High Rate of Cytomegalovirus Transmission in Breast Milk

Abstract & Commentary

Hal B. Jenson, MD, FAAP Chair, Department of Pediatrics, Director, Center for Pediatric Research, Eastern Virginia Medical School and Children’s Hospital of the King’s Daughters, Norfolk, VA, is Associate Editor for Infectious Disease Alert

Synopsis: Cytomegalovirus (CMV) is frequently transmitted via breast milk and poses risk for infant infection and symptomatic disease, especially for preterm infants. CMV is preferentially shed in breast milk without detectable CMV in peripheral blood.

Source: Meier J, et al. Human Cytomegalovirus Reactivation During Lactation and Mother-to-Child Transmission in Preterm Infants. J Clin Microbiol. 2005;43:1318-1324.

Human cytomegalovirus (cmv) shedding in breast milk was prospectively studied in 73 mothers and their 89 preterm infants in Berlin, Germany. Gestational age was 24-33 weeks (median, 28 weeks) and birthweight was 380-2010 g (median, 1,119 g). Feeding was initiated as early as possible, usually by 24-48 hours of life. Only blood products from CMV-seronegative donors were used for transfusions. Breast milk, pharyngeal (endotracheal, if intubated) aspirates, and urine were collected at day 3 of life, day 7, and then weekly through 64 days, and tested by PCR and viral culture.

By PCR, 48 of 73 (66%) mothers shed CMV in breast milk, which corresponds to the CMV seroprevalence of about 60% in the German population. The maternal CMV serostatus was known for 28 women; 19 of 20 (95%) seropositive women shed CMV in breast milk, compared to 1 of 8 (13%) seronegative women, suggesting acute CMV in one mother. Peripheral blood mononuclear cells were tested from 13 seropositive mothers with CMV in breast milk, and 6 seronegative who were also negative for CMV in breast milk. Only 2 of 13 (15) seropositive mothers had detectable virus in blood.

Overall, 20 of 73 (27%) mothers shed CMV in breast milk, and 21 of 89 (24%) infants acquired CMV. The transmission rate from CMV-positive mothers to their offspring was 38% (21 of 55 mother-infant pairs, including one set of twins). Postnatal CMV infection was confirmed in 12 of the 22 infected infants by negative urine PCR for CMV within the first 2 weeks, followed by positive urine PCR. The remaining 9 infants had positive urine PCR for CMV within the first 2 weeks of life, suggesting possible congenital infection. Only 2 of the 22 infected infants developed symptomatic disease, one with a sepsis-like illness and another with hepatitis and icterus.

Comment by Hal B. Jenson, MD, FAAP

Human CMV virolactia, or virus in breast milk, appears to be much more frequent than previously appreciated. In this study, nearly all lactating women who are seropositive (19 of 20, 95%) shed CMV in breast milk, as detected by PCR. Interestingly, this did not correlate with maternal CMV viremia, as only 2 of 13 seropositive women tested had CMV detected in peripheral blood. These results suggest that CMV replication may occur preferentially in breast tissues, which may represent a distinct virological compartment from bone marrow granulocytes and monocytes that are known to harbor latent CMV. These findings are consistent with other studies that CMV DNA in breast milk results from local virus replication rather than systemic virus replication and the transmission of virus via the egress of maternal leukocytes in breast milk.

Symptomatic CMV disease was documented in 2 of 22 infected infants (9%), or 2 of 12 (17%) infected infants with confirmed postnatal infection. Although this study used PCR, which may detect DNA but not infectious virus, the documented infections in 22 infants without another source supports the conclusion that viral DNA in breast milk represents intact CMV virions.

If extrapolated to all preterm infants, there is a large vulnerable population with a significant risk of symptomatic disease. If confirmed by larger studies, these results suggest that maternal breast milk for preterm infants should possibly be routinely frozen, pasteurized, or otherwise treated before use in preterm infants during the first few weeks of life. Although treatment should diminish the risk of postnatal, or acquired, CMV infection transmitted via breast milk, treatment would likely also adversely affect the immunoglobulins, leukocytes, and other immunological constituents in breast milk.