MR imaging of the fetal cerebellar vermis: Biometric predictors of adverse neurologic outcome.

Abstract

To provide normal biometry of the cerebellar vermis using fetal MR and determine threshold values associated with abnormal neurologic outcome. Cerebellar vermis biometry was applied in prospective, cross-sectional evaluation of fetal brains. Vermis length and inferior vermian distance were obtained in mid-sagittal planes using T2-weighted, single-shot sequences with 1.5 Tesla MR. Measurements were compared with reference nomograms from a retrospective review of fetal brains with normal intracranial anatomy. Observed and predicted measurements of the cerebellar vermis were recorded. Neurologic outcome was classified as normal or abnormal. Unpaired t-tests and discriminate analysis were applied to the two measurements and differences between the observed and predicted values. The reference group included 64 fetuses of 13 to 38 weeks gestation. Both vermis length and inferior vermian distance increased linearly with time (r = 0.92, P < 0.001; r = 0.32, P = 0.01). The prospective group included 64 additional fetuses with documented normal (39/64, 61%) and abnormal (25/64, 39%) outcomes. Significant differences were seen in vermis length, inferior vermian distance, and correlation with predicted values based on neurologic outcome (P < 0.001). Vermis length discrepancy ≥ 4 mm or inferior vermian distance ≥ 4 mm were associated with abnormal neurologic outcome. MR measurements of a short, raised vermis characterized by a vermis length discrepancy ≥ 4 mm or an inferior vermian distance ≥ 4 mm is associated with abnormal neurologic, syndromic, and developmental outcomes. J. Magn. Reson. Imaging 2016;44:1284-1292.