Thick fetal corpus callosum: an ominous sign?

Authors

  • T. Lerman-Sagie,

    Corresponding author
    1. Fetal Neurology Clinic, Wolfson Medical Center, Holon, Israel
    2. Pediatric Neurology Unit, Wolfson Medical Center, Holon, Israel
    3. affiliated to the Sackler School of Medicine, Tel-Aviv University, Tel-Aviv, Israel
    • Pediatric Neurology Unit, Wolfson Medical Center, 62 Halochamim St, Holon, Israel 58100.
    Search for more papers by this author
  • L. Ben-Sira,

    1. Pediatric Radiology Unit, Sourasky Medical Center, Tel-Aviv, Israel
    2. affiliated to the Sackler School of Medicine, Tel-Aviv University, Tel-Aviv, Israel
    Search for more papers by this author
  • R. Achiron,

    1. Department of Obstetrics and Gynecology, Sheba Medical Center, Ramat-Gan, Israel
    2. affiliated to the Sackler School of Medicine, Tel-Aviv University, Tel-Aviv, Israel
    Search for more papers by this author
  • L. Schreiber,

    1. Department of Pathology, Wolfson Medical Center, Holon, Israel
    2. affiliated to the Sackler School of Medicine, Tel-Aviv University, Tel-Aviv, Israel
    Search for more papers by this author
  • G. Hermann,

    1. Department of Pathology, Assaf Harofe Medical Center, Tzrifin, Israel
    2. affiliated to the Sackler School of Medicine, Tel-Aviv University, Tel-Aviv, Israel
    Search for more papers by this author
  • D. Lev,

    1. Fetal Neurology Clinic, Wolfson Medical Center, Holon, Israel
    2. Genetics Institute, Wolfson Medical Center, Holon, Israel
    3. affiliated to the Sackler School of Medicine, Tel-Aviv University, Tel-Aviv, Israel
    Search for more papers by this author
  • D. Kidron,

    1. Department of Pathology, Sapir Medical Center, Kfar-Saba, Israel
    2. affiliated to the Sackler School of Medicine, Tel-Aviv University, Tel-Aviv, Israel
    Search for more papers by this author
  • G. Malinger

    1. Fetal Neurology Clinic, Wolfson Medical Center, Holon, Israel
    2. Department of Obstetrics and Gynecology, Wolfson Medical Center, Holon, Israel
    3. affiliated to the Sackler School of Medicine, Tel-Aviv University, Tel-Aviv, Israel
    Search for more papers by this author

Abstract

Objective

Anomalies of the corpus callosum are frequently diagnosed during pregnancy, but a thick corpus callosum is a rare finding and its significance is not clear. We aimed to assess the significance of thick fetal corpus callosum by reviewing our experience of such cases.

Methods

The records of all fetuses with anomalies of the corpus callosum referred to the prenatal diagnosis units of two university hospitals from 2000 to 2007 were reviewed. Nine fetuses with a thick corpus callosum were identified.

Results

In all cases there were associated abnormalities: macrocephaly, ventriculomegaly, vermian agenesis, abnormal sulcation or encephalocele. Four pregnancies were terminated and in each of these cases the autopsy confirmed dysmorphic features and additional brain abnormalities. Five infants were delivered; two died shortly after birth, one suffers from mental retardation, one had neonatal convulsions and one is developing normally.

Conclusions

A thick fetal corpus callosum is usually associated with other brain anomalies and is part of a neurogenetic syndrome in most cases. Copyright © 2009 ISUOG. Published by John Wiley & Sons, Ltd.

Introduction

The corpus callosum is the largest commissure of the brain and develops at between 8 and 20 weeks of gestation. A disturbance of this process may lead to agenesis or partial agenesis (hypogenesis or dysgenesis) of the corpus callosum. While agenesis and dysgenesis of the corpus callosum are frequently described among patients evaluated for mental retardation, a thick corpus callosum is rarely mentioned. In a recent study of 142 cases with anomalies of the corpus callosum, 82 patients had agenesis of the corpus callosum, while 60 had hypogenesis of the corpus callosum1. An abnormally thick corpus callosum was not mentioned in any patient. In a study on magnetic resonance imaging (MRI) in the evaluation of fetuses referred for suspicion of abnormalities of the corpus callosum on ultrasound examination, a thick corpus callosum was never the cause of referral2. However, a literature review discloses a few case reports describing this anomaly3–6 and it seems that it is usually associated with other malformations. When associated with microcephaly it may be pathognomonic of Cohen syndrome7.

A thick corpus callosum has been linked to enlarged white matter and macrocephaly in patients with neurofibromatosis8 and has been described in patients with the macrocephaly–capillary malformation syndrome9.

In this study we describe the different types of abnormally thick corpus callosum, the associated findings and outcome of fetuses with this relatively rare anomaly of commissural development among all the fetuses referred to our units for suspected brain abnormalities in an 8-year period.

Methods

We reviewed the records of brain ultrasound examinations of all fetuses with anomalies of the corpus callosum diagnosed in two prenatal diagnosis units from January 2000 until December 2007. From a total of 182 callosal anomalies, we subjectively identified nine fetuses with an abnormally thick corpus callosum. This group of nine fetuses represents our study group; a summary of the ultrasound and MRI findings in each case is given in Table 1.

Table 1. Ultrasound and magnetic resonance imaging (MRI) findings and follow-up information in the nine fetuses with thick corpus callosum included in the study
FetusGACallosal involvementSize of lateral ventricles (mm)MacrocephalyOther abnormal findingsMRI findingsFollow-up
  1. —, not performed; GA, gestational age (weeks); TOP, termination of pregnancy; MCD, malformation of cortical development.

122 + 0Genu, bodyNormal, asymmetricNoPolyhydramniosPolyhydramniosNeonatal convulsions
222 + 6Genu, body10NoNoneTOP
323 + 2BodyNormalNoComplete vermian agenesisComplete vermian agenesisTOP
423 + 3Genu, bodyNormalYesNoneSevere mental retardation, blindness
526 + 2Body, spleniumNormalYesOverdeveloped sulcationMCDTOP
627 + 0Diffuse10.9–9.5, asymmetricYesOverdeveloped sulcationMCDNeonatal death
729 + 5Body, splenium12.8–10, asymmetricYesOverdeveloped sulcationMCDTOP
830 + 3Body11NoOccipital encephaloceleNeonatal death
930 + 5Genu, fornices12NoNoneNoNormal development at 2 years

We retrieved relevant data including maternal and family history and ultrasonographic reports and images of all the previous examinations. Our detailed neurosonographic evaluation utilized a combined transabdominal and transvaginal approach10. The ultrasonographic examinations were performed using an HDI 3000 (ATL, Seattle, WA, USA), a Logic 9 (GE Healthcare, Milwaukee, WI, USA), a Voluson 730 Expert (GE Healthcare) and/or a Voluson E8 (GE Healthcare).

The length of the entire corpus callosum and the thicknesses of the genu, body, and splenium were measured in the mid-sagittal plane as described by Rakic and Yakovlev11 and by Barkovich and Norman12. The corpus callosum length was measured from the most anterior part of the genu to the most posterior part of the splenium; the thickness of the genu and splenium were measured exactly at the same level, and the body was measured at the middle (Figure 1).

Figure 1.

Transvaginal ultrasound image showing the correct measurement of callosal length (equation image) and thickness (equation image) at the level of the genu (G), body (B) and splenium (S). As shown, calipers are correctly placed when touching the inner edges of the callosal sulcus and the inner echoes generated by the interface between the corpus callosum and the cavum septi pellucidi et vergae.

Increased thickness was defined in all or single parts of the corpus callosum as a thickness above +2 SD of the mean values described by Malinger and Zakut13. The shape and echogenicity were subjectively evaluated; the echogenicity was compared with the adjacent brain and graded as less echogenic than, equally echogenic to, or more echogenic than, the cingulate gyrus (normally the corpus callosum is equally echogenic to or less echogenic than the cingulate gyrus).

MRI was performed in six fetuses using a 1.5 T system (GE Healthcare). Following a localizing gradient-echo sequence, ultra T2-weighted single-shot fast-spin echo MR images were collected according to fetal position in the axial, coronal and sagittal planes (TR/TE, infinite/90; bandwidth 32 KHz; field of view, 16 × 28 cm; matrix, 256 × 192; slice thickness, 3–5 mm; gap 0–1 mm; number of excitations, 0.5). A torso phased array coil was used and an experienced pediatric radiologist analyzed the MR images.

In all cases the fetal neurology clinic team, which included specialists in obstetrics, pediatric neurology and genetics, gave counseling to the expectant parents regarding the possible significance and prognosis of the findings.

Four couples opted for termination of pregnancy, and two infants died shortly after delivery. Postnatal evaluation and brain imaging were performed in the three surviving infants.

Results

The presence of an abnormally thick fetal corpus callosum was identified at a mean gestational age of 26 + 2 (range 22 + 0 to 30 + 5) weeks. The patients were referred for consultation to our clinics because of abnormal fetal findings on ultrasound examination: suspected macrocephaly (n = 3); ventriculomegaly (n = 3); occipital encephalocele (n = 1); suspected corpus callosum agenesis (n = 1); and suspected vermian agenesis (n = 1). A positive family history of macrocephaly was identified in two patients. The patient histories were negative for any maternal diseases or pregnancy complications.

Neurosonographic findings

The length and thickness of the corpus callosum as measured at the index examination are presented in Figure 2.

Figure 2.

Corpus callosum length (a) and thickness at the level of the genu (b), body (c) and splenium (d) in nine fetuses with abnormally thick corpus callosum. The numbers in brackets represent the fetus as shown in Table 1. The mean ± 2 SD of normal values are represented by the solid and dashed lines, respectively, derived from the data of Malinger and Zakut13.

In four fetuses, two of which had macrocephaly, the corpus callosum was longer than expected, in one fetus the corpus callosum was considered to be abnormally short and malformed, and the length was considered normal in the remaining four fetuses.

The corpus callosum was uniformly thick in only one fetus (Figure 3), the genu and body were thick in three (Figure 4), the body and splenium in two (Figure 5), only the body in two (Figure 6) and only the genu in one (Figure 7). The fetus with the thick genu also had abnormally thick fornices. Using color Doppler ultrasonography, the pericallosal arteries were found to be present and normal in all the fetuses, and the callosal sulcus was clearly distinguishable between the corpus callosum and the cingulate gyrus. In Fetus 6, color Doppler imaging of the pericallosal arteries helped to clearly differentiate between the cingulate gyrus and a structure composed of aberrant cingulum, the indusium griseum and the corpus callosum.

Figure 3.

Fetus 6. (a) Transvaginal ultrasound midline sagittal image at 27 weeks' gestation. The corpus callosum is uniformly thick. Sulcation is overdeveloped and corresponds to a gestational age of more than 30 weeks. The third ventricle (3v) is dilated. (b) Transvaginal ultrasound image in the coronal plane at the same gestational age, showing the thick corpus callosum and the abnormal shape of the lateral ventricles. (c) Midline sagittal magnetic resonance image at 27 weeks' gestation, demonstrating thick corpus callosum with abnormal acute angle, and two different intensities in T2-weighted images: a thin hypointense lower band and a thick upper band isointense with the cingulate gyrus. (d) Macroscopic photograph of sagittal dissection of the brain following neonatal death after delivery at 34 weeks, showing the unusually thick corpus callosum, large third ventricle (3V) and overdeveloped sulcation. (e) Microscopic coronal cut specimen, showing that the structure identified as the corpus callosum by imaging was actually composed of a thick corpus callosum (CC) and heterotopic tissue originating from the cingulate gyri (HC). Note the position of the indusium griseum, the remnant gray matter normally found dorsal to the corpus callosum (arrows) between these structures. The pericallosal artery (PA) is normally positioned in the callosal sulcus. (H&E stain).

Figure 4.

Fetus 2. Transvaginal ultrasound image in the midline sagittal plane at 22 + 6 weeks' gestation, showing thickening of the genu and body of the corpus callosum. The corpus callosum is more echogenic than is the adjacent brain tissue.

Figure 5.

Fetus 7. (a) Transvaginal ultrasound image in the midline sagittal plane at 30 + 4 weeks' gestation, showing the thick body and splenium of the corpus callosum. Note the partial obliteration of the cavum septi pellucidi with a hyperechogenic focus in the center. The obliteration of the cavum septi pellucidi makes it difficult to differentiate between the inferior portion of the genu and the anterior body of the corpus callosum and to measure the thickness of the genu. (b) Midline sagittal magnetic resonance image at 30 weeks' gestation. The corpus callosum is difficult to differentiate from the cingulate gyrus and appears to be thickened throughout its length, though predominantly in its posterior part (5 mm).

Figure 6.

Fetus 3. (a) Transvaginal ultrasound image in the midline sagittal plane at 23 + 2 weeks' gestation, showing that the body of the corpus callosum is thicker than usual and is of similar echogenicity to the adjacent brain tissue. (b) Macroscopic photograph of dissected brain in a similar plane, following termination of pregnancy at 28 weeks, demonstrating the thick corpus callosum.

Figure 7.

Fetus 9. Transvaginal ultrasound image in the midline sagittal plane at 30 + 5 weeks' gestation. The genu of the corpus callosum is unusually thick and echogenic. A very prominent fornix connects the rostrum to the splenium.

The corpus callosum was more echogenic than the cingulate in two fetuses (Figure 4), similarly echogenic in six fetuses (Figures 3, 5, 6 and 7) and less echogenic in only one fetus.

All fetuses had associated central nervous system (CNS) findings: mild ventriculomegaly in five (in two it was unilateral); asymmetric ventricles without enlargement in one; the shape of the anterior horns was considered abnormal in one; macrocephaly in four, in three of which sulcation was prematurely developed for gestational age giving the impression of overdeveloped sulci; complete vermian agenesis in one and occipital encephalocele in one (Table 1).

Magnetic resonance imaging findings

MRI was performed in six fetuses and an abnormally thick corpus callosum was identified in all. The corpus callosum was uniformly thick in three fetuses (Figure 3), only the genu was thick in one, and only the splenium and posterior body in two (Figure 5).

Associated findings were noted in all fetuses: mild ventriculomegaly in four (in two it was unilateral); malformations of cortical development were suspected in three fetuses and vermian agenesis in one fetus.

In one fetus there were different signal intensities on T2-weighted imaging within the corpus callosum, a thin hypointense inferior band and a thick isointense band inseparable from the overlying cingulate gyrus (Figure 3).

Patient follow-up

Four women opted for termination of pregnancy and abnormal CNS findings were confirmed by autopsy in three fetuses; in one fetus the brain was not suitable for full examination but the weight was consistent with severe megalencephaly.

Five infants were delivered, two of which died in the early postnatal period. One of these infants showed a large occipital meningoencephalocele, but no autopsy was performed. The other, with macrocephaly and suspected malformation of cortical development, was born at 34 weeks' gestation and died unexpectedly. The autopsy findings were consistent with acute and chronic chorioamnionitis; the pancreas showed signs of inflammation, and the brain findings were of a macroscopically abnormally developed cortical pattern, thick corpus callosum with adjacent supracallosal white matter fibers (possibly heterotopic cingulum) separated by the indusium griseum, bilateral ventriculomegaly with bulging germinal matrix and enlarged third ventricle (Figure 3).

One of the infants that survived the neonatal period suffers from severe mental retardation and blindness. Postnatal MRI showed an abnormally thick and short corpus callosum with two different intensities on T2-weighted imaging and an inferior irregular border, bilateral optic nerve hypoplasia and a small coloboma of the left eye. Another infant developed neonatal convulsions that required antiepileptic treatment, but was later lost to follow-up. At the time of writing, one infant was developing normally at the age of 24 months.

Discussion

A thick corpus callosum is an infrequently described finding on brain imaging. Only single case reports have been published and in these reports it was invariably associated with additional brain abnormalities3–6. We identified a thick corpus callosum in 5% of fetuses with abnormalities of this structure. It is difficult to explain the discrepancy between the relatively high prevalence among our population and that of other pre- and postnatal series1, 2. Possible explanations include: the fact that evaluation of the corpus callosum itself is not a part of routine fetal ultrasound examination and the diagnosis of callosal anomalies is usually made based on the presence of indirect signs such as abnormal ventricular configuration, subjective assessment of the thickness of the corpus callosum without comparison to published nomograms, failure to consider increased thickness as pathological, definition of the study population as solely including fetuses with agenesis/hypogenesis of the corpus callosum1, 2, a postnatal decrease in the thickness of a malformed corpus callosum and early death of patients before radiological evaluation.

Increased thickness was associated with macrocephaly in four of our cases. This association can be explained by the existence, in some fetuses with macrocephaly, of an increased amount of white matter tracts that cross through the corpus callosum. This explanation has been suggested in patients with neurofibromatosis Type 18. A transient overproduction of axons in the corpus callosum during fetal development has been demonstrated in mammals14, which is due, in part, to a transient excess of neurons that send an axon through the corpus callosum. The excessive axons are later eliminated15. It is possible that in cases with macrocephaly the decreased apoptosis of neurons causes decreased elimination of crossing axons, thus resulting in a thickened corpus callosum.

Three of our four fetuses with macrocephaly had a similar pattern of radiological findings: abnormal cortical development—most prominent in the perisylvian region—suggestive of polymicrogyria, abnormal shape of the frontal horns and a thick corpus callosum with two different intensities on T2-weighted imaging. In the only fetus that underwent an autopsy, pathology demonstrated that the thickening of the corpus callosum was due to abnormal pericallosal white matter fibers continuous with the corpus callosum. Similar radiographic features were described in a fetus by Rypens et al.3 but this fetus was not macrocephalic. An analogous pathological pattern of supracallosal longitudinal fiber bundles was described by Hori and Stan16 in a patient with severe mental retardation, spasticity and microcephaly, and also as an incidental finding in another adult brain. According to Hori and Stan16 these fibers may represent an aberrant cingulum; while the normal cingulum is positioned just beneath the subcortical white matter of the gyrus cinguli, the aberrant cingulum is found more medially along the dorsal corpus callosum. In the same study the authors stated that supracallosal gray matter should be excluded from the differential diagnosis as it may represent an incomplete form of cleavage disturbance such as holoprosencephaly or a hyperplastic indusium griseum.

The association of macrocephaly and thick corpus callosum has recently been described in the macrocephaly–capillary malformation syndrome9. The authors showed increased thickness of the whole corpus callosum in seven out of 16 patients and increased thickness in part of the corpus callosum in an additional seven. Other radiological findings are frequent in this syndrome, including focal cortical dysplasia, polymicrogyria that primarily involves the perisylvian and insular regions, and cerebral and/or cerebellar asymmetric overgrowth.

It is possible that our cases with the combination of thick corpus callosum, macrocephaly and abnormal cortical development had macrocephaly–capillary malformation syndrome. However, it is more likely that they represent a new syndrome, since the thick corpus callosum in our cases was already present in the prenatal period, while the thickening of the corpus callosum was a progressive postnatal finding in patients with macrocephaly–capillary malformation syndrome9 and did not demonstrate different signal intensities on MRI.

None of our cases demonstrated microcephaly as seen in Cohen syndrome, which is the only syndrome with a consistent description of isolated enlarged corpus callosum7. This is quite surprising considering that Cohen syndrome is relatively prevalent in the Ashkenazi Jewish population. All our cases were referred owing to the additional findings and thus cases of isolated thickened corpus callosum would have escaped our detection.

The prognosis in 8/9 fetuses was poor based on the presence of associated brain malformations as demonstrated by ultrasound and MRI (which were ultimately confirmed at autopsy). Of the five infants delivered, early neonatal demise occurred in two, and two others suffered from neonatal seizures and severe mental retardation. Only one infant was developing normally at the time of writing. This fetus had a normal head circumference, the increased thickness was borderline and confined to the genu and fornices, and there were no associated cortical abnormalities.

We cannot exclude the possibility that we diagnosed only the more severe cases with thick corpus callosum, i.e. those with associated CNS anomalies, thus it is possible that we missed the more benign cases. Therefore, we can cautiously conclude that the prognosis of thick corpus callosum with abnormal head circumference and/or associated malformations is poor, but an isolated finding may be associated with a better outcome.

The frequent association of a thick corpus callosum with other brain malformations is not surprising since neuronal migration and development of the corpus callosum occur at approximately the same time.

In one of our cases we observed abnormally increased echogenicity of the corpus callosum. Hyperechogenicity of the corpus callosum is characteristic of pericallosal lipoma, however the hyperechogenicity in our case was completely different. We hypothesize that the increased echogenicity could have been caused by an increased number of aberrant fibers or associated edema.

In conclusion, it is possible and important to diagnose a thick corpus callosum in utero since most cases carry a poor prognosis. The diagnosis was implied by the associated anomalies and reached after the performance of dedicated neurosonography and MRI. A thick corpus callosum may be a radiological feature of different neurogenetic syndromes. The significance of an isolated thick corpus callosum remains unknown.

Ancillary