Prenatal ultrasound and magnetic resonance features in a fetus with Walker-Warburg syndrome. Letters to the editor

Prenatal ultrasound and magnetic resonance imaging features in a fetus with Walker–Warburg syndrome A 27-year-old woman was referred for targeted ultrasound examination at 26 weeks’ gestation following sonographic detection of fetal ventriculomegaly and posterior fossa abnormality. She and the father were Macedonian, non-consanguineous, healthy and had two healthy daughters. A mid-trimester abnormality scan had not been performed. On ultrasound examination the fetal cerebral ventricles were found to be enlarged (atrial width, 14 mm), the cerebellar vermis was not depicted and the fourth ventricle communicated with the cisterna magna, which was not enlarged (Figure 1). A median anechoic structure was interposed between the lateral ventricles, suggestive of a vein of Galen aneurysmal malformation. However, this diagnosis was easily excluded because the structure showed no signal on color Doppler imaging. Two days later, magnetic resonance imaging (MRI) confirmed the sonographic findings. Moreover, it showed a kinked Z-shaped brainstem, with bifid pons and medulla oblongata (Figure 2), and the mantle was thin with a simplified gyral pattern. The association of lissencephaly with cerebellar and brainstem anomalies suggested the diagnosis of Walker–Warburg syndrome (WWS). WWS is a rare autosomal recessive disorder characterized by congenital muscle dystrophy (CMD) and complex brain and eye abnormalities1,2. Additional ultrasound and MRI examinations showed progressive enlargement of the cerebral ventricles, the median anechoic structure and the cisterna magna (Figure 1c). At 31 weeks, retinal nonattachment/ detachment and asymmetry of the eye globes were observed (Figure 3). At 41 weeks a male fetus was vaginally delivered, with a birth weight of 3400 g and head circumference of 37.5 cm. The neonate showed spontaneous respiratory activity, but was deeply hypotonic and required gavage nutrition for impaired swallowing. Postnatal MRI showed active hydrocephalus (necessitating ventriculoperitoneal shunt placement), the typical ‘cobblestone’ appearance of lissencephaly and abnormal cerebellar gyration (Figure 4). Muscular biopsy on day 15 showed severely increased variability of the diameter of muscle fibers and endomisial fibrosis together with immunohistochemical reduction of glycosylated alpha-dystroglycan2. Molecular analysis showed a homozygous mutation in the protein-Omannosyltransferase 1 (POMT1) gene (c.1611C>G, p.Ser537Arg), which has previously been detected in other cases of WWS3,4. The infant died at 6 months of age. In fetuses postnatally proven to be affected by WWS, the cerebral anomalies detected by prenatal sonography are usually non-specific5 and suggestive of WWS only in cases with a positive family history. When familial history is uninformative, only the association of cerebral or cerebellar with ocular anomalies can suggest the correct diagnosis6,7. In the current case, sonography showed ventriculomegaly and dysplastic cerebellum, and the diagnosis of WWS was suspected after the detection of lissencephaly, bifid pons and medulla oblongata and kinked brainstem by MRI8,9. The ocular abnormalities, which confirmed the diagnosis6,7,9, were only observed 1 month later. Late-onset microphthalmia and late diagnosis of retinal detachment have been described previously5,10. Aside from the findings that were used to diagnose WWS, listed above, the most peculiar sonographic finding at presentation was the midline oblong anechoic structure. This feature was explained when MRI showed separation of the bifid portions of the brainstem, together with its downward displacement (Figure 2). Figure 1 Oblique axial (a) and axial (b) ultrasound images of the fetal head at 26 weeks’ gestation, showing enlargement of the lateral ventricles and the third ventricle (*). The cerebellar lobes (arrows) were separated, with no detectable vermian structure. The dotted arrow indicates the retropulvinar cistern. (c) Axial ultrasound image at 35 weeks, showing widening of the midline anechoic structures. Copyright  2009 ISUOG. Published by John Wiley & Sons, Ltd. LETTERS TO THE EDITOR 364 Letters to the Editor Figure 2 Single shot fast spin echo T2-weighted magnetic resonance images of the fetal head at 26 weeks. (a,b) Axial planes at different levels showing separated cerebellar lobes (solid arrows), bifid pons (arrowhead), third ventricle (*) and retropulvinar cistern (dotted arrow). (c) Sagittal view showing the kinked brainstem (box). The line indicates the plane at which the axial images shown in (b) and in Figures 1b and 1c were obtained. Figure 3 Fetal retinal non-attachment at 31 weeks, imaged using sonography, seen as a conical structure in the sagittal plane (a) and as concentric circles in the coronal plane (b). Figure 4 Magnetic resonance imaging on day 6 of postnatal life, showing severe enlargement of the lateral ventricles and the cisterna magna, together with the typical ‘cobblestone’ appearance of the cortical mantle (a) and abnormal cerebellar gyral pattern (b). The resulting enlarged retropulvinar cistern appeared in continuation with the enlarged third ventricle anteriorly and with the communication between the fourth ventricle and the cisterna magna posteriorly. We suggest that this characteristic might represent an indirect sign of maldevelopment of the brainstem. As it was detectable at an earlier gestational age than ocular anomalies, it may offer an earlier sonographic clue leading to suspicion of the diagnosis of WWS.