BACKGROUNDCongenital cytomegalovirus infection following maternal primary cytomegalovirus infection affects approximately 0.4% of newborns in the United States but may be hard to diagnose prenatally.OBJECTIVETo evaluate the current sensitivity and specificity of amniocentesis in detecting congenital cytomegalovirus infection.STUDY DESIGNSecondary analysis of a multicenter randomized placebo-controlled trial designed to evaluate whether cytomegalovirus hyperimmune globulin reduces congenital cytomegalovirus infection in neonates of individuals diagnosed with primary cytomegalovirus infection before 24 weeks of gestation. At randomization, subjects had no clinical evidence of fetal infection. Eligible subjects were randomized to monthly infusions of cytomegalovirus hyperimmune globulin or placebo until delivery. Although not required by the trial protocol, amniocentesis following randomization was permitted. The fetuses and neonates were tested for the presence of cytomegalovirus at delivery. Comparisons were made between those with and without amniocentesis and between those with cytomegalovirus-positive and negative results, using chi-square or Fisher exact test for categorical variables and the Wilcoxon rank sum test or t test for continuous variables. A P value of <.05 was considered significant.RESULTSFrom 2012 to 2018, 397 subjects were included, of whom 55 (14%) underwent amniocentesis. Cytomegalovirus results were available for 53 fetuses and neonates. Fourteen amniocenteses were positive (25%). Gestational age at amniocentesis was similar between those with and without cytomegalovirus present, as was the interval between maternal diagnosis and amniocentesis. The prevalence of fetal or neonatal infection was 26% (14/53). The neonates of all 12 subjects with a positive amniocentesis and available results had cytomegalovirus infection confirmed at delivery, as did 2 neonates from the group of 41 subjects with a negative amniocentesis, with a sensitivity of 86% (95% confidence interval, 57-98), specificity of 100% (95% confidence interval, 91-100), positive predictive value of 100% (95% confidence interval, 74-100), and negative predictive value of 95% (95% confidence interval, 83-99). Amniocentesis-positive pregnancies were delivered at an earlier gestational age (37.4 vs 39.6 weeks; P<.001) and had lower birthweights (2583±749 vs 3428±608 g, P=.004) than amniocentesis-negative pregnancies.CONCLUSIONAmniocentesis results are an accurate predictor of congenital cytomegalovirus infection.