Ultrasound prediction of Zika virus-associated congenital injury using the profile of fetal growth

Zika virus (ZIKV) is a mosquito-transmitted flavivirus, recently linked to microcephaly and central nervous system anomalies following infection in pregnancy. Striking findings of disproportionate growth with a smaller than expected head relative to body length have been observed more commonly among fetuses with exposure to ZIKV in utero compared to pregnancies without ZIKV infection regardless of other signs of congenital infection including microcephaly. This study’s objective was to determine the diagnostic accuracy of femur-sparing profile of intrauterine growth restriction for the identification of ZIKV-associated congenital injuries on postnatal testing. A retrospective cohort study of pregnant women with possible or confirmed ZIKV infection between January 1, 2016 and December 31, 2017 were included. Subjects were excluded if no prenatal ultrasound was available. A femur-sparing profile of growth restriction determined using INTERGROWTH-21(st) sonographic standard for head circumference to femur length (HC: FL). Congenital injuries were determined postnatally by imaging, comprehensive eye exam and standard newborn hearing screen. A total of 111 pregnant women diagnosed with ZIKV infection underwent fetal ultrasound and 95 neonates had complete postnatal evaluation. Prenatal microcephaly was detected in 5% of fetuses (6/111). Postnatal testing detected ZIKV-associated congenital injuries in 25% of neonates (24/95). A HC: FL Z-score ≤ -1.3 had a 52% specificity (95% CI 41–63%), 82% negative predictive value (NPV, 95% CI 73–88%) for the detection of ZIKV-associated congenital injuries in the neonatal period. A more stringent threshold with a Z-score ≤ -2 was associated with a 90% specificity (95% CI 81–95%), 81% NPV (95% CI 77–85%). Excluding cases of fetal microcephaly, HC: FL (Z-score ≤ -2) demonstrated a similar specificity (89%, 95% CI 81–95%) with superior NPV (87%, 95% CI 84–90%). The sonographic recognition of a normally proportioned fetus may be useful prenatally to exclude a wider spectrum of ZIKV-associated congenital injuries detected postnatally.