Prenatal diagnosis of rhombencephalosynapsis: neuroimaging features and severity of vermian anomaly

K. Krajden Haratz, P. Oliveira Szejnfeld*, M. Govindaswamy, Z. Leibovitz, L. Gindes, M. Severino, A. Rossi, D. Paladini, R. Garcia Rodriguez, L. Ben-Sira, T. Borkowski Tillman, R. Gupta, G. Lotem, N. Raz, T. E.N.K. Hamamoto, D. Kidron, A. Arad, R. Birnbaum, M. Brussilov, L. PomarY. Vial, R. J. Leventer, G. McGillivray, M. Fink, W. Krzeszowski, A. Fernandes Moron, D. Lev, M. Tamarkin, J. Shalev, J. Har Toov, T. Lerman-Sagie, G. Malinger

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


Objectives: To describe the prenatal neuroimaging spectrum of rhombencephalosynapsis (RES) and criteria for its classification according to the severity of vermian anomaly. Methods: In this multicenter retrospective study of fetuses with RES between 2002 and 2020, the medical records and brain ultrasound and magnetic resonance images were evaluated comprehensively to determine the severity of the vermian anomaly and the presence of associated brain findings. RES was classified, according to the pattern of vermian agenesis and the extent of the fusion of the hemispheres, as complete RES (complete absence of the vermis) or partial RES (further classified according to the part of the vermis that was missing and, consequently, the region of hemispheric fusion, as anterior, posterior, severe or mixed RES). Findings were compared between cases with complete and those with partial RES. Results: Included in the study were 62 fetuses with a gestational age ranging between 12 and 37 weeks. Most had complete absence of the vermis (complete RES, 77.4% of cases), a ‘round-shaped’ cerebellum on axial views (72.6%) and a transverse cerebellar diameter (TCD) < 3rd centile (87.1%). Among the 22.6% of cases with partial RES, 6.5% were classified as severe partial, 6.5% as partial anterior, 8.1% as partial mixed and 1.6% as partial posterior. Half of these cases presented with normal or nearly normal cerebellar morphology and 28.5% had a TCD within the normal limits. Infratentorially, the fourth ventricle was abnormal in 88.7% of cases overall, and anomalies of the midbrain and pons were frequent (93.5% and 77.4%, respectively). Ventriculomegaly was observed in 80.6% of all cases, being more severe in cases with complete RES than in those with partial RES, with high rates of parenchymal and septal disruption. Conclusions: This study provides prenatal neuroimaging criteria for the diagnosis and classification of RES, and identification of related features, using ultrasound and magnetic resonance imaging. According to our findings, a diagnosis of RES should be considered in fetuses with a small TCD (severe cerebellar hypoplasia) and/or a round-shaped cerebellum on axial views, during the second or third trimester, especially when associated with ventriculomegaly. Partial RES is more common than previously thought, but presents an extreme diagnostic challenge, especially in cases with normal or nearly-normal cerebellar morphobiometric features.

Original languageEnglish
Pages (from-to)864-874
Number of pages11
JournalUltrasound in Obstetrics and Gynecology
Issue number6
StatePublished - Dec 2021


FundersFunder number
Department of Neurogenetics
Division of Pediatric Radiology of Johns Hopkins Hospital
Kennedy Krieger Institute


    • aqueductal stenosis
    • cerebellum
    • fetus
    • prenatal diagnosis
    • rhombencephalosynapsis
    • vermis


    Dive into the research topics of 'Prenatal diagnosis of rhombencephalosynapsis: neuroimaging features and severity of vermian anomaly'. Together they form a unique fingerprint.

    Cite this