OBJECTIVES: To evaluate the in vitro activity of anti-leishmanial drugs against intracellular Leishmania donovani amastigotes in different types of macrophages. METHODS: Mouse peritoneal macrophages (PEMs), mouse bone marrow-derived macrophages (BMMPhi), human peripheral blood monocyte-derived macrophages (PBM Phi) and differentiated THP-1 cells were infected with L. donovani. Cultures were incubated with sodium stibogluconate, amphotericin B deoxycholate (Fungizone), miltefosine or paromomycin sulphate over six concentrations in 3-fold serial dilutions for 5 days. Analysis was based on percentage inhibition of infected macrophages and EC(50)/EC(90) values estimated using sigmoidal curve-fitting. RESULTS: The rank order of drug activity was the same in the different macrophage populations: amphotericin B > miltefosine > sodium stibogluconate > paromomycin. However, significant (P < 0.05) differences were observed between populations. Amphotericin B was more active in PEMs and BMM Phi (EC(50) 0.02-0.06 microM) compared with PBM Phi and differentiated THP-1 cells (EC(50) 0.08-0.40 microM) and miltefosine was more active in PBM Phi (EC(50) 0.16-0.74 microM) compared with PEMs and BMM Phi (EC(50) 2.60-7.67 microM). Sodium stibogluconate displayed highest activity in PBM Phi (EC(50) 1.38-1.89 microg Sb(v)/mL), followed by PEMs (EC(50) 21.75-27.79 microg Sb(v)/mL) and BMM Phi and differentiated THP-1 cells (EC(50) 28.96-112.77 microg Sb(v)/mL). Paromomycin showed highest activity in PBM Phi (EC(50) 80.03-104.38 microM) and PEMs (EC(50) 75.42-201.63 microM). CONCLUSIONS: In vitro activity of anti-leishmanial drugs is host cell dependent. This has implications for: (i) the evaluation of in vitro drug activity; (ii) the evaluation of drug susceptibility of clinical isolates; and (iii) the standardization of anti-leishmanial drug assays.