373: Cytomegalovirus infection in pregnancy: role of serial ultrasounds


      To evaluate the effectiveness of ultrasound in the antenatal prediction of symptomatic congenital cytomegalovirus (CMV) infection.

      Study Design

      A prospective study of women with primary CMV infection managed between 2000 and 2008 was performed. Primary CMV infection was defined: seroconversion to IgG positivity or anti-CMV IgG of low avidity with IgM. Serial targeted ultrasound examination was performed every fortnight until 24 weeks and every 4 weeks until delivery. Ultrasonographic markers of fetal infection were: brain calcifications, hydrocephaly, microcephaly, germinative cysts, ventricular dilatation, polymicrogyria hyperechogenic bowel, hepatomegaly, liver calcifications, IUGR, pericardial effusion, cardiomyopathy, ascite, enlarged placenta. Infection status was disclosed by viral isolation in urine, blood and saliva samples at birth or CMV tissue inclusions at placental histological exam or fetal autopsy. Neonatal follow-up included up to one year after delivery. Cross-tabs were used to calculate sensitivities, specificities, and positive and negative predictive values of ultrasound vs symptomatic congenital infection.


      A total of 88 women were included in the study, whose 26 were seroconversions. 11 cases were lost at follow up. There were 20 (24%) neonates with congenital CMV infection, in which 6 (33%) were symptomatic at 1-year-life follow-up. Table 1 shows comparison between symptomatic infected (ill) and asymptomatic infected and non infected (healthy) neonates.
      Table 1
      US findingSymptomatic infectionAsymptomatic infectionSensistivity (%)Specificity (%)PPV (%)NPV (%)


      Serial ultrasound identify 67% of symptomatic fetuses at risk of severe sequelae with a NPV of 98%.