Objective: Many physicians and patients still have concerns about the safety of breastfeeding in mothers infected with hepatitis B virus; we evaluate the safety of the newborn and the women with HBeAg positive and high viral load, who received nucleoside analogues to block maternal to child transmisssion and selected postpartum breastfeeding after drug discontinuance. Methods: This prospective, observational study enrolled 60 HBeAg positive patients and HBV-DNA >2*10E+ 5 IU/ml, all patients started antiviral treatment for blocking maternal to child transmission at 24-28 weeks of pregnancy. All the newborns received the active-passive immunization therapy with hepatitis B immunoglobulin (HBIG) and HBVac. After the delivery, patients with normal liver function discontinued the antiviral drug and selected breastfeeding voluntarily. The safety of breastfeeding were compared with patients selected artificial feedings, they were followed up for 7 months. Primary measurements were the proportion of mothers with abormal liver function after stopping the drug and the level of newborn’s anti-HBs at 6 months of age; secondary measurements were the positive rate of neonatal HBsAg and the HBV-DNA value of the patients at 6 weeks postpartum. Results: From December 1, 2015 to May 1, 2017, 415 patients were enrolled in Beijing You’an Hospital Affiliated to the Capital Medical University and all these patients were born following full-term single-child pregnancy. After the delivery and drug withdrawal, there was no significant difference in the incidence of ALT elevation between the breast fed group and the artificially fed group: 29 /220 versus 30/195, (χ²=0.411, P=0.521). Patients continued to take the antiviral medicines between the breast feeding group and the artificial feeding group: 15 /220 versus 20/195, (χ²=1.487, P=0.223), there were no significant differences between them (P> 0.05). At the month 7, there were no significant difference between the breast fed group and the artificially fed group (747.62±374.08 mlU/ml versus 709.76±374.32 mlU/ml, t-value: 0.309, P-value>0.05). At birth, hepatitis B surface antigen (HBsAg) was detected in 15/220 and 20/195 of newborns in the breast feeding and artificially fed groups, respectively. At month 7, an intention-to-treat analysis indicated 0/220 of HBsAg infants from the breast fed versus 0/195 in the artificially fed group (P>0.05) and no significant difference was found in the rate of positive HBsAg between the two groups. In the breast fed group, the mean HBV DNA at baseline was significantly higher than that of the artificially fed group: (1.17±1.82) E+ 8 IU/ml versus (1.12±0.39)E+ 8 IU/ml, and the difference was statistically significant. Conclusions We have not found the relationship between the rate of neonatal infection and the increase of maternal abnormal liver function in HBeAg positive and high viral load patients, who chose breastfeeding after drug discontinuance.