To evaluate the therapeutic efficacy and safety of lamivudine treatment in late pregnancy by analyzing the maternal-fetal outcomes of chronic hepatitis B (CHB) mothers featuring hepatitis B e antigen (HBeAg)-positivity and highly viremic status. A total of 256 pregnant women in the second or third trimester with monoinfected CHB, HBeAg-positivity, and HBV DNA more than 6 log10 copies/mL were divided into two groups: lamivudine (lam) treatment (n=164) or no treatment (controls; n=92). All infants were treated with hepatitis B immune globin (HBIg; 200 IU) within 12 hrs of birth and 15 days later, and were given the recombinant HBV vaccine (20 mug) at 0, 1 and 6 months. All infants were followed-up to at least seven months and hepatitis B surface antigen (HBsAg) and HBV DNA levels were used to determine perinatal transmission (PT) rates. The mothers' data from routine blood analysis, tests of hepatic and renal function, detection of HBV markers and HBV DNA were retrospectively analyzed to determine changes associated with the lam treatment. Correlations of lam treatment with HBV PT rate, alanine aminotransferase (ALT) normalization, adverse reactions, pregnancy complications, congenital deformities, and infants' growth/development were determined by statistical analyses. Prior to delivery, the lam-treated mothers had significantly lower HBV DNA levels (3.72+/-1.78 vs. controls: 7.83+/-0.67 log10 c/ml; t=-22.359, P less than 0.001). The rate of virological response in the lam-treated group was 97.56% (160/164). The lam-treated group had significantly higher ALT normalization rate (90.20% vs. controls: 55.88%; X2=13.349, P less than 0.001) and significantly lower HBeAg titer (957.73+/-458.42 vs. controls: 1296.35+/-383.14 S/CO; t=-5.410, P less than 0.001). At birth, the infants from lam-treated mothers had significantly lower HBsAg-positivity (15.24% (25/164) vs. controls: 30.43% (28/92); X2=8.284, P=0.004). By 7-12 months after birth, none of the infants born to lam-treated mothers tested positive for HBsAg, compared to 8.70% (8/92) of the infants born to mothers in the control group (X2=14.721, P less than 0.001). None of the lam-treated mothers required treatment discontinuation due to adverse events or lam-resistance. No congenital deformities were observed during the study and follow-up periods. There were no differences between the lam-treated and control groups for postpartum hemorrhage, gestational age, infants' height/weight or Apgar scores. In highly viremic HBsAg+ mothers with CHB, lam treatment in the second or third trimester of pregnancy is safe and effective for reducing HBV maternal-neonatal transmission.