Revisiting short- and long-term outcome after fetal first-trimester primary cytomegalovirus infection in relation to prenatal imaging findings

Objective To determine the short- and long-term outcome of pregnancies with proven first-trimester fetal cytomegalovirus (CMV) infection in a large prospective cohort. Methods This was a prospective cohort study of pregnancies with documented primary maternal CMV infection in the first trimester and evidence of fetal infection, referred for further evaluation between January 2011 and January 2018. Maternal serological diagnosis of primary CMV infection was documented by seroconversion. Vertical CMV transmission was identified by amniocentesis with polymerase chain reaction (PCR) for the CMV genome. After birth, fetal infection was re-tested by PCR in neonatal urine or saliva samples. All patients underwent serial prenatal ultrasound scans and fetal magnetic resonance imaging (MRI) at 32-33 weeks' gestation. All neonates underwent ocular fundus examination, an ultrasound brain scan and hearing evaluation, and were followed periodically for a median of 2 years (range, 6 months to 10 years). Follow-up information was obtained from hospital charts and by telephone interviews with parents. The CMV-associated outcomes assessed were sensorineural hearing loss (SNHL), neurodevelopmental abnormality, composite clinical outcome (including SNHL and neurodevelopmental abnormality) and composite outcome (additionally including termination of pregnancy (TOP)). The association between prenatal ultrasound or MRI findings and abnormal outcome was assessed. Results Primary CMV infection in the first trimester occurred in 123 patients. The rate of an abnormal ultrasound finding was 30.9%, and the rate of an abnormal MRI finding was 30.1% overall and 14.1% in the subgroup of patients with normal ultrasound. Of the 85 patients with normal ultrasound, 12 had an abnormal MRI finding, of whom five (5.9%) had true anatomical findings. Fifteen patients decided to terminate the pregnancy owing to abnormal prenatal findings on either ultrasound or MRI. Overall, the rate of CMV-associated postnatal and childhood sequelae was 27.8%, with a rate of 16.7% for SNHL and 11.1% for neurodevelopmental abnormalities, mostly slight motor or verbal delay. Approximately half of the cases with CMV-associated sequelae did not have any abnormal prenatal imaging findings. Abnormal prenatal findings on ultrasound were not associated significantly with SNHL, neurodevelopmental delay or composite clinical outcome (P = 0.084, 0.109 and 0.176, respectively), but they were associated with the composite outcome including TOP (P < 0.001). We identified a non-significant trend for a higher rate of SNHL in the group with abnormal ultrasound than in those with normal ultrasound. For abnormal MRI findings, we found a correlation only with neurodevelopmental abnormality and composite outcome (P= 0.014 and P< 0.001, respectively). Conclusions The risk of childhood sequelae after first-trimester fetal CMVinfection ismost often associated with abnormal prenatal imaging findings. However, normal imaging does not rule out the development of SNHL and minor neurodevelopmental abnormalities. Copyright (C) 2019 ISUOG. Published by John Wiley & Sons Ltd.