Halo sign in fetal cytomegalovirus infection: cerebral imaging abnormalities and postmortem histopathology in 35 infected fetuses.

OBJECTIVE

To evaluate the correlation of periventricular echogenic halo (halo sign) with histopathological findings and its association with other brain imaging abnormalities in fetuses with cytomegalovirus (CMV) infection.

METHODS

This was a retrospective study of fetuses diagnosed with severe CMV infection based on central nervous system (CNS) abnormalities seen on ultrasound, which had termination of pregnancy (TOP) or fetal demise at a single center from 2006 to 2021. All included cases had been evaluated by conventional complete fetal autopsy. A maternal-fetal medicine expert reanalyzed the images from the transabdominal and transvaginal neurosonography scans, blinded to the histological findings. The halo sign was defined as the presence of homogeneous periventricular echogenicity observed in all three fetal brain orthogonal planes (axial, parasagittal and coronal). Cases were classified according to whether the halo sign was the only CNS finding (isolated halo sign) or concomitant CNS anomalies were present (non-isolated halo sign). An expert fetal radiologist reanalyzed magnetic resonance imaging (MRI) examinations when available, blinded to the ultrasound and histological results. Hematoxylin-eosin-stained histologic slides were reviewed independently by two experienced pathologists blinded to the neuroimaging results. Ventriculitis was classified into four grades (Grades 0-3) according to the presence and extent of inflammation. Brain damage was categorized into two stages (Stage I, mild; Stage II, severe) according to the histopathological severity and progression of brain lesions.

RESULTS

Thirty-five CMV-infected fetuses were included in the study, of which 25 were diagnosed in the second and 10 in the third trimester. One fetus underwent intrauterine demise and TOP was carried out in 34 cases. The halo sign was detected on ultrasound in 32 (91%) fetuses (23 in the second trimester and nine in the third), and it was an isolated sonographic finding in six of these cases, all in the second trimester. The median gestational age at ultrasound diagnosis of the halo sign was similar between fetuses in which this was an isolated and those in which it was a non-isolated CNS finding (22.6 vs 24.4 weeks; P = 0.10). In fetuses with a non-isolated halo sign, the severity of additional ultrasound findings was not associated with the trimester at diagnosis, except for microencephaly, which was more frequent in the second compared with the third trimester (10/18 (56%) vs 1/8 (13%); P = 0.04). With respect to histopathological findings, ventriculitis was observed in all fetuses with an isolated halo sign, but this was mild (Grade 1) in the majority of cases (4/6 (67%)). Extensive ventriculitis (Grade 2 or 3) was more frequent in fetuses with a non-isolated halo sign (21/26 (81%)) and those without a periventricular echogenic halo (2/3 (67%); P = 0.032). All fetuses with an isolated halo sign were classified as histopathological Stage I with no signs of brain calcifications, white-matter necrosis or cortical injury. On the other hand, 25/26 fetuses with a non-isolated halo sign and all three fetuses without a periventricular echogenic halo showed severe brain lesions and were categorized as histopathological Stage II. Among fetuses with a non-isolated halo, histological brain lesions did not progress with gestational age, although white-matter necrosis was more frequent, albeit non-significantly, in fetuses diagnosed in the second vs the third trimester (10/15 (67%) vs 3/11 (27%); P = 0.06).

CONCLUSIONS

In CMV-infected fetuses, an isolated periventricular echogenic halo was observed only in the second trimester and was associated with mild ventriculitis without signs of white-matter calcifications or necrosis. When considering pregnancy continuation, detailed neurosonographic follow-up complemented by MRI examination in the early third trimester is indicated. The prognostic significance of the halo sign as an isolated finding is still to be determined. © 2023 International Society of Ultrasound in Obstetrics and Gynecology.