#40: Does Cytomegalovirus (CMV) Viral Load Correlate with Disease Severity in the Setting of Congenital CMV (cCMV) Infection? Results from a Universal cCMV Screening Study

Abstract

Abstract Background Consideration is being given to universal newborn screening for congenital CMV (cCMV). Early detection could improve neurodevelopmental and audiologic outcomes. For some viral infections, viral load is a biomarker predicting disease severity and directs the use of antiviral agents. cCMV is the most common congenital infection in the United States; its estimated prevalence is 4.5 per 1,000 live births. Since optimal screening strategies are not completely understood, this analysis was conducted to assess whether viral load at birth could stand as a biomarker of CMV symptomatology. Methods Newborn screening for cCMV was undertaken at five Twin Cities nurseries. Real-time PCR was performed on dried saliva swabs. Dried blood spot (DBS) PCR was performed independently by the UMN and CDC labs. Infections were categorized as: 1) moderately to severely symptomatic; 2) mildly symptomatic; 3) asymptomatic with isolated sensorineural hearing loss (SNHL); and 4) asymptomatic. Viral load analyses were conducted using Prism software. Results Among 16,092 newborns screened for cCMV, 72 cCMV cases were identified (prevalence of 0.44%). The sensitivity and specificity of CMV PCR were 92.9 and 99.9% for saliva, and 85.7 and 100% for DBS, respectively. 83% (60/72) of infants cCMV were classified as asymptomatic; the remaining 17% as symptomatic. In the DBS, a significant difference between viral loads in the symptomatic and asymptomatic groups was not seen (p=0.83; Mann-Whitney, Figure 1). However, in the saliva, a significant difference was seen. Mean saliva viral load in infants with confirmed cCMV and any symptomatic disease was 3.2×103 IU/ml, versus 7×106 IU/ml in asymptomatic infants (p<0.05, Mann-Whitney; Figure 1). Conclusions The prevalence of cCMV in the Twin Cities is approximately 0.5%; most infants are asymptomatic at birth. Surprisingly, an inverse relationship was observed between viral load at birth in the saliva samples and severity of CMV sequelae. This may be because maternal infection earlier during pregnancy has been correlated with more severe sequelae, but also induces a fetal immune response that limits viral replication. Hence, viral load at birth may be higher with asymptomatic infections that occur later in the pregnancy.