Cytomegalovirus Saliva Shedding in Congenitally Infected Children

Source: Pinninti S, Pati S, Novak Z, et al. Cytomegalovirus saliva shedding kinetics in children with congenital CMV infection. J Pediatric Infect Dis Soc. 2025;14(5):piaf040; doi: 10.1093/jpids/piaf040.Investigators from the University of Alabama at Birmingham conducted a study to evaluate the duration of viral shedding of cytomegalovirus (CMV) in children with congenital CMV (cCMV) infection who had been identified through a universal newborn screening program. Study participants were born between March 2007 and March 2012 at 1 of 7 US hospitals and had a positive PCR assay for CMV from a saliva sample as part of a newborn screening program. Patients with cCMV infection were classified as symptomatic at birth if they had petechiae/purpura, hepatomegaly, splenomegaly, direct hyperbilirubinemia, unexplained neurologic/central nervous system abnormalities, or chorioretinitis. Infants with cCMV infection were assessed for sensorineural hearing loss (SNHL) at 3 to 8 weeks of age, then every 6 months during the first 4 years of life. At each of these assessments, saliva samples for CMV were obtained. The primary outcome was duration of saliva CMV shedding, defined as the last visit at which the PCR assay was positive. Children with a negative PCR at 1 visit and a positive PCR subsequently were classified as having intermittent shedding. The duration of saliva CMV shedding was compared in children who had been symptomatic or asymptomatic in the newborn period and those with and without SNHL, using regression analyses and accounting for multiple observations in a participant. The change in viral load (VL) over the study period also was documented.Among 100,332 newborns screened, data were analyzed on 197 children with cCMV infection who had follow-up data from at least 5 study visits. There were 15 (7.6%) patients with symptomatic cCMV. SNHL was diagnosed in 24 (12.2%) children with cCMV infection, with 54.2% diagnosed in the newborn period and 45.8% with late-onset SNHL. The rate of saliva CMV shedding decreased during the study period; 53% of study participants were shedding CMV at 18 months of age and 9.5% at 4 years of age. Similarly, the median VL was 8.89 x 106 IU/mL at enrollment and 5.6 x 103 IU/mL at 18 months of age. The duration of saliva CMV viral shedding was not significantly different in children who were symptomatic or asymptomatic at birth (median, 20 months in both groups; P = 0.18) or between those with or without SNHL (median, 21 vs 20 months; P = 0.40). Among those with SNHL, the median duration of viral shedding was 22 months for those with congenital SNHL and 18 months for those with late-onset SNHL (P = 0.62). The duration of CMV viral shedding was significantly longer in patients with intermittent viral shedding than in those who did not have intermittent shedding (median, 30 and 15 months, respectively; P <0.0001).The authors conclude that CMV was detectable in saliva of children with cCMV for a median duration of 20 months.Dr Brady’s institution was funded as a site for the CHIMES study. Dr Brady received no personal funding and was not a co-investigator on this substudy. This commentary does not contain a discussion of an unapproved/investigative use of a commercial product/device.Newborn screening for cCMV is increasing in the US.1 Those with cCMV are at risk for SNHL that may present as a neonate or later. (See AAP Grand Rounds. 2019;42[5]:572 and AAP Grand Rounds. 2022;47[3]:29.)3 One-third to one-half of cCMV-associated SNHL develops beyond the newborn period, and 50% of those at-risk experience progressive SNHL during early childhood.4,5 Children with cCMV infection are recommended to have hearing evaluations once every 3–6 months up to 1 year of age, once every 6 months from 1 to 3 years of age, and annually from 3 to 6 years of age.6 For parents, this watch-and-follow approach can be anxiety-provoking.Are there markers that can be used to predict which asymptomatic infants with cCMV will develop SNHL later? In the current study, comparisons of saliva CMV viral load between those with symptomatic and asymptomatic cCMV and between those with normal hearing and SNHL were not different between groups. The authors reported similar results with urine CMV viral loads.7 Blood CMV viral loads also have not been predictive.8A strength of the current study was the recognition that 95.2% of infants with cCMV were still shedding CMV in saliva at 6 months. Newborns who fail the hearing screen and later have SNHL diagnosed may be several months old before CMV testing is performed. The study results suggest that the absence of CMV DNA in saliva at 6–7 months of age is likely to exclude a diagnosis of cCMV in such infants.CMV DNA was detectable in saliva for a median of 20 months, irrespective of newborn symptoms or hearing outcomes.The results of this study also have important implications for the potential prevention of transmission of CMV to women of childbearing age, whether in household or child care settings.