Surgical interventions for fetal hydrocephalus: systematic review

To evaluate the experience with and developmental outcomes of surgical interventions for fetal hydrocephalus.

A literature search was conducted in MEDLINE, CINAHL, EMBASE and the Cochrane Central Register of Controlled Trials (CENTRAL) databases from inception to 10 April 2024, using a combination of Medical Subject Headings and free text search in titles and abstracts. Case reports, case series, cohort studies and randomized controlled trials reporting on surgical interventions for fetal hydrocephalus in the English language were identified. Studies in which cephalocentesis was performed solely to facilitate vaginal delivery, studies focusing solely on hydrocephalus in the case of spina bifida and studies in which developmental outcome was not reported were excluded. The quality of research was evaluated using the Joanna Briggs Institute checklist.

We identified 12 studies: six case reports and six case series. In total, data were presented on 172 participants: 29 underwent cephalocentesis, 131 had a ventriculoamniotic shunt inserted, with or without valves, and 14 underwent attempted endoscopic third ventriculostomy, of which 11 procedures were successful. In the three participants in which endoscopic third ventriculostomy was unsuccessful, two had a ventriculoamniotic shunt inserted and are included in the case numbers for that procedure. Hydrocephalus was thought to be isolated antenatally in 160 (93.0%) cases, but 18 (11.3%) of these had a further abnormality detected postnatally. There were no perinatal complications of cephalocentesis or endoscopic third ventriculostomy procedures. Of the 131 fetuses that had ventriculoamniotic shunts fitted, perinatal complications were seen in nine (6.9%): four fetuses died in utero and three were born preterm within 48 h of the shunt insertion procedure and died as a consequence. Nine postnatal deaths were reported 48 h after the procedure and were not directly attributable to it: five deaths were related to preterm birth and four were from a congenital abnormality. Ventriculoamniotic shunt failure occurred in 42/131 (32.1%) cases. The developmental outcome was reported in 137 participants between 1 and 38 months of age, of which 63 (46.0%) infants had a normal developmental outcome, 12 (8.8%) had mild developmental abnormalities and 62 (45.3%) had moderate/severe developmental difficulties. Of the 113 participants with truly isolated hydrocephalus and a reported developmental outcome, 63 (55.8%) reported normal development, eight (7.1%) had mild developmental difficulties and 42 (37.2%) had moderate-to-severe developmental difficulties. Of the 24 participants with other abnormality diagnosed antenatally or postnatally and a reported developmental outcome, none had normal development, four (16.7%) had mild developmental difficulties and 20 (83.3%) had moderate-to-severe developmental difficulties.

The effectiveness of surgical intervention for fetal hydrocephalus is affected by historical problems with identification of hydrocephalus cases and high rates of morbidity and mortality following preterm birth. Developmental outcomes were better for isolated hydrocephalus, such as from aqueductal stenosis, but the quality of developmental outcome data was poor and only assessed in the short term, and there was no control group for comparison. Improvements in in-utero magnetic resonance imaging, whole genome/exome studies and new fetal surgery techniques may allow for re-evaluation of surgical intervention in the context of randomized controlled trials to result in more meaningful conclusions. © 2025 International Society of Ultrasound in Obstetrics and Gynecology.