MRI analysis of neurodevelopmental anatomy in myelomeningocele: prenatal vs postnatal repair.

OBJECTIVE

Compared with postnatal repair, prenatal myelomeningocele (MMC) repair is associated with improved motor function and decreased need for cerebrospinal fluid (CSF) diversion. It is unknown how prenatal surgery alters neuroanatomical structures identifiable on magnetic resonance imaging (MRI). The purpose of this study was to use MRI to compare neurodevelopmental anatomy in patients undergoing fetal MMC repair compared with those undergoing postnatal repair.

METHODS

This was a retrospective review of neonates who underwent prenatal or postnatal MMC repair at our institution between 2016 and 2021. Imaging data, including prenatal ultrasound and pre- and postnatal MRI examinations, if available, were retrieved. We analyzed anatomical findings characteristically seen on MRI of the neuroaxis in patients with MMC and compared imaging findings between patients with prenatal vs postnatal MMC repair.

RESULTS

The study population included 61 patients who underwent surgical repair for MMC during the 6-year study period, of whom 25 underwent prenatal repair and 36 postnatal repair. CSF diversion was required in 24% of the prenatally repaired cohort vs 67% of the postnatally repaired cohort (P = 0.001). On postnatal MRI, a syrinx was present in 13% of the prenatally repaired cohort vs 42% in the postnatally repaired cohort (P = 0.02). Postnatal corpus callosal (CC) morphology was abnormal in 54% of the prenatally repaired cohort vs 53% of the postnatally repaired cohort (P = 0.92), while falx morphology was normal in 92% of the prenatally repaired cohort vs 34% of the postnatally repaired cohort (P < 0.001). On postnatal MRI, patients in the prenatally repaired cohort had a shorter tentorium-to-foramen-magnum distance compared with those in the postnatally repaired cohort (mean, 18.43 mm vs 22.42 mm; P = 0.01), a larger foramen magnum diameter (mean, 22.87 mm vs 18.94 mm; P < 0.001) and a smaller degree of hindbrain herniation (mean, 1.53 mm vs 8.72 mm; P < 0.001). The cerebral aqueduct was patent in 79% of the prenatally repaired cohort vs 100% of the postnatally repaired cohort on postnatal MRI (P = 0.008). Between the two cohorts, at postnatal MRI there was no significant difference in the presence of gray-matter heterotopia, presence of the septum pellucidum or size of the massa intermedia.

CONCLUSIONS

We report variations in developmental neuroanatomy in patients with MMC, including rates of CC dysgenesis, gray-matter heterotopia and additional cranial and spinal MRI findings. Compared to postnatal surgery, prenatal surgery is associated with changes to infratentorial anatomy, with minimal effect on supratentorial brain development. This information will be useful in counseling parents affected by fetal MMC and in understanding how prenatal repair of MMC affects brain development. © 2024 International Society of Ultrasound in Obstetrics and Gynecology.