliteits-indeks (mca PI) van fetusse met sentrale senustelselafwykings

ondersoek (Hoofstuk 8). In fetusse met abnormaliteite van die sentrale senustelsel, is

‘n verhoogde mca PI geassosieerd met ‘n verhoogde risiko op ‘n agterstand in neuro-

| Afr

ikaa

nse

op

som

min

g

| Ap

pen

dix

196

ontwikkeling. Dit geld nie vir fetusse met spina bifi da of veelvuldige abnormaliteite

nie. ‘n Onverwagte bevinding was dat ‘n verhoogde mca PI wat later verminder

het, geassosieerd was met malformasies van kortikale ontwikkeling, mikrosefalie

of Westsindroom. Of dit verband hou met ‘n onderliggende oorsaak soos bloeding,

vaskulitis of ‘n vertraagde ontwikkeling van die arteriële bloedtoevoer, of ‘n toevallige

bevinding is, moet nog duidelik word.

‘n Skielike styging van die mca PI is gesien in fetusse met ‘n hoë risiko vir sterfte, veral

in fetusse met hidrosefalus. Dat ‘n styging in mca PI ‘n indikasie vir bevalling is, is dus

nie deur ons bevindings bevestig nie omdat dit dan al te laat is. ‘n Moontlike verklaring

hiervoor is dat die arteriële bloedtoevoer eers by ernstig verhoogde intrakraniële

drukke afneem, wat dui op ‘n ver gevorderde stadium van die siekte. ‘n Ander indeks

van arteriële vloei soos die transsistoliese tyd (25) of kenmerke van veneuse vloei (26)

sou miskien van groter waarde wees, maar sal prospektief getoets moet word.

Die aanleiding om ons ervaring met met sonar, MRI en multidissiplinêre besprekings

te evalueer, was ‘n geval van ‘n fetus met malformasies van kortikale ontwikkeling

na intra-uteriene transfusie vir ernstige anemie as gevolg van Parvovirus B19-infeksie

(Hoofstuk 9). Vir die evaluasie het ons die diagnostiese akkuraatheid van sonar, MRI

en multidissiplinêre besprekings voor en na MRI bepaal. Ons het ‘n retrospektiewe

kohort van pasiënte ondersoek by wie van 2000 tot 2008 ‘n MRI van die fetale sentrale

senustelsel verrig is. Ons het ook die akkuraatheid van standaard sonarondersoeke met

die van neurosonografi e vergelyk. Die diagnose van multidissiplinêre besprekings was

akkuraat in 62% voor en in 73% na MRI. Die diagnose was onveranderd in 23 van die

28 pasiënte. In vier pasiënte was die diagnose na MRI meer akkuraat, en in een minder

akkuraat.. Standaard sonarondersoeke was akkuraat in 42% en neurosonografi e

in 63% van pasiënte. Die hoogste diagnostiese akkuraatheid (78%) is verkry met ‘n

multidissiplinêre bespreking na neursonografi e en MRI (Hoofstuk 10). Die akkuraatheid

na MRI het nie verbeter in pasiënte met ‘n neuraalbuisdefek nie, en die grootste

verbetering is gesien in pasiënte met ‘n sonografi ese diagnose van ventrikulomegalie.

Van belang vir toekomstige navorsing na die nut van MR spektroskopie om die graad

van fetale asfi ksie te bepaal, is die bevinding dat die vrugwater laktaat:kreatinien-

verhouding ‘n goeie voorspeller van fetale laktaatasidemie is (Hoofstuk 11).

A

197

AANBEVELINGS EN TOEKOMSTIGE ONTWIKKELINGS

• Dit is moontlik om duidelike beelde van die ontwikkelende serebrale

ventrikelsisteem by agt weke swangerskap te verkry uit ‘n enkele 3D sonarvolume

met behulp van outomatiese volume-berekeningsagteware. Die waarde van

hierdie tegniek vir die diagnose van intra- en ekstrakraniële afwykings behoort

verder ondersoek te word

• Dit is moontlike om fetale kortikale ontwikkeling met ‘n eenvoudige telsisteem

te gradeer. Die telsisteem behoort verder evalueer te word om te bepaal of dit

nuttig is om fetale malformasies van kortikale ontwikkeling te diagnoseer.

• Dit is moontlik om die fetale blootstelling en totale ondersoekstyd te beperk

deur die gebruik van 3D sonografi e.

• Albei laterale ventrikels kan meestal met transabdominale eggografi e beoordeel

word deur die midkoronale vlak op 2D sonar te verkry en met die hulp van 3D

volumekontras.

• Asimmetrie is ‘n kenmerk van normale fetale breinontwikkeling en is

mees uitgesproke in kortikale ontwikkeling, en tot ‘n mindere mate in

ventrikelontwikkeling. Daar behoort ‘n groep fetusse voor en na geboorte

opgevolg te word om ‘n verklaring te vind vir die verskillende prevalensie van

asimmetrie van die brein gedurende swangerskap en postnataal.

• Die serebellêre volume kan met dieselfde betroubaarheid met beide VOCALTM

en multiplanêre metings bepaal word. Die waarde van die gebruik van

volumemetings om abnormaliteite van serebellêre ontwikkeling te diagnoseer

behoort prospektief bepaal te word.

• Die fetus met ‘n vermoedelike afwyking van die sentrale senustelsel moet

multidissiplinêr bespreek word. As fetale neurosonografi e onvoldoende inligting

verskaf, moet ‘n MRI gedoen word.

• In fetusse met spina bifi da kan die oorlewing akkuraat met sonar voorspel word,

maar toekomstige verstandelike of motoriese funksie nie.

• ‘n Verhoogde mca PI in ‘n fetus met ‘n sentrale senustelselabnormaliteit is

geassosieerd met ‘n verhoogde risiko op sterfte of abnormale neurologiese

ontwikkeling. Die evaluasie van die fetale serebrale sirkulasie verdien verdere

navorsing; enersyds om te bepaal of dit van waarde is om kortikale malformasies

te diagnoseer, en andersyds om te bepaal wat die optimale tyd van verlossing

van ‘n fetus met hidrosefalus is. 3D sonartegnieke soos vaskularisasie en vloei-

| Afr

ikaa

nse

op

som

min

g

| Ap

pen

dix

198

Doppler indekse (27) en MRI-tegnieke om areas van hiper- of onderperfusie in

die fetale brein te visualiseer of suurstof- of pH-vlakke te bepaal (28;29) kan

hierby van waarde wees.

• Fetale laktaatasidemie (as mate van asfi ksie) kan voorspel word deur die laktaat-

kreatinien-verhouding in vrugwater. Dit moet nog bepaal word hoe om dit non-

invasief met MR spektroskopie te meet (30;31).

• Die veiligheid in swangerskap van MRI-toerusting met veldsterktes sterker as

die huidig aanvaarde 1,5 Tesla moet eers bevestig word voordat dit prenataal

gebruik kan word.

• Dit is ook heel moontlik dat evaluasie van fetale bewegings op sonar en MRI

in kliniese praktyk gebruik sal word om nie net ‘n anatomiese en strukturele

diagnose te maak nie, maar ‘n volle fetale neurologiese ondersoek te verrig

(32).

• Daar is nog behoefte aan verdere prospektiewe, goed beplande evaluasie

van die waarde van sonar en MRI in die diagnose van fetale sentrale

senustelselafwykings.

199

LITERATUUR

(1) Pasto ME, Kurtz AB. Ultrasonography of the normal fetal brain. Neuroradiology 1986; 28(5-6):380-385.

(2) Kalache KD, Eder K, Esser T, Proquitte H, Stoltenburg-Didinger G, Hartung JP et al. Three-dimensional

ultrasonographic reslicing of the fetal brain to assist prenatal diagnosis of central nervous system anomalies. J

Ultrasound Med 2006; 25(4):509-514.

(3) McCarthy SM, Filly RA, Stark DD, Hricak H, Brant-Zawadzki MN, Callen PW et al. Obstetrical magnetic resonance

imaging: fetal anatomy. Radiology 1985; 154(2):427-432.

(4) Raybaud C, Levrier O, Brunel H, Girard N, Farnarier P. MR imaging of fetal brain malformations. Childs Nerv Syst

2003; 19(7-8):455-470.

(5) Malinger G, Ben Sira L, Lev D, Ben Aroya Z, Kidron D, Lerman-Sagie T. Fetal brain imaging: a comparison between

magnetic resonance imaging and dedicated neurosonography. Ultrasound Obstet Gynecol 2004; 23(4):333-

340.

(6) Brugger PC, Stuhr F, Lindner C, Prayer D. Methods of fetal MR: beyond T2-weighted imaging. Eur J Radiol 2006;

57(2):172-181.

(7) Abramowicz JS. Prenatal exposure to ultrasound waves: is there a risk? Ultrasound Obstet Gynecol 2007;

29(4):363-367.

(8) Kanal E. Pregnancy and the safety of magnetic resonance imaging. Magn Reson Imaging Clin N Am 1994;

2(2):309-317.

(9) Hagmann CF, Robertson NJ, Leung WC, Chong KW, Chitty LS. Foetal brain imaging: ultrasound or MRI. A

comparison between magnetic resonance imaging and a dedicated multidisciplinary neurosonographic opinion.

Acta Paediatr 2008; 97(4):414-419.

(10) Blaas HG, Eik-Nes SH, Berg S, Torp H. In-vivo three-dimensional ultrasound reconstructions of embryos and early

fetuses. Lancet 1998; 352(9135):1182-1186.

(11) Tanaka H, Senoh D, Yanagihara T, Hata T. Intrauterine sonographic measurement of embryonic brain vesicle.

Hum Reprod 2000; 15(6):1407-1412.

(12) Malinger G, Monteagudo A, Pilu G, Timor-Tritsch IE, Toi A. Sonographic examination of the fetal central nervous

system: guidelines for performing the ‘basic examination’ and the ‘fetal neurosonogram’. Ultrasound Obstet

Gynecol 2007; 29(1):109-116.

(13) Bernard C, Droulle P, Didier F, Gerard H, Larroche JC, Plenat F et al. [Echographic aspects of cerebral sulci in the

ante- and perinatal period]. J Radiol 1988; 69(8-9):521-532.

(14) Cohen-Sacher B, Lerman-Sagie T, Lev D, Malinger G. Sonographic developmental milestones of the fetal cerebral

cortex: a longitudinal study. Ultrasound Obstet Gynecol 2006; 27(5):494-502.

(15) Monteagudo A, Timor-Tritsch IE. Development of fetal gyri, sulci and fi ssures: a transvaginal sonographic study.

Ultrasound Obstet Gynecol 1997; 9(4):222-228.

| Afr

ikaa

nse

op

som

min

g

| Ap

pen

dix

200

(16) Toi A, Lister WS, Fong KW. How early are fetal cerebral sulci visible at prenatal ultrasound and what is the normal

pattern of early fetal sulcal development? Ultrasound Obstet Gynecol 2004; 24(7):706-715.

(17) Cardoza JD, Goldstein RB, Filly RA. Exclusion of fetal ventriculomegaly with a single measurement: the width of

the lateral ventricular atrium. Radiology 1988; 169(3):711-714.

(18) Achiron R, Yagel S, Rotstein Z, Inbar O, Mashiach S, Lipitz S. Cerebral lateral ventricular asymmetry: is this a

normal ultrasonographic fi nding in the fetal brain? Obstet Gynecol 1997; 89(2):233-237.

(19) Shen EY, Huang FY. Sonographic fi nding of ventricular asymmetry in neonatal brain. Arch Dis Child 1989;

64(5):730-732.

(20) Chang CH, Chang FM, Yu CH, Ko HC, Chen HY. Assessment of fetal cerebellar volume using three-dimensional

ultrasound. Ultrasound Med Biol 2000; 26(6):981-988.

(21) Hata T, Kuno A, Dai S-Y, Inubashiri E, Hanaoka U, Kanenishi K et al. Three-dimensional sonographic volume

measurement of the fetal cerebellum. J Med Ultrasonics 2007; 34(1):17-21.

(22) Araujo JE, Guimaraes Filho HA, Pires CR, Nardozza LM, Moron AF, Mattar R. Validation of fetal cerebellar volume

by three-dimensional ultrasonography in Brazilian population. Arch Gynecol Obstet 2007; 275(1):5-11.

(23) McCann E, Pilling D, Hesseling M, Roberts D, Subhedar N, Sweeney E. Pontomedullary disconnection: fetal and

neonatal considerations. Pediatr Radiol 2005; 35(8):812-814.

(24) Johnsen SD, Bodensteiner JB, Lotze TE. Frequency and nature of cerebellar injury in the extremely premature

survivor with cerebral palsy. J Child Neurol 2005; 20(1):60-64.

(25) Hanlo PW, Gooskens RH, Nijhuis IJ, Faber JA, Peters RJ, van Huffelen AC et al. Value of transcranial Doppler

indices in predicting raised ICP in infantile hydrocephalus. A study with review of the literature. Childs Nerv Syst

1995; 11(10):595-603.

(26) Pooh RK, Pooh KH, Nakagawa Y, Maeda K, Fukui R, Aono T. Transvaginal Doppler assessment of fetal intracranial

venous fl ow. Obstet Gynecol 1999; 93(5 Pt 1):697-701.

(27) Chang CH, Yu CH, Ko HC, Chen CL, Chang FM. Three-dimensional power Doppler ultrasound for the assessment

of the fetal brain blood fl ow in normal gestation. Ultrasound Med Biol 2003; 29(9):1273-1279.

(28) Levine D. Science to practice: can MR oxygenation imaging be used to assess at-risk pregnancies? Radiology

2006; 238(3):765-766.

(29) Wedegartner U, Tchirikov M, Schafer S, Priest AN, Kooijman H, Adam G et al. Functional MR imaging: comparison

of BOLD signal intensity changes in fetal organs with fetal and maternal oxyhemoglobin saturation during

hypoxia in sheep. Radiology 2006; 238(3):872-880.

(30) Borowska-Matwiejczuk K, Lemancewicz A, Tarasow E, Urban J, Urban R, Walecki J et al. Assessment of fetal

distress based on magnetic resonance examinations: preliminary report. Acad Radiol 2003; 10(11):1274-1282.

(31) Roelants-van Rijn AM, Groenendaal F, Stoutenbeek P, van der GJ. Lactate in the foetal brain: detection and

implications. Acta Paediatr 2004; 93(7):937-940.

(32) de Vries JI, Fong BF. Changes in fetal motility as a result of congenital disorders: an overview. Ultrasound Obstet

Gynecol 2007; 29(5):590-599.

A

y

j

201

ACKNOWLEDGMENTS

To the fi rst author of my life: thank you, Jesus, for all the challenging chapters, and the

peaceful paragraphs in between;

Aan Ronel, Wilma en Hantie: dankie dat julle nou al meer as helfte, of julle hele lewe,

uithou met my terwyl ek probeer om te “work quietly and humbly to realize my

delusions of grandeur.” Julle het die kosbare vermoë om terselfdertyd my geloof in

waarmee ek besig is in stand, my voete op die aarde, en die koffi ebekers vol te hou.

Spesiaal dankie ook Ronel, vir die voorblad en die afskaal van sosiale aktiwiteite (veral

laasgenoemde was sekerlik nie altyd maklik nie), Wilma, vir die musiek, Hantie, vir die

uitnodigings.

Aan al my leermeesters en mentore vir hulle rol om my professioneel te (probeer)

ontwikkel: van die tyd as student toe Jan van der Merwe en W.B.D.I. Evans letterlik

of fi guurlik kop en skouers bo vele ander uitgestaan het. Frieda Coetzee en Yvette

Mouton, julle het my in my huisdokterjaar nie net verloskunde geleer nie, maar

ook om verloskunde lief te hê. Kobie van der Merwe: my lewe is onmeetbaar ryker

danksy ons families se interaksie deur die generasies heen. Al die ander konsultante

van my Tygerberg-tyd: Thinus Kruger, Bob Pattinson, Pohl de Villiers, Hein Odendaal,

Wilhelm Steyn, Gerhard Theron, Bert Schaetzing, Karin Norman vir julle toewyding en

‘n ongeloofl ike vermoë om eerstewêreldstandaarde in die derde wêreld te handhaaf

en natuurlik Linnie Muller vir die fondasie van my sonarvermoëns. Peter Soothill you

have always been there in the background and far too seldom in the foreground. Jan

Deprest voor onkreukbare integriteit en vriendelijke toegankelijkheid. My ex-Kalafong

kollegas: Bob (weer!), Graham Howarth, Gerald Mantel, Eric de Jonge, Henri de Wet;

duidelik het ons daar ‘n internasionale perspektief ontwikkel. Eddie Mhlanga, thanks

for speaking grace over Kalafong. Ernest & Patricia, you helped us to discover where

friendship really comes from. (That was worth going to Hull for!) Peter Brugger, for

your infectious enthousiasm for digging deeper into fetal diagnosis.

En vooral natuurlijk iedereen die dit proefschrift mogelijk maakte: Gerard, je

enthousiasme, optimisme en ervaring waren de benzine voor de motor; Linda, voor je

bereidheid om altijd weer mee te kijken en mee te denken, Rob, Floris en Philip, voor

jullie pragmatisme, realisme, hulp en nuchterheid (vooral om mij te leren dat tussen de

zwart en wit van echo vele grijze nuances liggen); Edu, je heb echt een leeuwenaandeel

gehad aan alles van statistieken tot punctuatie, met ook nog een hele scheut fi losofi e

ertussen! It all really started with the impossible review: a review with both an MRI

| Ack

now

led

gmen

ts

| Ap

pen

dix

202

and ultrasound expert as co-authors: Daniela and Gustavo, thank you for making this

all possible: not only the article, but for the amazing work you both put in to strive for

perfection in fetal imaging. Daniela, thank you also for participating in the assessment

commission, not to mention the symposium. It proves again: if you need someone to

help, fi nd someone who is already too busy. Ook dank aan de andere medeauteurs:

Wendy, Marjet, Sylwia, Manon, Sanne, Marc, Jaime, Timo, Lieve, Malgosia, Dick en Ron:

ieder van jullie heeft geholpen om het bootje in veiligheid te brengen (al voelde het

soms op dekstoelen schikken op de Titanic!) Hein, dank voor je grote hulp met het

Nederlands, het gaat nog om Nederlands op zijn Afrikaans te praten, maar te schrijven

helemaal niet.

Voor Nils en je collegae laboranten, dank voor alle MRI’s, en het helpen uitzoeken wat

kan en wat niet. Ook aan Helen en Jeroen: de spectroscopieën zijn niet gelukt, en wat

twee proefschriften zou moeten zijn, is uiteindelijk een artikel geworden, maar aan

jullie enthousiasme ligt het niet! Dank aan alle echoscopisten voor alle medewerking,

in de bijzonder om proefpersonen te werven. Een heel groot woord van dank aan alle

proefpersonen. Zonder jullie getrouwheid was de hele afdeling over fysiologie niet

mogelijk. Mag alle kleine proefpersoontjes die onwetend meededen, jullie nog veel

vreugde verschaffen! Dank ook aan de andere ouders voor toestemming tot publicatie.

Mag het goed blijven gaan!

En dan mijn collegae in het UMC Utrecht: Gerard, Hein, Martijn, Lieve, Carla, Philip, Jan,

Anneke, Wendy, Michelle, Monique: voor alle samenwerking en collegialiteit, voor tijd

vrij om dit proefschrift mogelijk te maken: de helft van jullie ken ik niet half zo goed

als ik zou willen; en van minder dan de helft van jullie houd ik half zoveel als jullie

verdienen. Bertina, Ans, Lot; jullie altijd vriendelijke en effi ciënte ondersteuning was

op alle verschillende stadia onmisbaar.

Dankie, familie en vriende, wat ons moes mis terwyl ek na Nederland gekom het vir

verskeie redes, waarvan hierdie proefskrif ’n groot deel uitgemaak het. Thank you, all

my friends in Crossroads, especially the life group: Bas, Brigitte, Eric, Gert-Jan, Lillian,

Linda, Marco, Nicole, Petra, Rebecca, Seantel, Sylvie, thanks for all your prayers and

support.

Dan aan de paranimfen Wendy en Corstiaan, voor jullie vermogen om te regelen, te

ondersteunen, noch eens weer een extra mijl te lopen, en niet alleen collegae maar

vrienden te zijn: heel erg bedankt.

To all, aan almal en iedereen: thank you, baie dankie, dank je wel, tov toda, grüß Gott.

203

CURRICULUM VITAE

The author of this thesis was born on the 27th April 1963 in Potchefstroom, South

Africa. He matriculated at the D.F. Malan High School, Bellville in 1980. Medical studies

at the University of Pretoria led to the degree MBChB cum laude in 1986. After an

internship in Boksburg-Benoni Hospital and national service (most of which was

spent as medical offi cer in Tsumeb, Namibia), he moved back to Bellville to study

gynaecology at the Tygerberg Hospital of the University of Stellenbosch. This led to the

degree MMed (O&G) and FCOG (SA) in 1993. After a year as visiting registrar in the Hull

Maternity Hospital in Kingston-upon-Hull in the UK, he took up a post as consultant

gynaecologist at the Kalafong Hospital of the University of Pretoria. From 1998 to 2003

he ran a private maternal and fetal medicine practice in Sandton, Johannesburg, in

the process introducing laser therapy for twin-twin transfusion in South Africa. Since

January 2004 he has been working as obstetrician and latterly perinatologist in the

University Medical Centre Utrecht.

He married Ronel (néé Nel) on the 5th April 1987 and they have two children: Wilma,

born in 1991, and Hantie, born in 1995.

De auteur van dit proefschrift werd geboren op 27 april 1963 in Potchefstroom, Zuid-

Afrika. In 1980 behaalde hij zijn eindexamen gymnasium aan de D. F. Malan middelbare

school in Bellville. Daarna volgde hij zijn geneeskunde studie aan de Universiteit van

Pretoria wat leidde tot een cum laude doctoraal in 1986. Na een periode als AGNIO

(assistent geneeskunde niet in opleiding) gewerkt te hebben in het Boksburg-Benoni

Hospital en het vervullen van de nationale dienstplicht (waarvan de meeste tijd

doorgebracht werd als basisarts in Tsumeb, Namibië), verhuisde hij terug naar Bellville,

waar hij startte met zijn opleiding tot gynaecoloog in het Tygerberg Hospital aan de

universiteit van Stellenbosch. Deze werd in 1993 afgerond. Na een fellowship van

een jaar in het Hull Maternity Hospital in Kingston-upon-Hull te Engeland, was hij

werkzaam als gynaecoloog in het Kalafong Hospital (universiteit van Pretoria). Van

1998 tot 2003 had hij een privépraktijk voor perinatologie in Sandton, Johannesburg. In

deze periode introduceerde hij de lasertherapie voor behandeling van het transfuseur-

transfusée syndroom bij tweelingen in Zuid-Afrika. Sinds januari 2004 versterkt hij

de staf verloskunde (eerst als obstetricus, later als perinatoloog) van het Universitair

Medisch Centrum Utrecht. Hij is getrouwd met Ronel Nel en samen hebben zij twee

kinderen; Wilma, geboren in 1991, en Hantie, geboren in 1995.

| Cu

rric

ulu

m v

itae

205

LIST OF PUBLICATIONS

First author

(1) Pistorius LR, Page-Christiaens GC. First-trimester septated cystic hygroma:

prevalence, natural history, and pediatric outcome. Obstet Gynecol 2005

Dec;106(6):1415-6.

(2) Pistorius LR, Hartman CR. Sonographic diagnosis of subacute puerperal uterine

inversion. J Obstet Gynaecol 1998 Sep;18(5):483.

(3) Pistorius L. Appalling ultrasound knowledge? S Afr Med J 2001 Apr;91(4):272-3.

(4) Pistorius LR, Howarth GR. Failure of amniotic septostomy in the management of

3 subsequent cases of severe previable twin-twin transfusion syndrome. Fetal

Diagn Ther 1999 Nov;14(6):337-40.

(5) Pistorius LR. Routine ultrasound dating has not been shown to be more accurate

than the calendar method. Br J Obstet Gynaecol 1998 Sep;105(9):1039-40.

(6) Pistorius LR, Pattinson RC, Howarth GR, Funk M. Cost-effective on-site screening

for anaemia in pregnancy. S Afr Med J 1997 Aug;87(8):1024.

(7) Pistorius LR, Kruger TF, de VA, van der Merwe JP. A comparative study using

prepared and unprepared frozen semen for donor insemination. Arch Androl

1996 Jan;36(1):81-6.

(8) Pistorius LR, Funk M, Pattinson RC, Howarth GR. Screening for anemia in

pregnancy with copper sulfate densitometry. Int J Gynaecol Obstet 1996

Jan;52(1):33-6.

(9) Pistorius LR, Sweidan WH, Purdie DW, Steel SA, Howey S, Bennett JR, et al.

Coeliac disease and bone mineral density in adult female patients. Gut 1995

Nov;37(5):639-42.

(10) Pistorius LR, Christianson AL. Screening for Down syndrome. S Afr Med J 1995

Sep;85(9):934-6.

(11) Pistorius LR, Kruger TF, van der Merwe JP. [Artifi cial insemination with donor

semen]. S Afr Med J 1993 Jun;83(6):381-2.

(12) Pistorius LR, Pattinson RC. Rational antenatal care. S Afr J Epidemiol Infect

1995;10:66-7.

| Lis

t o

f p

ub

licat

ion

s

| Ap

pen

dix

206

Co-author

(1) Hack KE, Kaandorp JJ, Derks JB, Elias SG, Pistorius L, Visser GH. The value

of umbilical artery Doppler velocimetry in the antenatal surveillance of

uncomplicated monochorionic twin pregnancies. Ultrasound Obstet Gynecol

2008 Jun;31(6):662-8.

(2) Kuchenbecker WK, Pistorius LR, Pattinson RC. The tap test--an accurate fi rst-line

test for fetal lung maturity testing. S Afr Med J 2002 Sep;92(9):720-3.

(3) Jeffery BS, Tsuari M, Pistorius LR, Makin J, Pattinson RC. The impact of a pregnancy

confi rmation clinic on the commencement of antenatal care. S Afr Med J 2000

Feb;90(2):153-6.

(4) Howarth GR, Pistorius LR, Combrink W, Hartman C. Management of early onset

severe twin-twin transfusion syndrome in the absence of fetoscopic equipment

by exteriorisation, ligation and replacement of the umbilical cord of the sacrifi ced

twin. S Afr Med J 1998 Mar;88(3):286.

(5) Howarth GR, Pistorius L, Mantel G, Funk M, Pattinson RC. Induction of labour at

term--misoprostol, effi cacy, economics and ethics. S Afr Med J 1996 Sep;86(9

Suppl):1174, 1176.

(6) Pattinson RC, de JE, Pistorius LR, Howarth GR, de WH, Bremer P, et al. Practical

application of data obtained from a Perinatal Problem Identifi cation Programme.

S Afr Med J 1995 Mar;85(3):131-2.

(7) Funk M, Pistorius LR, Pattinson RC. Antenatal screening for bacterial vaginosis

using the amine test. S Afr J Epidemiol Infect 1996;11:74-6.

(8) Howarth GR, Funk M, Steytler P, Pistorius LR, Makin J, Pattinson RC. A randomised

controlled trial comparing vaginally administered misoprostol to vaginal

dinoprostone gel in labour induction. J Obstet Gynecol 1996;16(6):474-8.

Chapters in books

(1) Pistorius LR. Late reproductive failure. In: Van der Spuy Z, Anthony J, editors.

Handbook of obstetrics and gynaecology.Cape Town: Oxford University Press;

2002. p. 172-6.

(2) Pistorius LR. Hypertensive conditons. In: de Kock J, van der Walt C, editors.

Maternal and newborn care.Lansdowne: Juta; 2004. p. 20-1-20-11.

(3) Pistorius LR. The compromised fetus. In: de Kock J, van der Walt C, editors.

Maternal & newborn care.Lansdowne: Juta; 2002. p. 24-1-24-20.

207

(4) Pistorius LR. Prenatale screening. In: Bakker J, van den Boogaard MHWA, de

Lange B, Rommes JH, van der Voort PHJ, editors. Intensive care capita selecta

2007.Utrecht: Uitgeverij Venticare; 2007. p. 107-16.

| Lis

t o

f p

ub

licat

ion

s

Wat ek geskrywe het,

het ek geskrywe.

Pontius Pilatus