Quantitative fetal magnetic resonance imaging assessment of cystic posterior fossa malformations

Ultrasound Obstet Gynecol. 2020 Jul;56(1):78-85. doi: 10.1002/uog.21890.

Abstract

Objective: Normal cognitive development usually requires a structurally intact and complete cerebellar vermis. The aim of this study was to evaluate whether quantification by fetal magnetic resonance imaging (MRI) of vermis- and brainstem-specific imaging markers improves the definition of cystic posterior fossa malformations (cPFM).

Methods: Fetuses diagnosed with cPFM that had an available midsagittal plane on T2-weighted MRI were identified retrospectively and compared with gestational-age (GA) matched brain-normal controls. Fetuses with cPFM were assigned to three groups, according to standard criteria (vermian size and brainstem-vermis (BV) angle): normal vermian area and BV angle < 25° (Group 1); reduced vermian area and/or BV angle of 25-45° (Group 2); and reduced vermian area and BV angle > 45° (Group 3; Dandy-Walker malformation (DWM) group). The number of differentiable vermian lobules and the areas of the vermis, mesencephalon, pons and medulla oblongata were quantified, correlated with and controlled for GA, and compared between the study groups.

Results: In total, 142 cases of cPFM were included, with a mean GA of 25.20 ± 5.11 weeks. Cases comprised Blake's pouch cyst (n = 46), arachnoid cyst (n = 12), inferior vermian hypoplasia (n = 5), megacisterna magna (n = 35) and classic DWM (n = 44). In the control group, 148 fetuses were included, with a mean GA of 25.26 ± 4.12 weeks. All quantified areas and the number of differentiable vermian lobules had a significant positive correlation with GA. The number of vermian lobules and the areas of all quantified regions, except for that of the medulla oblongata, differed significantly between the study groups (P ≤ 0.015 for all). The control group had the highest number of differentiable vermian lobules and the DWM group had the lowest (P < 0.01).

Conclusions: Prenatal MRI assessment of vermian lobules is a useful addition to standard neuroradiological and neurosonographic techniques. The quantification of vermian lobules using fetal MRI allows further differentiation of cPFM into subgroups and thereby improves the classification of hindbrain malformations. © 2019 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of the International Society of Ultrasound in Obstetrics and Gynecology.

Keywords: MRI; brain segmentation; cerebellar dysplasia; cerebellar vermis; fetal MRI; hindbrain malformation; neurodevelopment; vermian lobulation.

Publication types

  • Evaluation Study

MeSH terms

  • Adult
  • Cranial Fossa, Posterior / abnormalities
  • Cranial Fossa, Posterior / diagnostic imaging
  • Dandy-Walker Syndrome / diagnostic imaging*
  • Female
  • Gestational Age
  • Humans
  • Infant, Newborn
  • Magnetic Resonance Imaging
  • Nervous System Malformations / diagnostic imaging*
  • Predictive Value of Tests
  • Pregnancy
  • Prenatal Diagnosis