Evaluation of saliva pools method for detection of congenital human cytomegalovirus infection.

Human cytomegalovirus (HCMV) is the most frequent cause of congenital infection, leading to long-term sequelae especially sensorineural hearing loss (SNHL). Since 5-15 % of the asymptomatic newborns will develop late sequelae, the implementation of a universal screening would allow the identification of infected children and early intervention. The aim of this study was to validate the use of saliva pools of 10 and 20 samples for the detection of HCMV congenital infection. Four spiking samples (negative saliva matrix added with known concentration of AD169 strain culture supernatant) and a set of 12 saliva samples, collected from newborns with confirmed congenital infection in their first three weeks of life, were tested individually and after dilution in 10 and 20 pools by an "in-house" RT-PCR. Both pool methodologies, 10-pool and 20-pool samples, had 100 % sensitivity and specificity when compared with individual samples. This methodology could allow a cost reduction close to 85 % and 89 %, respectively for the 10-pool and 20-pool approach, when compared with testing each sample individually. This significant reduction may open the possibility to perform the newborn screening for HCMV in a large-scale.