Institutional members access full text with Ovid®

Share this article on:

Non-Primary Maternal CMV Infection Following Viral Shedding in Infants

Boucoiran Isabelle MD MSc; Mayer, Bryan T. PhD; Krantz, Elizabeth M. MS; Marchant, Arnaud MD, PhD; Pati, Sunil PhD, MBA; Boppana, Suresh MD; Wald, Anna MD, MPH; Corey, Larry MD; Casper, Corey MD, MPH; Schiffer, Joshua T. MD, MSc; Gantt, Soren MD, PhD, MPH
The Pediatric Infectious Disease Journal: Post Acceptance: January 08, 2018
doi: 10.1097/INF.0000000000001877
Original Studies: PDF Only


Most infants with congenital cytomegalovirus (CMV) infection are born to seropositive women as a result of maternal CMV non-primary infection (reinfection or reactivation). Although infected children are known to transmit CMV to their seronegative mothers, the frequency and magnitude of non-primary maternal CMV infection after exposure to viral shedding by children in their household has not been characterized.


A cohort of Ugandan newborns and their mothers were tested weekly for CMV by quantitative PCR of oral swabs. Infant primary infection and maternal non-primary infection were defined by the onset of persistent high-level oral CMV shedding. Strain-specific antibody testing was used to assess maternal reinfection. Cox regression models with time-dependent covariates were used to evaluate risk factors for non-primary maternal infection.


Non-primary CMV infection occurred in 15 of 30 mothers, all following primary infection of their infants by a median of 6 weeks (range 1 to 10) in contrast to none of the mothers of uninfected infants. The median duration of maternal oral shedding lasted 18 weeks (range: 4 - 42) reaching a median maximum viral load of 4.69 log copies/ml (range 3.22 - 5.64). Previous-week infant CMV oral quantities strongly predicted maternal non-primary infection (hazard ratio 2.32 per log10 DNA copies/swab increase; 95% CI 1.63, 3.31). Maternal non-primary infections were not associated with changes in strain-specific antibody responses.


Non-primary CMV infection was common in mothers following primary infection in their infants, consistent with infant-to-mother transmission. Since infants frequently acquire CMV from their mothers, e.g. through breast milk, this suggests the possibility of “ping-pong” infections. Additional research is needed to characterize the antigenic and genotypic strains transmitted among children and their mothers.

Sources of support/funding: This work was supported by grants from the NIH – Roadmap KL2 Clinical Scholar Training Program KL2-RR025015-01 (SG), University of Washington Center for AIDS Research New Investigator Award P30-AI027757 (SG), P01-AI 030731 (LC, AW, SS, JTS, MLH); R01-CA138165 (CC), and P30-CA015704 (CC and LC) – and by a Child & Family Research Institute Salary Award (SG).

Correspondence: S. Gantt, BC Children’s Hospital Research Institute, 950 W 28th Ave, Rm A5-144, Vancouver, BC V5Z 4H4, Canada (

Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.