BACKGROUND: Short-term antiretroviral therapy (START) to prevent mother-to-child transmission (MTCT) is currently recommended for all HIV-1-infected pregnant women. The objective of this study was to assess the effect on CD4 cell counts and viral load dynamics the withdrawal of START after birth could generate. METHODS: This was a 5-year cohort study involving HIV-1-infected pregnant women who presented with CD4 counts >300 cells/microL and had received START to prevent MTCT. RESULTS: Seventy-five pregnancies were assessed. In 24 cases, there was a history of antiretroviral therapy prior to prophylaxis. The median baseline CD4 count was 573 cells/microL. In 75% of cases, prophylaxis was started after 26.6 weeks of gestation. The median CD4 cell count increase over baseline during prophylaxis was 24.5%. In only five cases did HIV-1 viral load remain detectable during prophylaxis. After START, CD4 cell counts did not drop significantly, and the HIV-1 viral load plateau was near the baseline level. The estimated mean time for CD4 count to fall below 300 cells/microL was 3.5 years and was directly associated with high baseline CD4 cell count, as well as with CD4 increase after prophylaxis, whereas it was negatively correlated with previous use of antiretroviral (ARV) drugs and persistence of detectable HIV-1 viral load during prophylaxis. CONCLUSIONS: A potent, well-tolerated prophylactic ARV regimen can improve CD4 cell counts during and after START. In women receiving such prophylaxis, there is a remarkable time interval for CD4 cell counts to drop to levels that indicate treatment.