MR imaging quantitative analysis of fetal chiari II malformations and associated open neural tube defects: Balanced SSFP versus half‐fourier RARE and interobserver reliability

Abstract

To evaluate bSSFP (balanced steady state free precession) and half-Fourier RARE (rapid acceleration with relaxation enhancement) MRI sequences in their ability to demonstrate fetal anatomic landmarks, quantify the degree of cerebellar herniation in Chiari II malformations and level and length of the associated open neural tube defects, as well as evaluate interobserver reliability of these measurements. Two independent observers retrospectively reviewed MRIs of 37 fetuses with Chiari II malformations and associated open neural tube defects (mean gestational age: 27 weeks 2 days). Comparison of bSSFP and RARE included: (i) Ability to confidently identify anatomic landmarks of the posterior fossa and spine; (ii) Measurements of the foramen magnum, cerebellar tonsillar herniation length, intervertebral disc space level of tonsillar herniation, open neural tube defect length, and disc space start and end level of the open neural tube defect; (iii) Observed conspicuity of anatomic landmarks. There was no significant difference in assessment of cerebellar tonsillar herniation or open neural tube defect level between bSSFP and RARE for either observer. Intervertebral discs were more conspicuous on bSSFP while cerebellar tonsils were more conspicuous on RARE (P < 0.05). Interobserver reliability was strong for both sequences in assessing the foramen magnum (r = 0.95, 0.94), tonsillar herniation length (r = 0.93, 0.95), and open neural tube defect length (r = 0.97, 0.96). Despite improved conspicuity of the intervertebral discs with bSSFP and cerebellar tonsils with RARE, there is no significant difference in measurement of hindbrain herniation or open neural tube defect level; interobserver reliability is excellent for both sequences.