RESUMO
OBJECTIVE: Prenatal imaging has several critical roles in the diagnosis and management of myelomeningocele, including specific family counseling and the selection of fetal surgery or postnatal repair. In this study, the authors compared the accuracy of fetal MRI and prenatal ultrasonography (US) in predicting the spinal lesion level and assessed the correlation between imaging findings and motor function as independently evaluated by a physical therapist (PT) after birth. METHODS: A retrospective review of demographic and clinical data was performed to identify children who had been treated with postnatal myelomeningocele closure at a single institution between March 2013 and December 2018. Patients were eligible for inclusion if they had all of the following: prenatal US identifying the neural tube defect level, fetal MRI identifying the neural tube defect level, and postoperative PT evaluation identifying the motor deficit level. Statistical analysis was performed using Cohen's kappa coefficient to compare the US- and MRI-demonstrated lesion level and correlate these findings with the motor level assigned postnatally by a PT via manual muscle testing. RESULTS: Thirty-four patients met the inclusion criteria. The mean gestational age at US was 23.0 ± 4.7 weeks, whereas the mean gestational age at MRI was 24.0 ± 4.1 weeks. The mean time from surgery to the PT evaluation was 2.9 ± 1.9 days. Prenatal US and MRI were in agreement within one spinal level in 74% of cases (25/34, k = 0.43). When comparing the US-demonstrated spinal level with the PT-assigned motor level, the two were in agreement within one level in 65% of cases (22/34, k = 0.40). When comparing MRI-demonstrated spinal level with the PT motor level, the two were in agreement within one level in 59% of cases (20/34, k = 0.37). MRI and US were within two spinal levels of the PT evaluation in 79.4% and 85.3% of cases, respectively. MRI and US agreed within two levels in 97.1% of cases. Prenatal US and MRI were equivalent when comparing the difference between the imaged level and the postnatal motor deficit level (mean level difference: 1.12 ± 1.16 vs 1.17 ± 1.11, p = 0.86). CONCLUSIONS: Prenatal US and MRI equivalently predicted the postnatal motor deficit level in children with myelomeningocele. These data may be valuable in prenatal prognostication.