BOSTON – A real-time saliva-based polymerase chain reaction assay identifies more newborns with congenital cytomegalovirus infections than a standard rapid-culture assay, investigators reported.
"A PCR [polymerase chain reaction]-based assay would be a better newborn screen as the sample is convenient and noninvasive to collect, and we have already shown that elimination of the DNA extraction step does not interfere with the sensitivity or the specificity of the assay," said Dr. Swetha Pinninti, a pediatric infectious disease fellow at the University of Alabama at Birmingham.
Dr. Swetha Pinninti
The PCR assay lends itself to high throughput methods, is less expensive than culturing, and could help clinicians identify an additional 400-1,200 infants with congenital cytomegalovirus (CMV) annually, Dr. Pinninti said at the annual meeting of the Pediatric Academic Societies.
Congenital CMV is the leading cause of nongenetic sensorineural hearing loss, but may go undetected, because approximately 90% of infected children have no clinical findings at birth. From 10%-15% of children who are infected but asymptomatic, and 60% of those who are symptomatic will develop hearing loss. In about one-third of cases, the development of sensorineural hearing loss has a delayed onset, Dr. Pinninti said.
Although detection of virus from culture of urine or saliva samples is the current gold standard, the investigators previously showed that PCR of saliva is comparable to culture-based methods (N. Engl. J. Med. 2011;364:2111-8).
In the current study, they tested whether saliva PCR could be a better screening tool for congenital CMV infection than rapid culture (RC)-based techniques.
To do so, they looked at two cohorts of infants. The first, screened from June 2008 through December 2009, included newborns who had saliva samples collected by swab and tested with both PCR and RC.
Due to a change of protocol, a second cohort of infants screened from January 2010 through November 2011 had saliva PCR testing, with RC performed on all samples that tested positive by PCR.
Infants who were positive for CMV by either screening were re-evaluated for confirmation, with testing of saliva and urine samples collected at enrollment. Infants confirmed to have congenital CMV infections were enrolled in follow-up to monitor their hearing outcomes.
Of the 35,334 infants screened in the first cohort, 161 tested positive for CMV and were enrolled for confirmation. Of this group, 152 had confirmed CMV, and 9 were confirmed to be CMV-negative.
Of the 152 confirmed CMV-positive, 147 were positive by both screening methods, 4 were PCR-positive but RC-negative, and 1 was PCR-negative but RC-positive,
In the second cohort, 37,250 infants were screened with PCR, and 123 were enrolled for confirmation. Of this group, 114 were confirmed to be CMV-positive, and 9 were confirmed to be CMV-negative. Of those confirmed positive, 105 were positive on both screening methods, and 9 were PCR-positive but RC-negative.
Of the 14 PCR/RC discordant samples, PCR detected CMV copies in varying amounts in 13, whereas RC detected cells in only 1 of the 14 (4 cells per slide).
Dr. Pinninti acknowledged a few study limitations, including the fact that infants with initially negative screens were not enrolled, and that samples were transported to a central lab, which could have degraded the quality of the cultured samples.
"This is really exciting," commented Dr. Morven S. Edwards, professor of pediatrics at Baylor College of Medicine, Houston, who comoderated the session in which the study was presented, but was not involved in the research.
The study was supported by the National Institute on Deafness and Other Communication Disorders. Dr. Pinninti and Dr. Edwards reported having no conflicts of interest.