Background An zidovudine (AZT)-substitution regimen containing 24-week stavudine (d4T) followed by long-term AZT for HIV therapy is potential to trade off short-term AZT-related anemia and long-term risks associated with d4T in resourcelimited settings.However,evidence is scarce.This study aims to assess the efficacy and safety of AZT-substitution regimen,aiming to find a regimen with better efficacy,less adverse events,and more affordability in resource-limited settings.Methods This prospective,multicenter study enrolled 499 (190 on d4T regimen,172 on AZT regimen,and 137 on AZT-substitution regimen) HIV-1-infected subjects who initiated combined antiretroviral therapy and attended follow-up visits over 96 weeks from 2009 to 2011.Lamivudine (3TC) and either nevirapine (NVP) or efavirenz (EFV) were the other two drugs in the antiretroviral regimens.Virologic and immunologic responses and adverse events were monitored at baseline and at weeks 4,12,24,36,48,60,72,84,and 96.Results In terms of hematological adverse effects,AZT-substitution group had similar safety profiles to d4T group and was superior to AZT group.In comparison with AZT-substitution group,AZT group was associated with higher risk of developing anemia (adjusted hazard ratio (aHR)for anemia ≥ grade Ⅱ,8.44,95％ CI 1.81-39.46) and neutropenia (aHR for neutropenia ≥ grade Ⅱ,1.86,95％ CI 1.19-2.93).The prevalence of lipodystrophy in d4T group was 19.5％,while that in AZT-substitution group was zero.As to antiretroviral efficacy,these three groups showed no differences.Conclusion AZT-substitution regimen provides a relatively safe and effective first-line antiretroviral strategy in resource-limited settings.