OBJECTIVETo study the bacterial culture results of expressed breast milk.

METHODSA total of 1178 expressed breast milk samples were collected for bacterial culture. The breast milk sampled from the mothers of preterm neonates (n=615) and term neonates (n=563) who were hospitalized between May 2014 and April 2015.

RESULTSThere was no significant difference in bacterial counts between the preterm and term sample groups (P>0.05). Potential intestinal pathogens were found in 55 samples (4.63%) of the 1178 samples, with no significant difference between the preterm and term sample groups (P>0.05). The second expressed milk samples from 33 mothers were cultured. Only 10 samples (30%) were found to have the same bacteria as the first time. The detection rate of bacterial load of ≥ 10⁵ CFU/mL was higher in those samples with potential intestinal pathogens, as compared with those samples without potential intestinal pathogens (43.64% vs 14.87%; P<0.05). There was no correlation between the incidence of neonatal infections and potential intestinal pathogens in breast milk.

CONCLUSIONSBreast milk is not sterile. Bacterial loads and phylotypes are variable. Random breast milk cultures can neither describe bacterial colonies in breast milk, nor be a predictor of neonatal infection.