INTRODUCTION

MR imaging of the fetal brain is rapidly being embraced in clinical practice. Ultrasonography (US) is the first step in the in vivo evaluation of fetal central nervous system (CNS). Two- or three-dimensional (2D or 3D) ultrasound is nowadays the standard imaging technique used for prenatal diagnosis of suspected fetal abnormalities. However it is not always easy – or even possible – to perform the US examination because of maternal obesity, oligo- or ahydramnios or fetal position (1). Magnetic resonance imaging (MRI) then becomes the only method of the assessment. It is also regarded as a method solving diagnostic problems in cases with unclear US picture or when sonographic findings are suggestive, but not definitive (2, 3, 4). Fetal MR imaging is proving to be a powerful modality to evaluate the fetal brain and is a valuable adjunct to prenatal ultrasound. The development of ultrafast T2-weighted techniques (known as single-shot rapid acquisition with refocused echoes, (i.e. Single-Shot Fast Spin Echo [SSFSE] or Half-Fourier Acquired Single-Shot Turbo Spin Echo [HASTE]) has contributed to increasing clinical use of fetal MRI. The method allows direct visualization of the developing brain parenchyma, is not susceptible to the same limitations as ultrasound, and has higher contrast resolution than prenatal sonography, thereby allowing better differentiation of normal from abnormal tissue. Structural abnormalities, such as cerebral malformations and destructive lesions, can be sonographically occult on prenatal ultrasound yet detectable by fetal MRI. Fetal MR imaging is primarily used to confirm and characterize brain abnormalities detected by routine prenatal sonography. When a sonographically suspected abnormality is confirmed, MRI is also used to identify any additional sonographically occult central CNS abnormalities. Although no formal data exist, it is well accepted that ultrasound is limited in its ability to detect many of the destructive and developmental lesions that occur prenatally.

MRI has also proved useful in volumetric analysis of important structures like the germinal matrix and lateral ventricles (5). Sectional MRI data provide information on neuronal migration processes, revealing a layered appearance of the cortex and germinal matrix (6). As far as CNS abnormalities are concerned MRI is particularly useful in the evaluation of midline structures, of the posterior cranial fossa and of vertebral canal and the spinal cord.

Information provided by MRI is crucial for the work of the interdisciplinary team – a group of specialists making decisions about pregnancy management, pre- and postnatal management and parental counseling at the tertiary referral centres, where women with complicated pregnancies are referred. The team consists of obstetricians, neonatologists, paediatricians, neurologists, geneticists, radiologists, surgeons and other specialists, depending on the kind of detected fetal abnormality. The decisions made by the team, concerning the choice of place, time and mode of delivery, including termination of pregnancy, require detailed knowledge of the actual status of the fetus. Assessment of CNS is one of the elements of this knowledge. Recognition of cerebral and cerebellar abnormalities requires in turn detailed knowledge of the normal development of fetal brain.

AIM

The purpose of this study was to show the sequence of fetal brain maturation on MR images on the basis of own material and equipment. To our knowledge there are only single publications covering this subject (7,8). Our study, showing the progress of sulcation and gyration week by week, is a very helpful tool in everyday clinical practice.

MATERIAL AND METHOD

In the centres represented by the authors we have 10-year experience with fetal MRI. Up to now over 800 examinations have been performed so far. 259 fetuses have been examined in order to confirm or rule out CNS pathology. 75 of these fetuses, without cerebral abnormalities, were selected for further analysis of T2-weighted images of the brain.

MRI was performed using General Electric (GE) Signa scanners with the magnetic fields’ strength of 1,5T. Four-channel surface phased-array coils were used for scanning.

The pregnant women were positioned feet-first in the magnet, in supine position with the right side slightly elevated (nearly left decubitus position) to reduce the obstruction of the inferior vena cava.

No maternal sedation was used. The women were offered headphones with the chosen music to minimize the noise and to reduce stress.

Three-plane localizer sequence, with a field of view (FOV) of 48 x 48 cm and slice thickness of 10 mm, with the interslice gap of 5 mm, was performed first, to visualize the fetal head. It was followed by the proper sequences with a smaller FOV, but not smaller than 30 x 30 cm to avoid wrap-around artifacts, and thinner slices: 3 mm routinely and 2 mm when smaller details were necessary to assess. The slices in all planes covered the whole fetal brain.

Single-Shot Fast Spin Echo sequence (SSFSE) was applied in T2-weighted images in axial, sagittal and coronal planes in all fetuses. Fast Spoiled Gradient echo (FSPGR) T1-weighted images were also obtained. The attempts were made to obtain truly axial, sagittal and coronal images, but it was not always possible due to fetal movements during the study, so in some cases oblique sections were performed.

RESULTS

On the basis of our studies the sequence of fetal brain maturation during the second half of pregnancy was established. This sequence is presented in Table I.

The collection of MR images was created beginning with 19th week of gestation, showing details of brain structure, gyration and sulcation (Fig.1-13).

DISCUSSION

MR imaging is the first imaging modality to be able to demonstrate the morphological and structural changes in the maturing brain of the fetus. The fetal brain cellularity, the mass production of myelin with its effects on T1 and T2 relaxation times, produce age-specific images that make it possible to evaluate the course of brain maturation. Several studies have been conducted from the first papers describing normal fetal anatomy and brain maturation examined in vivo (9, 10, 11, 12) to the last informations about the newest MRI techniques (13, 14, 15, 16). The results of MRI were compared to pathological findings (6, 17, 18). Post-mortem MR examinations of the fetuses were performed as early as in 1987 (19).

The quality of images is crucial for the correct assessment. The quality in the old studies is not sufficient for this purpose today (9). With modern MRI equipment the quality improves and newer studies provide new details of fetal anatomy.

Fast imaging techniques are applied in order to achieve good quality of the images. T2-weighted images are basic part of MR examination. They provide information about brain morphology, gyration and sulcation and allow biometry. In our study we used Single-Shot Fast Spin Echo sequence with a short scanning time of 16 s. This allowed to obtain sufficient image quality in most cases although even with such a short time the movements of the fetus which causes blurred images require a repeated scan. But short scanning time allows repetition without additional maternal stress due to prolonged examination.

Positioning of the fetal brain within the coil and the magnet and accurate localization of the image slices are another condition for achieving good image quality.

Using high-spatial-resolution MRI and a 4.7T magnet with a gradient power of 20 G cm−1 it is possible to early visualize the internal structure of fetal brain specimen (17). The spatial resolution is then sufficient for a detailed detection of the layers of the brain tissue. In the developing brain, seven transient layers are present in the cerebral cortex (i.e. ventricular, periventricular, subventricular, intermediate, subplate zones, cortical plate and marginal zone) and good correlation has been found between histological and MRI data on T1-weighted images using 2.0T magnet (6). It demonstrates the utility of MRI for studying brain development. Girard and Raybaud described a multilayered pattern on T1-weighted images beginning at approximately 20 weeks of gestation and persisting until approximately 28 weeks. It is characterized by three layers of alternating T1 signal intensity between the germinal matrix and developing cortex.

For imaging fetuses in vivo 1.5 T MRI systems should be used to obtain prenatal images of excellent quality without the necessity of maternal (and fetal) sedation. Still, in some centers sedating the mother with flunitrazepam is used (7). In our centres the sedatives are not used as they introduce a risk to the fetus or to the mother (10).

T1-weighted images are essential to evaluate the progress of white matter myelination in the first months of life (up to six months of postnatal life). They show myelination more clearly than T2-weighted images. The myelinated structures display high signal intensity and appear bright. On T2-weighted images the myelinated white matter is dark. T1-weighted images depict also haemorrhagic lesions, calcifications and lipomas. Their acquisition lasts longer, so they are more prone to motion artifacts and their quality is poorer, it was poorer also in our material.

Anatomical T2-weighted images were used to evaluate structural development of the fetal brain and provided unique information about its development.

• Layering

The ventricular zone, or germinal matrix, forms a smooth band lining the lateral ventricles from early gestation. The germinal matrix gradually regresses during the third trimester, persisting in the roof of the temporal horn, in the lateral wall of the occipital horn until about 33 weeks, and in the caudothalamic groove until several months after birth. The germinal matrix is the source of neuroectodermal elements that constitute the brain parenchyma, giving rise to neuronal and nonneuronal cells. The first group of neurons to migrate out radially from the germinal matrix forms the preplate zone of the neural tube. The second group splits the preplate zone into the marginal zone (future layer I) and the subplate. The cortical plate (layers II-VI) then develops in an inside-out pattern, with new arriving cells migrating over older formed layers, and then stopping before the marginal zone, which contains Cajal-Retzius cells. In human beings, most neuronal migration is complete by approximately 24 weeks of gestation (20).

In human fetuses, from the 12th to the 16th week of gestational age, high-field MRI provides a detailed analysis of the layered cerebral cortex, but only for specimens in vitro and MR field strength over 2.0 T.

In the in vivo examinations using 1.5 T scanner the multilayer structure of the cerebral parenchyma is evident after 16 weeks of gestation. Before 16 weeks of gestation, the spatial resolution does not permit to separate the matrix from the migrating cells. At 19 weeks of gestation, the layer of migrating cells is probably so close to the matrix that the deep intermediate zone could not be differentiated. At 22 weeks of gestation, the migrating cells are adjacent to the matrix; thus, the inner layer on MR imaging sections corresponds to the matrix and to the deeper part of the layer of migrating cells. At 27 weeks of gestation, some migrating cells are also included in the inner layer. Li et al. found three layers in the cerebral cortex at 20–26 weeks of gestation. Approximately between the 20th and 28th week of gestation T2-weighted images show the multilayer structure of brain that represents different zones of the developing cerebrum.

From inside to outside these layers are as follows:

– hypointense germinal matrix (ventricular zone) around the ventricles, highly cellular, with the local thickenings corresponding to ganglionic eminences,

– hyperintense periventricular zone, rich in fibres, moderately hypointense layer consisting of subventricular and intermediate zones (intermediate zone corresponds to the fetal white matter),

– hyperintense subplate zone,

– hypointense cortical plate consisting of marginal zone and developing cerebral cortex, with densely packed cells (6, 21).

There is no definitive explanation of the signal intensities of the different layers. Investigators in a histological study excluded the hypothesis of a difference in myelination status (22). Girard et al. surmise that a close correlation exists between signal intensity and cellular density. The germinal matrix and the cortical plate, which have high cellular densities, exhibit low intensity signals on T2-weighted images. However, the relationship between the relaxation times and the cellular density remains unclear (23, 24). McArdle et al. suggested that the higher interstitial water content in the immature brain could explain the long T1 and T2 relaxation times of white matter (25). True histological architecture is therefore reflected in the radiological images. Kostovic et al. described this multilayer pattern on the basis of T1-weighted images. These were the images of perfect quality, obtained in the formalin-fixed aborted fetuses in a 2.0T scanner (6). In our material these anatomical details were visualized in T2-weighted images in vivo, with smaller field strength (1.5T). Similar results were published by Glenn and Barkovich (21). Also Perkins and Hughes used magnetic resonance imaging of the fetal brain to assess cortical subplate evolution between 20 and 35 weeks gestation. They obtained two-dimensional measures of diameter for the cortex, subplate and fetal white matter. The subplate was originally seen as a continuous band at early gestations measuring up to 4.5 mm (26).

After, approximately, the 28th week of gestation the zonal architecture of the brain is gradually lost due to disappearance of germinal matrix and subplate zone. It becomes magnetic resonance invisible, initially from the depths of the sulci and then from the tops of the gyri. Disappearance of the subplate is regional, involuting most rapidly in the parietal lobe and remaining prominent in the anterior temporal lobe up to 35 weeks (26). This results from the migration processes. The cortex and white matter develop and form a more mature picture of the cerebrum.

• Sulcation

During the early weeks of gestation, the surface of the cerebral hemispheres is smooth. The interhemispheric fissure and the primitive Sylvian fissure appear during the 5th gestational month. Sylvian fissures can be visualized as early as 15 weeks gestational age (12).

After 20 weeks of gestation the corpus callosum can be identified.

At the gestational age of 20-21 weeks, except for the interhemispheric fissure, the surface of the brain appears smooth and gives the impression of “lissencephalic”brain which should not be mistaken for true lissencephaly. The Sylvian fissures are widely open. Pericerebral fluid spaces are also wide. Apart from interhemispheric fissure, the parieto-occipital fissure is also seen. At the age of 22-23 weeks calcarine and hippocampal fissures as well as callosal sulci start to be visible. From 24 to 26 weeks of gestation, the cerebral cortex had a mostly smooth surface, with a few shallow grooves in the central sulcus, in the interparietal sulci, or in the superior temporal sulci. At 24-25 weeks opercularization of Sylvian fissure starts. Calcarine fissure and cingular sulcus appear. At 26 weeks of gestation central and collateral sulci are present. At gestational age of 27 weeks marginal and precentral sulci are detected, followed by postcentral and intraparietal sulci at week 28 and by superior and inferior frontal sulci at week 29. The opercularization progresses and the Sylvian fissure is narrower. The corpus callosum is well seen. The white matter is bright on T2-weighted images, cortical ribbon is seen as a dark layer. From 30 weeks of gestation, the cortex begins to undergo infolding, which is first apparent in the occipital lobe, particularly medially, in the region of the calcarine fissure. At 32 weeks superior and inferior temporal sulci and at 33 weeks – the external occipitotemporal sulcus, are visible. At 34-35 weeks the final shape of gyration is almost achieved. At 36 weeks of gestation opercularization of Sylvian fissure is finished and the cortex is extensively and compactly folded.

• Myelination

Microscopically, myelination is already detectable at 20 weeks of gestation in the medial longitudinal fasciculus of the medulla and pons. T2-weighted images show adequate myelination as an area of low signal intensity. With the progress of myelination subcortical fibres and corona radiata become dark on T2WI.

In the last weeks of gestation (38-40) the posterior limbs of internal capsules also become dark (27).

• Ventricular size

In the fetuses of 12-23 weeks gestational age, the cerebral ventricles are large, which corresponds to the relatively normal fetal hydrocephalus. At the gestational age of 30-31 weeks the ventricular system is narrower. The ventricles gradually became smaller at 33-38 weeks gestational age (12).

• Subarachnoid space

The subarachnoid space along the lateral aspect of the convexities appears as a high-signal-intensity rim of cerebrospinal fluid on T2-weighted images. The width of the subarachnoid space in the middle fossa anterior to the temporal lobe is greater than that in the cisterns and in the dependent portions of the subarachnoid space. The subarachnoid space overlying the convexities is slightly dilated at all gestational ages. Marked dilatation is seen frequently from 21 to 26 weeks of gestation (12). At the gestational age of 36-37 weeks pericerebral fluid spaces become narrow.

Our data are consistent with earlier studies.

It should be kept in mind that – like the process of myelination – the development of the cerebellum is not finished during pregnancy and is continued after birth. The cerebellum is therefore small when compared to the fetal cerebrum and it should not be mistaken for hypoplasia (21).

Ultrasonography, in the hands of expert, may also provide accurate information about gyration (28, 29). However, it is unable to show other maturation processes such as cell migration, axon pathway formation and myelination.

The MR images of subsequent phases of the maturing brain illustrate the capabilities of the modern neuroimaging techniques and provide unique in vivo information about the human fetal brain development. For the radiologists involved in the maternal and fetal medicine, they may serve as a reference in the evaluation whether the fetal brain development is normal or not.

CONCLUSIONS

1.    Ultrasonography still remains the method of choice in the evaluation of fetal brain development, especially in the first half of pregnancy.

2.    MRI is the most useful method of normal fetal brain development assessment. It should be used as an additional tool in every case of diagnostic difficulties or uncertainty.

..............................................................................................................................................................

REFERENCES

1.    Kubik-Huch R.A., Huisman T.A.G.M., Wisser J., Gottstein-Aalame N., Debatin J.F., Seifert B., Ladd M.E., Stallmach T., Marincek B.: Ultrafast MR Imaging of the Fetus. Am. J. Roentgenol., 2000, 175, 1599-1606.

2.    Blaicher W., Prayer D., Mittermayer C., Weninger M., Birnbacher R., Deutinger J, Bernaschek G.: The clinical impact of magnetic resonance imaging in fetuses with central nervous system anomalies on ultrasound scan. Ultraschall Med., 2005, 26(1), 29-35.

3.    Rubod C., Robert Y., Tillouche N., Devisme L., Houfflin-Debarge V., Puech F.: Role of fetal ultrasound and magnetic resonance imaging in the prenatal diagnosis of migration disorders. Prenat. Diagn., 2005, 25(13), 1181-1187.

4.    Hengstschlaeger M.: Fetal magnetic resonance imaging and human genetics. Eur. J. Radiol., 2006, 57(2), 312-315.

5.    Kinoshita Y., Okudera T., Tsuru E., Yokota A.: Volumetric analysis of the germinal matrix and lateral ventricles performed using MR images of postmortem fetuses. Am. J. Neuroradiol., 2001, 22, 382-388.

6.    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, 536-544.

7.    Girard N., Gambarelli D.: Normal Fetal Brain. Magnetic Resonance Imaging. An atlas with anatomic correlations. Francis Brunelle, Donald Shaw, 2001.

8.    Garel C.: MRI of the fetal brain. Normal development and cerebral pathologies. Springer-Verlag Berlin Heidelberg, 2004, p. 35-85.

9.    Girard N., Raybaud C., Poncet M.: In vivo MR study of brain maturation in normal fetuses. Am. J. Neuroradiol., 1995, 16(2), 407-413.

10.    Yamashita Y., Namimoto T., Abe Y., Takahashi M., Iwamasa J., Miyazaki K., Okamura H.: MR imaging of the fetus by a HASTE sequence. Am. J. Roentgenol., 1997, 168(2), 513-519.

11.    Amin R.S., Nikolaidis P., Kawashima A., Kramer L.A., Ernst R.D.: Normal anatomy of the fetus at MR imaging. Radiographics, 1999, 19(Spec no), S201-214.

12.    Lan L.M., Yamashita Y., Tang Y., Sugahara T., Takahashi M., Ohba T., Okamura H.: Normal fetal brain development: MR imaging with a half-Fourier rapid acquisition with relaxation enhancement sequence. Radiology, 2000, 215, 205-210.

13.    Girard N., Gouny S.C., Viola A., Le Fur Y., Viout P., Chaumoitre K., D’Ercole C., Gire C., Figarella-Branger D, Cozzone P.J.: Assessment of normal fetal brain maturation in utero by proton magnetic resonance spectroscopy. Magn. Reson. Med., 2006, 56(4), 768-775.

14.    Prayer D., Kasprian G., Krampl E., Ulm B., Witzani L., Prayer L., Brugger P.C.: MRI of normal fetal brain development. Eur. J. Radiol., 2006, 57(2), 199-216.

15.    Manganaro L., Perrone A., Savelli S., Di Maurizio M., Maggi C., Ballesio L., Porfiri L.M., De Felice C., Marinoni E., Marini M.: Evaluation of normal brain development by prenatal MR imaging. Radiol. Med., (Torino), 2007, 112(3), 444-455.

16.    Rutherford M., Jiang S., Allsop J., Perkins L., Srinivasan L., Hayat T., Kumar S., Hajnal J.: MR imaging methods for assessing fetal brain development. Dev. Neurobiol., 2008, 68(6), 700-711.

17.    Sbarbati A., Pizzini F., Fabene P.F., Nicolato E., Marzola P., Calderan L., Simonati A., Longo L., Osculati A., Beltramello A.: Cerebral cortex three-dimensional profiling in human fetuses by magnetic resonance imaging. J. Anat., 2004, 204(6), 465-474.

18.    Bendersky M., Musolino P.L., Rugilo C., Schuster G., Sica R.E.: Normal anatomy of the developing fetal brain. Ex vivo anatomical-magnetic resonance imaging correlation. J. Neurol. Sci., 2006, 250(1-2), 20-26.

19.    Mintz M.C., Grossman R.I., Isaacson G., Thickman D.I., Kundel H., Joseph P., DeSimone D.: MR imaging of fetal brain. J. Comput. Assist. Tomogr., 1987, 11(1), 120-123.

20.    Glenn O.A.: Fetal Central Nervous System MR Imaging. Neuroimag Clin. N. Am., 2006, 16, 1-17.

21.    Glenn O.A., Barkovich A.J.: Magnetic resonance imaging of the fetal brain and spine: an increasingly important tool in prenatal diagnosis, part I. Am. J. Neuroradiol., 2006, 27, 1604-1611.

22.    Larroche J.C.: The development of the central nervous system during intra-uterine life. In: Human development (Falkner F., ed.). Philadelphia, Pa, Saunders, 1966, p. 257-276.

23.    Girard N.J., Raybaud C.A.: In vivo MRI of fetal brain cellular migration. J. Comput. Assist. Tomogr., 1992, 16, 265-267.

24.    Girard N., Raybaud C., D‘Ercole C., Boubli L., Chau C., Cahen S., Potier A., Gamerre M.: In vivo MR imaging of the fetal brain. Neuroradiology, 1993, 6, 43l-436.

25.    McArdle C.B., Richardson C.J., Nicholas D.A., Mirfakhraee M., Hayden C.K., Amparo E.G.: Developmental features of the neonatal brain: MR imaging. I. Gray-white matter differentiation and myelination. Radiology, 1987, 162, 223-229.

26.    Perkins L., Hughes E., Srinivasan L., Allsop J., Glover A., Kumar S., Fisk N., Rutherford M.: Exploring Cortical Subplate Evolution Using Magnetic Resonance Imaging of the Fetal Brain. Dev. Neurosci., 2008, 30, 211-220.

27.    Glenn O.A., Barkovich A.J.: Magnetic resonance imaging of the fetal brain and spine: an increasingly important tool in prenatal diagnosis: part II. Am. J. Neuroradiol., 2006, 27, 1807-1814.

28.    Ruiz A., Sembely-Taveau C., Paillet C., Sirinelli D.: Sonographic cerebral sulcal pattern in normal fetuses. J. Radiol., 2006, 87(1), 49-55.

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.

..............................................................................................................................................................

Author’s address:

Monika Bekiesińska-Figatowska
Zakład Diagnostyki Obrazowej
Instytut Matki i Dziecka
ul. Kasprzaka 17a, 01-211 Warszawa
tel. (0-22) 32-77-156
fax: (0-22) 32-77-195
zaklad.rtg@imid.med.pl