SEARCH

SEARCH BY CITATION

Keywords:

  • 2D/3D sonography;
  • fetal face;
  • prenatal diagnosis

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Objectives

To detail fetal facial examination in utero using ultrasound, to distinguish between requirements for routine screening and those for precise analysis of fetal facial features, and to assess the use of three-dimensional (3D) ultrasound imaging in fetal facial examination.

Methods

This was a retrospective study, based on the examination of approximately 10 500 fetuses. The usual age at screening ultrasound examination was 22 ± 1 gestational weeks.

Results

The sonographic anatomy of the normal fetal face was depicted, and the relevance of the three reference scanning planes, sagittal, coronal and axial, was specified.

Conclusions

At routine screening using two-dimensional sonography, at least two selected views must be imaged: the mid-sagittal plane (search for facial dysmorphology) and the anterior ‘nose–mouth’ coronal plane (search for disruption in lip continuity or deformation of the narinal bend). Precise analysis of fetal facial anomalies requires visualization and thorough step-by-step analysis in the three planes. The difference between routine screening and precise facial features' analysis has been greatly decreased since 3D and 4D sonography have become readily available. Using this modality, after surface analysis of the fetal face, the three reference planes are simultaneously imaged, using the multiplanar modality, and detailed. 3D/4D ultrasound imaging allows easier, more rapid screening and more precise evaluation of the different facial features. Copyright © 2004 ISUOG. Published by John Wiley & Sons, Ltd.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Prenatal recognition of facial anomalies during pregnancy has many benefits. It can lead to the diagnosis of various genotypic syndromes and chromosomal anomalies1–5. More comprehensive diagnostic imaging planes can be achieved to assist therapeutic options after birth. Also, it allows more adequate counseling and preparation of the parents-to-be6–10.

However, in many institutions, there is no recommendation to include the fetal face in routine ultrasound screening during pregnancy, nor are all proposed examination protocols fully adequate to provide fetal facial anomaly screening.

The detection rate of fetal facial anomalies in utero increases dramatically when ultrasound examination of the face includes an analysis in the three traditional sonographic examination planes, i.e. sagittal, coronal and axial. For example, the generally reported rates of prenatal recognition of cleft lip range between 21% and 30%3, 4, 8, 11–14. However, when all three planes of the face are systematically examined, the detection rate increases to 88%15. This figure is close to rates reported in tertiary centers, where facial examination is part of the work-up for fetuses suspected to have an anomaly on a previous study or for high-risk patients16.

The purpose of our study was to depict the sonographic anatomy of the fetal face, and to propose protocols to either screen for or analyze abnormalities of the fetal face in utero. This report focuses on fetuses with normal facial anatomy.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Study population

Over the 7-year study period (1996–2002), approximately 10 500 fetuses were evaluated sonographically. Women came essentially from a local population for whom the fetal ultrasound evaluation was one of the three ultrasound screening examinations offered to all pregnant women according to routine antenatal care in France. For the study of fetal morphology, which takes place at the second ultrasound screening examination, examinations were preferentially performed at 22 ± 1 gestational weeks. This examination included assessment of the head and face, thorax and heart, abdominal wall and viscera, spine and limbs. The study population also comprised patients referred by other centers because of difficulties during sonographic examination and, in cases when an anomaly was suspected, for confirmatory advice and presurgical evaluation.

The accuracy of the sonographic diagnosis was ascertained by correlation with postnatal findings. It was not possible to obtain postnatal information for all fetuses examined in our center, and therefore the study cannot address the false-negative rate of detection of fetal craniofacial malformation.

Ultrasound examination

Ultrasound examinations were performed using a Combison 530 MT or a Combison 730 (KretzTechnik, Zipf, Austria) ultrasound system, fitted with a 5–8-MHz electronic linear transducer.

Initially, two-dimensional (2D) ultrasound was used (Figure 1). First, the transducer was positioned to visualize the mid-sagittal plane. This plane along with the left and right parasagittal planes were imaged. The transducer was then repositioned to view the mid-sagittal plane, and rotated 90° around its axis, at the level of the nasal tip (a landmark situated approximately midway between vertex and chin). The axial planes were then imaged serially from forehead to chin. The transducer was then maintained in the axial plane positioned at the level of the alveolar ridge. This landmark was chosen as the palate is the central focal point of the examination. The examination plane was rotated a further 90° by sweeping the transducer in an arc over the woman's abdomen to visualize the coronal planes. During the sweep, the fetal palate was used as a reference mark. Usually, the probe was moved in order to insonate the fetal chin first. However, depending on the fetal position, it sometimes proved easier to invert the direction of the sweep so that the coronal planes were imaged first at the level of the junction of the nasal bones with the frontal bone, and proceeding to the palate through the nasal septum. In both cases, serial coronal planes were then imaged, starting at the skin level and moving backwards into the bony structures.

thumbnail image

Figure 1. Examination protocol of the fetal face with two-dimensional ultrasound.

Download figure to PowerPoint

The introduction of three-dimensional (3D) ultrasound allowed a major simplification in protocol. The mid-sagittal view was obtained and the 3D volume acquired. The 3D volume dataset was displayed on the screen using the standard multiplanar reconstruction mode and the views were analyzed following the same systematic examination technique as that used with 2D sonography. The examiner rotated the volume to an optimal position, so that the images were produced in the three planes of interest, i.e. sagittal, axial and coronal. The examiner then scrolled through the volume and was able to move forwards and backwards, visualizing successive parallel planes.

As necessary, more specific 3D modalities were also used. Images with various filter settings were used to differentiate between the bony and soft tissue components. The 3D surface rendered image modality was useful for imaging the soft tissue surface features of the fetal face. The surfacing or the transparency mode was used to image the bony structures, such as the palate (coronal views).

More recently, our protocol was modified to include the use of the 4D-ultrasound (i.e. real-time 3D imaging) modality. The 2D mid-sagittal view was obtained, and the 4D surface-rendered mode switched on, which allowed inspection of the face. One of the recorded volumes was then chosen and analyzed in the multiplanar mode, as described above.

In general, the best volume reconstructions for facial analysis were obtained when the 3D or 4D volume was acquired starting from a mid-sagittal plane. When the face could not be examined, in particular when it was obscured by its proximity to the uterine wall or the placenta, we tried to move the fetus by tapping the maternal abdomen or by having the woman walk around. When this failed, the patient was asked to come back after 1 week.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

A representative view of a normal fetal face as it appears in a mid-sagittal plane is shown in Figure 2. From top to bottom, the following landmarks that should be analyzed are: forehead, nasal soft tissues and nasal bones, upper lip, hard palate, oral cavity, inferior lip and chin. The normal appearance of these features is described in Table 1. Special attention should be directed to the hard palate (Figure 2c). This description of the various features is valid only in the mid-sagittal plane. Indeed, a feature appearing normal in the parasagittal planes is not necessarily normal in the mid-sagittal plane. In the parasagittal planes, the upper lip, the nasal base and the ears must be analyzed (Table 2).

thumbnail image

Figure 2. Representative ultrasound images of a normal fetal face, demonstrating sagittal section planes. (a) Mid-sagittal view (1, forehead; 2, nasal soft tissues; 3, upper lip; 4, chin). (b) Superior facial angle. (c) Hard palate (1, maxilla; 2, maxilla, palatal process; 3, palatal bone, horizontal plate; 4, vomer).

Download figure to PowerPoint

Table 1. Normal appearance of fetal facial features on ultrasound in the mid-sagittal section plane
StructureNormal appearance
Forehead (Figure 2a)Almost linear immediately above the articulation between the nasal bones and the frontal bone, followed by a smooth backwards bend. This view allows measurement of the thickness of the frontal skin (at the level of the middle of the frontal bend).
Nasal bones (Figure 2b)Oblique along a frontocaudal direction. This view allows measurement of the length of the nasal bones, and of the superior facial angle (angle between the vertical section of the frontal bone and the nasal bones).
Nasal soft tissuesThe columella is oblique or horizontal but should not be vertical.
Upper lipThe philtrum is linear, and should present no bulging. This view allows measurement of the length of the philtrum.
Secondary palate (Figure 2c)Thick echoic line, beginning at the alveolar level, and extending horizontally backwards. Its middle is marked by a notch, present on both the superior and inferior edges, and corresponds to the tranverse palatal suture32. The notch is mostly visible on the superior edge.
Oral cavityThe tongue is slightly oblique upwards (10–15° ). Its tip lies immediately behind the alveolar ridge.
Inferior lipRests edge to edge with the upper lip. Both lips are arranged along the same axis; there should be no anteroposterior shift between them.
ChinAt the level of the vertical line traced on the prefrontal skin (‘esthetic vertical line of the face’).
Table 2. Normal appearance of fetal facial features on ultrasound in the parasagittal section planes
StructureNormal appearance
Upper lipThe labial arc is continuous over all the scans. The upper lip has a slightly bulging appearance on these scans.
Nasal base, naresNasal base is smoothly curved, with a concavity directed downwards. There is no lack of continuity on the following scans. Narinal bends are identical on both sides. Both nares look downwards.
EarsAnalysis of size, form and position of the ears complete fetal facial examination. These views allow measurement of the size (length, width) and position of the ears.

A representative view of a normal fetal face as it appears in the axial section planes is shown in Figure 3 and the features are described in Table 3. Similarly, analysis in the coronal planes is shown in Figure 4 and described in Table 4.

thumbnail image

Figure 3. Representative ultrasound images of a normal fetal face, demonstrating axial section planes. (a) Orbits (calipers indicate external interorbital diameter). (b) Nasal septum, malar arches (1, malar bone; 2, nares; 3, skin of the jaw). (c) Upper lip, maxilla, hard palate (1, secondary palate; 2, maxilla; 3, upper lip). (d) Upper lip, alveolus (1, alveolus; 2, upper lip). (e) Oral cavity (1, tongue; 2, alveolus; 3, lip). (f) Mandible (1, mandible; 2, inferior lip).

Download figure to PowerPoint

thumbnail image

Figure 4. Representative ultrasound images of a normal fetal face, demonstrating coronal section planes. (a) Coronal nose–mouth view (1, nares; arrow, nose). (b) Coronal nose–mouth view (3D surface analysis). (c) Aveoli, nasal septum (3D surface analysis; arrow, gingiva and alveolus). (d) Secondary palate (3D surface analysis; arrow, secondary palate). (e) Retropalatal region (3D surface analysis; arrow, posterior part of the secondary palate, supporting the vomer).

Download figure to PowerPoint

Table 3. Normal appearance of fetal facial features on ultrasound in the axial section planes
StructureNormal appearance
Orbits (Figure 3a)The interorbital axis is perpendicular to the strict sagittal axis. This view allows measurement of the inner and outer interorbital lengths.
Nasal septum, molar arches (Figure 3b)The nasal septum is perpendicular to the axial plane. The two malar arches are symmetrical with regard to the nasal septum.
Upper lip, maxilla, alveolus, secondary palate (Figures 3c and d)There is no lack of continuity of the upper lip. The maxilla appears as a regular, U-shaped echoic bend. The alveolus and tooth buds appear as hypoechoic spots regularly layered along the alveolar ridge. There is no shift between adjacent alveoli. The hard palate appears as an echoic structure, with a complex shape. The anterior part is semicircular and lies immediately posterior to the alveolar ridge. The posterior part presents as a rectangular-shaped figure, with a notch on its distal side. This view allows measurement of the size of the maxilla.
Oral cavity (Figure 3e)The tongue occupies the totality of the oral cavity, and is glued to the alveoalar ridge. Posteriorly, the tongue ends at the oropharynx level. This view allows measurement of the width and length of the tongue.
Mandible (Figure 3f)Appears as a regular V-shaped echoic image. Both hemimandibles are almost rectilinear. The symphysis menti is clearly visible. The alveoli appear as regularly layered hypoechoic spots. This view allows measurement of the size of the mandible (for example, mandible width and computation of mandible width/maxilla width ratio10).
Table 4. Normal appearance of fetal facial features on ultrasound in the coronal section planes
StructureNormal appearance
Nose–mouth view (Figure 4a and b)The nostrils are symmetric. The columella is visualized as an intact line.
Alveolus, nasal septum (Figure 4c)The alveolar ridge is regular, and appears as an echoic band where tooth buds appear as hypoechoic holes. The nasal septum is perpendicular to the alveolar ridge lining.
Hard palate (Figure 4d)Bow-shaped, with a symmetrical bend. On each side, it is in continuity with the malar bone.
Retropalatal region (Figure 4e)The vomer bone appears as a medial echoic spot isolated in an empty space, with no visible supporting structure.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Fetal facial screening and/or evaluation with ultrasound

Fetal facial examination with ultrasound is done mainly in two situations: routine screening for the presence of facial anomalies and facial feature analysis in order to delineate precisely the anomaly. Presence of a defect may lead to the discovery of a chromosomal anomaly or a pathomorphological state, whether syndromic or polymalformative. Adequate knowledge of facial status can also help in the planning of the diagnostic work-up and in preparing for the birth.

The recommendations for fetal facial imaging with sonography have changed with the improvement of sonographic equipment. Earlier reports advocated various views, alone or in combination: sagittal17, axial18, coronal19, 20, sagittal and axial21, coronal and sagittal22 and coronal and axial15, 23. However, it rapidly became obvious that precise facial analysis required the examination of all three reference views, coronal, sagittal and axial16, 17, 24–26. The work by Babcook et al.15 brought confirmatory experimental evidence to this opinion. These investigators submitted specimens from the Carnegie collection to ultrasound examination, computed tomography and magnetic resonance imaging and were able to demonstrate that complete analysis of the various pertinent structures of the face required the visualization of all three scanning planes. These experimental data were confirmed in fetuses scanned in vivo.

The mid-sagittal plane allows the sonographer to look for dysmorphology of the facial profile, analyze the profile of the face, and measure various biometric parameters, such as facial angles and nasal bone length (Table 1). The anterior coronal nose–mouth plane (Table 3) is essential to look for disruption in lip continuity, deformation of the narinal bend, and defective alveolar ridge alignment, all being major signs, for example, to assert the absence of a facial cleft15, 16, 23, 24. Serial axial images (Table 3) allow analysis of the various constituents of the face, i.e. eyes, lips, maxilla, tongue and mandible15, 16, 23, 24.

However, in a screening setting, the examination protocol must compromise between high sensitivity and reasonable examination time. Only selected views should be analyzed, and only specific information sought in each view. When 2D sonography is the only available mode, we propose that at least two views should be systematically examined: the mid-sagittal and anterior coronal planes. When precisely analyzing the fetal facial features, imaging of serial axial planes, and also serial sagittal and coronal planes becomes necessary (Figure 5). Table 5 gives examples of the relevance of the different scanning planes to various pathological situations.

thumbnail image

Figure 5. Screening vs. analysis of facial features.

Download figure to PowerPoint

Table 5. Relevance of the different ultrasound screening section planes for the analysis of various fetal facial features
Section planeObjectiveExamples of pathology
Mid-sagittalSearch for facial dysmorphologiesAbnormalities of the nasal bones (hypoplasia or agenesis), protruding maxillary prolabium, position of the tongue (displacement forwards or backwards), retroceding chin
Nose–mouth coronalSearch for facial cleftsLabial clefts, palatal clefts
Staged axialSearch for morphological anomalies of facial partsOrbital anomalies (increased or decreased interorbital length, abnormal orbital diameter, crystalline lenses agenesis), cleft lip and cleft alveolus, (disruption and anteroposterior shift in the alveolar ridge), tongue size (hypo- or hypertrophy), abnormalities of the maxilla or mandible (hypoplasia or agenesis), tongue hypertrophy

Three-dimensional ultrasound

The benefit of 3D ultrasound for fetal facial analysis is generally acknowledged23, 27–31, although recently it has been challenged24.

The surface-rendering mode in the coronal plane is useful for imaging the soft tissue surface features of the fetal face and allows rapid evaluation of the presence or absence of facial dysmorphology or clefting. 3D multiplanar imaging allows standardization of the section planes, simultaneous visualization of the three reference orthogonal planes, and easy step-by-step analysis in each of the planes. Also, 3D multiplanar reconstructions bring more precise information. For instance, this modality is specifically valuable for examination of the posterior coronal views. As such, multiplanar imaging is of high importance for cleft component analysis and in the search for clefting of the secondary palate. Assessment of palate integrity can be facilitated by the use of bony views, in particular in the coronal plane.

The introduction of 3D/4D sonography was a turning point in fetal facial examination. The differentiation between screening and analysis has in part lost relevance. A systematic use of 3D/4D sonography can thus be proposed (Figure 5). Once the mid-sagittal plane is obtained, the volume dataset is acquired (in 4D, one of the recorded blocks of data is selected). First, the surface-rendering mode is used to search for facial dysmorphologies and clefts (analysis of the lips, nasal base, and eyelid obliquity). Then the multiplanar reslicing mode is used, and the three reference planes, sagittal, axial and coronal, are simultaneously displayed and scrutinized.

However, 3D ultrasound analysis has the same drawbacks as occur with 2D visualization. When the fetus is leaning against the uterine wall or the placenta, or14pc in the case of decreased amniotic fluid volume, visualization in all modalities is obscured. Also, specific artifacts exist in 3D/4D ultrasound imaging. However, experience in the technique will soon allow the practicing sonographer to recognize them. Introduction of new algorithms for numerical processing of the data (thick slices, bony views) will in the near future lead to major improvements in the quality of the information obtained by 3D ultrasound analysis.

In conclusion, examination of the fetal face is an integral part of fetal ultrasound examination during pregnancy, whether in a screening setting or during targeted analysis. The detection of fetal facial anomalies can lead to the diagnosis of chromosomal anomalies or various polymalformative states. Isolated fetal facial anomalies can require specific perinatal measures or lead to medical termination of pregnancy in cases of severe and complex defects. The use of 3D and 4D ultrasound imaging allows easier and more rapid diagnosis in the screening setting, and a more precise evaluation of facial features in the analysis setting.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We thank Sylvie Carillo for dedicated management of the database, Christine Nunez Moreno and Sophie Rodange for typing the manuscript, and Sharon Deslignières for linguistic review of the manuscript.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References