June 13, 2011

Saliva Testing Catches CMV Infection in Newborns

Photo of a mother’s hands touching her infant Anthea Sieveking, all rights reserved by Wellcome Images

A saliva sample from a newborn can be used to quickly and effectively detect cytomegalovirus (CMV) infection, a major cause of hearing loss in children. Better CMV screening could help doctors know which patients to monitor for symptoms so they can be treated as quickly as possible.

CMV is the most common infection passed from a mother to her unborn child. Between 20,000 and 30,000 infants are infected with the virus upon birth, and 10% to 15% of them are at risk for developing hearing loss. Monitoring infected children for signs of hearing loss as they grow is the best way to ensure they get early treatment. But infected babies often show no symptoms, so screening for CMV at birth is critical.

CMV infection is currently detected using the “rapid culture” test. However, this method isn’t easily automated to allow for widespread screening. A team at the University of Alabama at Birmingham, led by Dr. Suresh Boppana and Dr. Karen Fowler, set out to develop a fast, accurate test to identify CMV-infected newborns. They used a technique called polymerase chain reaction (PCR), which amplifies virus DNA and could be used to screen large numbers of infants for CMV. The study was funded by NIH’s National Institute on Deafness and Other Communication Disorders (NIDCD).

The researchers had previously shown that testing heel-stick blood by PCR accurately predicted CMV infection only 30% to 40% of the time, but 95% accuracy is required for a screening test. In an attempt to improve this percentage, the team tested saliva instead. They collected saliva via mouth swab from nearly 35,000 newborn infants at 7 hospitals and sent them to the laboratory for testing. In the first phase of the study, the samples were stored in liquid, and in the second phase, they were air-dried. The findings were published in the June 2, 2011, issue of the New England Journal of Medicine.

Among liquid saliva samples, the PCR and rapid culture tests both identified CMV in the same 85 infants, suggesting that the PCR test is 100% accurate compared to the current standard. Among dried saliva samples, rapid culture identified 76 infected infants. PCR identified 74 of these same infants, dropping its accuracy rate slightly to 97.4%.

Sixteen additional liquid saliva samples tested positive with the PCR test but not the rapid culture one. Follow-up analysis revealed that 3 of these were indeed positive, leading the researchers to conclude that PCR may be superior to rapid culture for CMV screening. For the remaining 13 samples, CMV may have been present in the mother’s breast milk or secretions in the birth canal that showed up in the baby’s mouth swab. For this reason, the researchers suggest that all positive results be confirmed with follow-up testing within the first 3 weeks of life.

“We now know that we have a test with saliva that works,” Boppana says. “The challenge is, unlike the dried blood spot, which is already used for newborn screening in hospitals across the country, we don’t have a system in place for the collection of saliva. But we’ve shown that if you wanted to test a lot of babies for congenital CMV infection, it can be done.”

The CMV-infected infants from this study are now enrolled in a follow-up program to see how strongly CMV infection at birth contributes to hearing loss in early childhood.

Related Links