Rapid culture testing of saliva or urine is the standard method for identification of infants with congenital cytomegalovirus (CMV) infection, but it is labor- and resource-intensive and not suitable for mass screening. Dried blood spots (DBS) are routinely collected from newborns, and there has been increasing interest in the United States for the use of DBS polymerase chain reaction (PCR)-based methods for CMV screening in newborns. This study examined the diagnostic accuracy of 2 DBS real-time PCR assays for mass screening of newborns for CMV infection, and compared the results of PCR screening with rapid tissue testing. A total of 20,448 infants born at 7 US medical centers had saliva specimens tested by rapid culture and were divided into 2 groups. The tissue culture screening results were compared with single-primer DBS (group I, n = 11,442) in 2007 and with double-primer DBS (n = 11,442, group 2) in 2008. Infants with positive saliva or DBS screening specimens had the congenital infection confirmed in a follow-up study using rapid culture testing on saliva or urine. Of the 20,448 infants in both groups, 92 (0.45%; 95% confidence interval [CI], 0.36%–0.55%) had confirmed congenital CMV infection. Screening with saliva rapid culture correctly identified 91 of 92 infants with confirmed congenital CMV infection. The single-primer DBS PCR identified 28% (17/60) of group 1 infants confirmed to have CMV infection, and the 2-primer DBS PCR identified 34% (11/32) group 2 infants with confirmed infection. With respect to the single-primer DBS PCR assay, the sensitivity and specificity in identifying infants with confirmed CMV infection were 28.3% (95% CI, 17.4%–41.4%) and 99.9% (95% CI, 99.9%–100%), respectively; the positive and negative predictive values were 80.9% (95% CI, 58.1%–94.5%) and 99.6% (95% CI, 99.5%–99.7%), respectively; and the positive and negative likelihood ratios were 804 (95% CI, 279–2318) and 0.7 (95% CI, 0.6–0.8), respectively. With respect to the double-primer DBS PCR assay, the sensitivity and specificity for identifying infants with confirmed infection were 34.4% (95% CI, 18.6%–53.2%) and 99.9% (95% CI, 99.9%–100.0%), respectively; the positive and negative predictive values were 91.7% (95% CI, 61.5%–99.8%) and 99.8% (95% CI, 99.6%–99.9%), respectively; and the positive and negative likelihood ratios were 3089 (95% CI, 411–23,227) and 0.7 (95% CI, 0.5–0.8), respectively. These findings demonstrate that DBS real-time PCR has limited value as a mass screening test of newborns for CMV infection. It has low sensitivity and misses about two-thirds of the infections.
Read full abstract