For a study, researchers sought to create and internally evaluate a noninvasive approach for predicting congenital cytomegalovirus (CMV) infection following primary maternal CMV infection. They did a secondary analysis of a multicenter randomized placebo-controlled study of CMV hyperimmune globulin to prevent congenital infection. Women were eligible if they had primary CMV infection, defined as detectable plasma CMV-specific immunoglobulin (Ig)M and CMV-specific IgG with avidity less than 50% before 24 weeks of gestation or IgG seroconversion before 28 weeks, and were carrying a singleton fetus with no ultrasonographic findings suggestive of CMV infection. Antibody testing was carried out in a single reference laboratory. CMV detection in amniotic fluid, neonatal urine or saliva, or postmortem tissue was characterized as congenital infection. They built logit models for predicting congenital infection using parameters known at randomization using backward elimination. The model’s performance was evaluated using leave-one-out cross-validation (a method of internal validation).

About 344 (86%) of the 399 women participating in the experiment provided usable data for this analysis. In 68 pregnancies, congenital infection occurred (20%). The best-performing model comprised government-assisted insurance, an IgM index of 4.5 or above, and IgG avidity of less than 32%, and whether CMV was detectable in maternal plasma by polymerase chain reaction at the time of randomization. The average area under the curve after cross-validation was 0.76 (95% CI 0.70–0.82), demonstrating moderate discriminating ability. One-, two-, and three-factor models fared much worse than the four-factor model. The probability of congenital infection was 0.69 (95% CI 0.53–0.82) for a woman with government-assisted insurance, avidity less than 32%, IgM index 4.5 or higher, and detectable plasma CMV; for a woman with private insurance, avidity 32% or greater, IgM index less than 4.5, and undetectable plasma CMV, the probability of infection was 0.03 (95% CI 0.02–0.07).

In the absence of ultrasonographic evidence indicative of congenital CMV infection, They created models to predict congenital CMV infection in the context of primary maternal CMV infection. These models might help with patient counseling and decision-making.