Human milk is an important source of microorganisms for infant gut colonisation. Although the maternal antibiotic prophylaxis is an important strategy to prevent maternal/neonatal sepsis, it has to be investigated how it may affect the human milk microbiota, especially the genus Bifidobacterium, which has been associated to health benefits. Here, we investigated the impact of the maternal antibiotic prophylaxis on the human milk Bifidobacterium spp. and total bacteria counts, in the first week (short-term) and first month (medium-term) after delivery. Human milk samples were collected from 55 healthy lactating women recruited from the University Hospital of the University of São Paulo at days 7±3 and 30±4 after vaginal delivery. Twenty one volunteers had received maternal antibiotic prophylaxis (MAP group) and 34 had not received MAP (no-MAP group) during or after labour. Total DNA was isolated from milk samples, and the bacterial counts were estimated by quantitative PCR (qPCR). We found lower levels of Bifidobacterium in the MAP group in the first week after delivery (median = 2.1 vs 2.4 log of equivalent cells/ml of human milk, for MAP and no-MAP groups, respectively; P=0.01), although there were no statistical differences in total bacteria count. However, no differences were found in Bifidobacterium counts between the groups at day 30±4 (median = 2.5 vs 2.2 log of equivalent cells/ml of human milk, for MAP and no-MAP groups, respectively; P=0.50). Our results suggest that MAP has a significant impact on Bifidobacterium counts in human milk, reducing this population in the first week after delivery. However, throughout the first month after delivery, the Bifidobacterium counts tend to recover, reaching similar counts to those found in no-MAP group at day 30±4 after delivery.