Fetal neurosonography as accurate tool for diagnosis of brain involvement in tuberous sclerosis

Objective: To demonstrate the potential utility of dedicated neurosonography for the diagnosis of fetal brain involvement in tuberous sclerosis complex. Methods: This was a multicenter retrospective study of fetuses at high risk for tuberous sclerosis complex. Dedicated neurosonographic, fetal magnetic resonance imaging (MRI) and postnatal reports were reviewed. Data collected included reason for referral, gestational age at which cardiac rhabdomyoma was first suspected and final number of cardiac rhabdomyomas detected on dedicated imaging. We searched for tuberous sclerosis complex-related brain involvement, defined as the presence of one or more of the following findings: white-matter lesions; subependymal nodules; cortical/subcortical tubers; and subependymal giant-cell astrocytoma. Results: We included 20 patients at high risk of tuberous sclerosis complex, of whom 19 were referred for the presence of cardiac rhabdomyomas and one for a deletion in chromosome 16 involving the tuberous sclerosis complex gene locus. Cardiac rhabdomyomas were diagnosed at a mean gestational age of 27 + 2 weeks (range, 16 + 0 to 36 + 3 weeks) and the mean number of cardiac rhabdomyomas per patient was 4 (range, 1–10). Brain involvement was present in 15 fetuses, in 13 of which the disease was confirmed in one or more of the following ways: chromosomal microarray analysis (n = 1), exome sequencing (n = 7), autopsy (n = 4), clinical tuberous sclerosis complex in the newborn (n = 4) and a sibling diagnosed with clinical tuberous sclerosis complex (n = 1). In two cases, the disease could not be confirmed: one was lost to follow-up and autopsy, following termination of pregnancy, was not performed in the other. Among the five cases without brain findings, tuberous sclerosis complex was confirmed in three by exome sequencing (n = 2) and/or autopsy findings (n = 2). The two remaining cases had normal exome sequencing; one case had five cardiac rhabdomyomas, which was a highly suggestive finding, while in the final case, the autopsy was considered normal, representing the only false-positive case in our cohort. Conclusions: Contrary to current literature, dedicated neurosonography appears to be effective in the diagnosis of brain involvement in fetuses at risk of tuberous sclerosis complex and should be used as the first-line approach. Although the number of cases in which MRI was performed was small, it seems that, in the presence of ultrasound findings, the added value of MRI is low.