Prenatal Ultrasound Screening for Corpus Callosum Anomalies: A Narrative Review

1. Introduction

The corpus callosum (CC) is the largest commissure connecting the two cerebral hemispheres, and consists of decussating interhemispheric axons that permit learning and memory to be shared between the cerebral hemispheres [1]. The CC consists of four components, namely, rostrum, genu, corpus and splenium [2].

It is feasible to visualise the CC on prenatal ultrasound in a mid-sagittal view of the fetal brain, as a narrow anechoic area, delineated superiorly and inferiorly by 2 echogenic lines, overlaying the quadrigeminal plate of the mesencephalon [3]. The CC forms the roof of the frontal horns, cavum septi pellucidi (CSP), and ventricular bodies [4]. The frontal horns, CSP, and third ventricle are imaged as anechoic spaces deep to the CC [5]. The pericallosal sulcus (and its vessels) is seen as a thin, hyperechoic line that separates the corpus callosum from the cingulate gyrus [5]. The latter appears as a broaden hypoechoic band that is cocurvilinear with the CC and lies one level more superficial [5].

Prenatal detection of agenesis of CC (ACC) or other CC anomalies is important as they are among the most common malformations affecting the central nervous system (CNS) with a reported prevalence ranging from 3 to 7 per 1,000 births [6]. Besides, they have a strong association with chromosomal abnormalities, genetic syndromes, and neurodevelopmental delay [7]. Even isolated partial ACC is associated with mild to severe disabilities including speech disorders, behaviour and motor deficit disorders [8]. Anomalies of the CC can occur in isolation or be associated with other cerebral, structural, and genetic abnormalities [9,10]. A recent study showed that 43% of prenatally diagnosed ACC had a pathogenic/likely pathogenic variant identified on exome sequencing following negative chromosomal microarray analysis [10].

Prenatal diagnosis of the CC anomalies is a challenge and can cause anxiety to the parents [11,12,13]. First, although a correct diagnosis of anomalies of the CC requires direct examination of the CC in the mid-sagittal plane of fetal brain, the International Society of Ultrasound in obstetrics and Gynecology (ISUOG) guidelines on mid-trimester morphology scan 2021, and on screening examination of the fetal CNS 2020 do not recommend such direct examination of the CC in low-risk populations because of the associated technical difficulties [14,15]. Instead, ISUOG recommends three standard axial views of the fetal brain which can be easily obtained for screening examination [14,15], but the CC, being ‘comma’ shape’, is not well visualized in these axial views. Anomalies of the CC may be suspected only if indirect signs are present and noted in these standard axial views [14,15]. However, a review of twelve studies involving 544 fetuses with suspected anomalies of CNS showed that the pooled sensitivity of prenatal ultrasound diagnosis of ACC was 0.72 [16]. To improve the prenatal diagnosis of partial ACC, examination of the anterior and posterior complex has been proposed since 2015 [17]. Furthermore, a step-by-step approach has recently been recommended to get the most from the axial views of the fetal brain to improve the detection of indirect signs of callosal anomalies, among other brain anomalies [18].

Second, although it is feasible to visualise the CC on prenatal two-dimensional (2D) ultrasound in a mid-sagittal view of the fetal brain, it may be difficult to do so because of unfavourable fetal head position, calcification of the cranium, fetal movements, and thick abdominal wall [19,20]. To overcome these difficulties, using three-dimensional (3D) ultrasound reconstruction of the mid-sagittal plane has been proposed as an alternative for the evaluation of CC [19,20,21].

Third, a review of fetal pathologies of the CC reported in the prenatal imaging literature over a period of almost 30 years showed large heterogeneity in the definitions and terminology used to describe similar pathologies [22]. To achieve a more standardized classification system with agreed criteria, Solomon et al. proposed a simple classification: complete ACC, partial ACC, or dysraphic CC [23]. The absence of all four segments of the CC is called complete ACC whereas the absence of one or more of the four segments is called partial ACC [23,24]. It is important to avoid using ccomplicated ultrasound descriptions and simple sub-threshold biometry because they can generate considerable anxiety experienced by the parents-to-be of a fetus with a suspected or confirmed CC anomaly [23].

To improve the prenatal screening for CC anomalies, routine direct assessment of the CC with the mid-sagittal view has been recommended by several international experts in a recent consensus statement [25]. The implementation of such routine direct assessment is not easy given the difficulties encountered in obtaining the mid-sagittal view of the CC [19,20]. As such, it is the time to revisit the various two-dimensional (2D) and three-dimensional (3D) ultrasound techniques with a view to improve visualisation of the CC.

This is a narrative review. Databases including PubMed and Google were searched by using keywords including callosal dysgenesis; corpus callosum; obstetric ultrasound; screening; second trimester; sonography; three-dimensional. Relevant articles, including guidelines, observational studies, and reviews published from 1 January 2006 to 1 April 2026, were included. The aim of this review article is to discuss the use of various prenatal ultrasound screening methods of callosal anomalies including standard axial views, a more detailed axial views, the mid-sagittal view, transvaginal approach, and 3D reconstruction. New insights on screening methods are also shared.

2. Standard Approach: Axial Views

According the ISUOG Guidelines [14,15], standard scanning planes for the basic examination of the fetal brain include three axial planes, namely, the trans-ventricular and trans-thalamic planes, and trans-cerebellar plane. Lateral ventricles, CSP, midline falx, and thalami should be evaluated [14] to look for “indirect” signs of callosal anomalies.

These indirect signs include an abnormal or absent CSP, dilatation, or a tear-shaped appearance and malalignment of the lateral ventricles, a high-riding and dilated third ventricle (between right and left thalami), and a widened interhemispheric fissure [24,26,27]. If any of these signs is found, a targeted fetal neurosonographic examination should be performed [15].

However, there are limitations of using these indirect signs to detect ACC. In a study of 118 cases of partial ACC, prenatal ultrasound revealed the presence of abnormal CSP, distention of the interhemispheric fissure, dilated and elevated third ventricle, and ventriculomegaly in 86.4%, 77.1%, 47.4%, and 35.6% of cases, respectively, in the transverse plane of the brain [28].

Furthermore, these indirect signs can be subtle and sometimes not present at all before 24 weeks’ gestation, particularly in the diagnosis of partial ACC [24,29,30]. In particular, a square-shaped CSP, or tear-shaped ventricles may not be obvious to an inexperienced sonographer [31]. Besides, a high-riding and dilated third ventricle can be mistaken as CSP [24]. Furthermore, a third of examinations in fetuses with partial ACC may not show any abnormality in transventricular screening view < 24 weeks [24]. In particular, no ventriculomegaly or absence of the CSP was found [24]. Thus, the diagnosis of ACC may be missed at mid-trimester screening ultrasound [24].

According to the United Kingdom 20-week screening scan standard protocol [32], American Institute of Ultrasound in Medicine (AIUM) standard protocol [33], Australian Society of ultrasound in Medicine (ASUM) guidelines [34], ISUOG guidelines [14], Asean Oceana Federation of Obstetrics and Gynaecology (AOFOG) guidelines [35], all of these professional organisations do not recommend directly assessing the CC through the mid-sagittal plane of the fetal head as a screening for the CC anomalies in low-risk populations. However, targeted neurosonographic examination (of the CC, among other brain structures) should be performed in high-risk populations with scan findings suspicion of CNS anomaly, a family history of inheritable CNS malformations, a previous pregnancy affected by a fetal CNS anomaly, a present pregnancy exposure to teratogens known to affect neurogenesis, and in the presence of fetal heart defect, infection or genetic disorder [15].Complex or difficult cesareans can be divided into four categories [6].

3. A More Detailed Axial Views than the Standard Approach

3.1. Anterior and Posterior Complex

Given the limitations of standard axial views to detect CC anomalies, evaluation of the anterior complex (AC) and the posterior complex (PC) during routine basic ultrasound examination of the fetal brain with axial views has been proposed to improve the prenatal detection of CC anomalies and other midline anomalies [F17,36] [R6–8]. In the standard trans-ventricular plane, the anterior complex can be visualised, comprising the group of structures including, from anterior to posterior, the inter-hemisphere fissure (IHF), the pericallosal sulcus, the genu of the CC, the CSP and, laterally, the anterior horns of the lateral ventricles [17]. Above the trans-ventricular plane, the posterior complex can be visualised, comprising, from anterior to posterior, the splenium of the CC, the pericallosal sulcus, the IHF, the parieto-occipital fissure and, laterally, the medial wall of the lateral ventricles [17].

In addition to the ‘indirect’ signs which can be observed in the standard axial views including abnormal or absent CSP, dilatation, or a tear-shaped appearance and malalignment of the lateral ventricles, a high-riding and dilated third ventricle, and a widened IHF [24,26,27], abnormalities of the pericallosal sulcus, the genu and the splenium of the CC, anterior horns and medial wall of the lateral ventricle (a comma or triangular shape) can be assessed in more details in these axial views [17].

3.2. A Step-by-Step Approach

A step-by-step approach has been proposed to provide a detailed anatomical description of the supratentorial fetal brain, to get most information, and to assist in detection of the CNS malformations including the CC anomalies using the axial views of the fetal brain during a mid-trimester screening assessment of the fetal brain [18]. This approach is based on a series of nine clinical questions. Of these nine questions, five are related to the detection of the CC anomalies [18].

First, is the IHF visible and undistorted, and does it separate the two cerebral hemispheres symmetrically? The presence of an echogenic mass in the IHF is a feature of lipoma of the CC [18].

Second, is the CSP visible, rectangular or trapezoidal in shape and anechoic? Is the length-to-width ratio of CSP≥ 1.5? Absence of CSP and a square-shaped CSP is a feature of complete ACC and partial ACC, respectively [18]. Hyperechogenic CSP is a feature of obliterated CSP or lipoma of the CC [18].

Third, are the anterior (or frontal) horns of the lateral ventricles visible, symmetrical and anechoic, and do their medial walls approach the cavity of the CSP? Displacement of the anterior horns from the midline is a feature of ACC [18].

Fourth, is the third ventricle visible, anechoic and of normal width? Upward displacement of the third ventricle is a feature of ACC [18].

Fifth, are the posterior (or occipital) horns of the lateral ventricles of normal width and anechoic, and do they contain a homogeneous hyperechogenic choroid plexus? Teardrop or dilated ventricle is a feature of ACC [18].

However, more time, effort and skills are required to perform such detailed examinations. Although a more detailed examination including evaluation of the anterior complex and posterior complex by an expert can improve the prenatal diagnosis of CC anomalies, there are still limitations and are uncertainties whether a non-expert can identify abnormal findings as an expert [17]. 3. A more detailed axial views than the standard approach

4. Direct Visualization of the CC as Part of Routine Screening

4.1. Mid-Sagittal View

According to a recent consensus statement by a panel of international experts, a sagittal view of the brain should be obtained as part of routine screening to ensure comprehensive evaluation (72.4% agreement among the experts) [25]. This was also supported by national societies [37]. Routine direct visualization of the CC has the potential to improve the detection rate of callosal dysgenesis [38]. Although cases of complete ACC or large partial ACC are more easily detected by standard axial views of the fetal brain, cases with a small partial ACC might easily be missed [39]. Although the present international guidelines do not recommend routine CC visualization in low-risk pregnant women [14,32,33,34,35], routine CC visualization is performed by many professionals who practice prenatal ultrasound [39].

The gold standard to visualize the CC by fetal ultrasound is the mid-sagittal plane which can allow visualization of its entire length with four components (the rostrum, genu, body, and splenium) [38]. On the other hand, the CC cannot be well visualized in the standard axial views of the fetal brain because of its ‘comma’ shape. Only a small portion of it can be visualised in coronal planes at a time. Coronal sonograms can show the genu of CC that crosses the midline above the CSP and the two, closely apposed frontal horns [5].

The mid-sagittal view of the CC can be obtained using either the transvaginal or the transfundal approach. Whatever the approach, proper alignment of the probe along the correct section planes usually requires gentle manipulation of the fetal head using the free hand [39]. In fetuses in breech presentation, a transfundal approach is used, positioning the probe on the uterine fundus, parallel instead of perpendicular to the abdomen [39].

4.2. The Quality of CC Visualization

The quality of CC visualization is affected by many factors including fetal lie and head position, maternal body mass index (BMI), and the gestational age [38]. As the fetal position is often variable throughout an ultrasound examination, an opportunity to examine the CC should be taken when the fetal position is favourable.

There are three approaches. The first is the anterior-posterior (AP) approach which is commonly used [40]. After the fetal profile is examined in the standard sagittal view, the transducer is then angled towards the anterior fontanelle which is used as an acoustic window to demonstrate the CC [19,38]. Fine side-to-side movements may be required to achieve an ideal image of the CC [19]. The complete visualisation rate was 68% by the AP approach because the splenium is particularly difficult to include in this position [38].

The second is the cranio-caudal approach which utilizes the craniocaudal sagittal non-ossified suture window [40]. Although such approach is less commonly used than the AP approach, such approach is the most successful in obtaining complete visualization of the CC with a success rate of 90% [38,40].

The third is the posterior anterior (PA) approach whereby the posterior fontanelle is used as an acoustic window to demonstrate the CC [40]. The complete visualisation rate was 67% by the PA approach because of shadowing from the occipital bone unless the CC was imaged directly through a posterior suture [38,40].

Alternatively, a coronal section of the anterior horns of the lateral ventricles and the CSP is first obtained through the anterior fontanelle with the CSP oriented as close to vertically as possible, and the anterior horns on the same horizontal level. The transducer is then rotated 90◦ to obtain the mid-sagittal plane of the CC [19].

Image quality of the fetal CC can be adversely affected in obese pregnant women with BMI ≥30.0 kg/m2 due to ultrasound beam attenuation caused by thick abdominal wall adipose tissue [40]. The image quality can be optimised by using a low-frequency transducer (1.5–2 MHz) to increase penetration, using transvaginal approach when the fetus is in the cephalic position, and alternating the woman’s position or scanning through the maternal umbilicus or suprapubic area or franks to reduce the distance from the transducer to the area of interest [41]. Such skills and techniques can be developed with an appropriate training [40].

The optimal period for ultrasound examination of the fetal CC is after 20 weeks’ gestation [40]. The odds of non-visualization of the CC were 2.6 times higher before 20 weeks’ gestation (7.4%) than at or after 20 weeks’ gestation (3.0%) [40]. The CC develops between 12 and 16 weeks’ gestation, with all components formed by 18 weeks [42].

In a previous study, the rate of complete visualization of the corpus callosum could be increased from 23% to 71% after intensive training including didactic lecture on neuroanatomy and pathology and a one-on-one clinical training session on probe manipulation [38,43]. In another study, complete CC visualization significantly improved from 71.3% to 92.1% after initial protocol implementation and training [40]. In another study, direct visualization of the CC in the sagittal plane was achieved in 95% of cases from a cohort of 38,922 examinations [29]. The high rate of visualization may be attributed to improved scan technique over time, as well as improving ultrasound technology with better resolution [40].

4.3. CC Biometry and Doppler

Care should be taken in using corpus callosal biometry (length or thickness) to diagnose callosal anomalies, since a short, thin or thick corpus callosum does not necessarily mean anomalies [39]. Measurements of the CC are not mandatory for routine screening [25] because perfectly designed standards of CC biometry have not been produced and threshold values consistent with the rarity of anomalies have not been implemented [23]. Thus, a qualitative assessment (whether all its four components are present and normal) is much more important than a quantitative one (biometry) [39]. After birth, the CC is considered to be normally developed if its four components are present [23]. On the other hand, when there is an objective or subjective suspicion of an anomaly in the shape, length or thickness of the CC, referral to a specialist is recommended to ensure proper diagnosis and management [25].

There are two potential markers of partial ACC. The first marker is the ratio between the CC length and the internal cranial occipitofrontal dimension (ICOFD) in the mid-sagittal plane [39]. A previous study showed that such ratio was constant throughout the pregnancy in normal populations (2.35 ± 0.11) but significantly higher in pregnancies with CC anomalies (3.20 ± 0.84) [39]. Measuring this ratio may enable rapid evaluation of the CC without the need to refer to biometry tables [39].

Another potential marker is the distance between the distal part of the CC and the choroid plexus of the third ventricle which was relatively fixed in normal fetuses, but was reduced in fetuses with partial ACC [31]. Further studies are required to evaluate the usefulness of these two potential markers.

It is feasible to use power Doppler or microvascular flow imaging to demonstrate the pericallosal artery and its branches. However, its role is marginal in the assessment of the CC [39].

4.4. Disadvantages of Routine Examination of the CC

There are several disadvantages of routine examination of the CC using the mid-sagittal view. First, although routine direct visualization is feasible, obtaining the mid-sagittal view of the CC requires technical skill, and can be difficult [19]. The non-visualization rate of the splenium with or without other components of the CC varied from 7.9% to 29% even after intensive training [38]. Second, the mean time to perform the CC views was near one minute but was not reduced even after intensive training [38]. Third, implementation of routine visualisation of CC is difficult especially when resources or operator expertise are limited [25]. Fourth, mis-interpretation may lead to overdiagnosis and unnecessary parental anxiety. Thus, interpretation of the midsagittal view should be cautious, taking into account the full clinical context, local infrastructure and counselling capacities [25].

It is difficult to completely visualize CC, quickly and easily in a screening examination because of an unfavourable fetal position [38]. As time is important in a screening examination, prolonged time cannot be spent trying to get difficult views of the CC with the risk of paying less attention to other important structures. Alternatively, a more detailed examination of the fetal brain by axial planes including assessment of anterior and posterior complex, and the step-by-step approach as discussed above can be performed carefully to look for indirect signs of callosal anomalies [17,18,36]. Such alternative approach can avoid recalling the patient for assessment of the CC alone and the related costing. If routine sagittal CC view is to be implemented, audits and quality management of this examination are required [38].

4.5. Transvaginal Approach

Although the transvaginal approach is the preferred method to perform an adequate high-resolution targeted neurosonographic examination [44], systematic use of a transvaginal ultrasound approach is not necessary in standard screening practice [25]. When the fetus is in vertex presentation, and transabdominal views are inadequate or nondiagnostic, a transvaginal approach provides significant advantages over the transabdominal one including higher resolution, due to the higher emission frequency, and scanning through sagittal and coronal planes to avoid acoustic shadowing produced by the calvarium is circumvented [39]. In a previous study, a transvaginal approach was required in a small proportion (4.4%) of second trimester screening examinations [40].

5. Use of 3D Reconstruction

Although systematic use of a 3D ultrasound approach is not necessary in standard screening practice [25], 3D ultrasound reconstruction may be useful in selected cases when difficulties in obtaining sagittal view of the CC through 2D ultrasound are encountered. 3D ultrasound allows acquisition of volumes in the standard axial planes with subsequent multiplanar reconstruction of the midsagittal plane [44]. Furthermore, the quality of the 3D images of CC can be improved by multiplanar reconstruction techniques like VCI (Volume Contrast Imaging) mode which allows displaying thicker ‘slices’ of the CC, and enhancing the signal-to-background noise ratio with significant enhancement of image quality [39].

The extent to which the CC can be assessed in 3D reconstructed sagittal planes is entirely dependent on the plane of volume acquisition [45]. The 3D mid-sagittal reconstructed plane will likely be obtained if the 3D volume is acquired at the mid-sagittal plane with the angle between the transducer and the direction of the fetal nose ranging from 0◦ to 179◦ and from 330◦ to 359◦ [45]. Another study also showed that the best technique for visualization of the CC in a single image involved 3-D acquisition in a sagittal plane through the sagittal suture [44]. These are the planes at which the CC can be visualised by 2D ultrasound though the metopic suture, anterior fontanelle, sagittal suture or posterior fontanelle as discussed in the section 4. These results are expected as obtaining a good 2D view before volume acquisition is a pre-requisite to obtain good 3D images after reconstruction.

The 3D mid-sagittal reconstructed plane will also likely be obtained if the 3D volume is acquired at an oblique plane around the crown–rump axis with an angle from the mid-sagittal plane of less than 30◦ [45]. Thus, when 2D plane is still slightly oblique around the crown rump axis with an angle from but not exactly at the mid-sagittal plane despite appropriate manipulation, a 3D volume acquisition can be attempted to get a mid-sagittal reconstructed view subsequently. This may save time and effort in getting the mid-sagittal view of the CC through 2D ultrasound.

However, if 3D volume is acquired at the axial BPD plane [46], reconstruction of the midline sagittal plane will likely show a hyperechogenic artifact instead of the corpus callosum which should be anechogenic or hypoechogenic [F47] [R24]. Such hyperechogenic artifact can be confused with a hyperechogenic CC which is a feature of callosal lipoma [47]. In another study, 3D volume acquisition from axial planes at the level of BPD plane with an angle of 90 degrees to the midline are insufficient to evaluate the CC [48].

There are several disadvantages of such 3D approach. First, some sonologists may not be familiar with the use of 3D ultrasound and its related multiple processing techniques [49]. These techniques are not easy to learn and are time consuming in the absence of specific training [49]. Second, the quality of the 3D volume acquired depends on several factors including maternal obesity, depth of imaging, fetal lie, fetal movements, the angle and the speed of volume acquisition [49].At present, pre-operative ultrasonographic assessment is not a routine practice. There is lack of studies in this issue [5]. Without any preceding obstetric ultrasound, unexpected problems like PAS, placenta previa, severe pelvic adhesions, uterine fibroids, membranous fetal vessels, or large superficial vessels in the LUS, or breech presentation may be found during subsequent cesarean section [5,9,10,15,29,42]. Rarely, PAS can be found in women without any risk factors [10]. Given that cesarean section is a common obstetric procedure performed in an elective or an emergent situation, it is worthwhile to consider performing assessment of LUS and placenta localization during commonly indicated obstetric scan. More studies on this issue are required.

6. New Insights

In a previous study, using the posterior-anterior approach or anterior-posterior approach, complete visualization of the CC was achieved in only 67% or 68% of the studied pregnancy women [38]. Despite three of the four components of the CC were visualised, the splenium was not visualized in 13% of the studied pregnant women even after intensive training [38]. Efforts should be made to image the splenium which is commonly absent in partial ACC.

The splenium can be visualized when the posterior complex is examined [17]. In my experience, with an axial view at the level of posterior complex, rock the transducer to face the posterior fontanelle can show the splenium between the echogenic pericallosum sulcus and the anechoic cavum vergae. Similarly, with an axial view at the level of anterior complex, rock the transducer to face the anterior fontanelle can show the rostrum between the echogenic pericallosum sulcus and the anechoic CSP. In either of these two views, fan the transducer can show both the rostrum and the splenium simultaneously. Rotating the transducer 90◦ can then show the mid-sagittal plane of the CC.

If a 3D volume is acquired at either of these two tilted axial planes with both rostrum and splenium in view, a 3D reconstructed mid-sagittal view of the CC will be visualized as an anechogenic shadow. In a previous study, the CC visualization rate was 42% when the 3D volume was acquired at axial at the level of BPD plane with an angle of 0-290 from the midline [45]. Further studies are required to investigate whether these modified 2D or 3D approaches (as discussed above) in combination with other approaches can improve the visualisation rate of the CC in difficult situations.

Recently, a deep learning–based artificial intelligence (AI)–assisted diagnostic framework CC-FocusNet for the automated classification of CC developmental status has been investigated [50]. As the present screening examination is based on axial views of the fetal brain, and some indirect signs of the CC anomalies are too subtle to an experienced sonographer, it is worthwhile to investigate whether a deep learning machine can detect these subtle indirect signs by analysing a video clip of a slow axial sweep of the fetal brain.

7. Conclusions

In this review article, various prenatal ultrasound screening methods of CC anomalies including standard axial views, a more detailed axial views, the mid-sagittal view, transvaginal approach, and 3D reconstruction are discussed. Key features of these methods are summarised in Table 1. New insights on screening methods are also shared.