Abstract 224

Background: As we have recently demonstrated, postnatal CMV-transmission to preterm infants by breastfeeding is a potential health risk (Pediatr Infect Dis J 1998; 17: 53-8).

Aim: To test the effectiveness of heating and freezing on CMV-inactivation in breast-milk.

Methods: 20 ml breast milk samples spiked with different concentrations of laboratory strain AD 169 was subjected to either 72°C for 10 sec. or to -20°C for 3 to 21 days. A 2×8 well plaque assay was used for virus detection in milk whey following immediate early antigen staining.

Results: A reduction in infectivity of 70-80% was observed by adding CMV to native human milk (n=8). Absolutely no infection could be detected after heating, whereas freezing only partially reduced CMV infectivity, especially when virus load was high.

Conclusion: Only short-time pasteurisation of breast milk does effectively inactivate CMV and therefore can prevent transmission to preterm infants. However, the impact of this method on immuno-protective components in breast milk has not been evaluated yet.