The analytical variability of the new commercially available Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR) assay, Amplicor HIV-1 Monitor, has been assessed to establish criteria for assessing the significance of HIV-1 RNA level measurements. Estimations of the standard deviations (SD) of log-copies in inter-assay (mean 0.09 log) and in inter-laboratory (mean 0.14 log) reproducibility experiments demonstrated that the assay can discriminate with 95% confidence between 3-fold (inter-assay) and 5-fold differences (inter-laboratory). The inter-lot reproducibility (mean 0.10 log) was similar to the inter-assay reproducibility. The HIV-1 RNA concentrations measured in plasma collected in potassium EDTA anticoagulant were slightly higher than those measured in plasma collected in sodium citrate. The HIV-1 RNA concentrations measured in sera were about 50% of the HIV-1 RNA concentrations measured in paired plasma samples. However, there was a strong correlation between these two measurements (P < 0.0001). The assay was used to measure viral RNA in the plasma of 50 HIV-1 positive individuals at different stages of infection. All the individuals had detectable HIV-1 RNA (300-957000 copies/ml). There was no correlation between HIV-1 RNA and Immune Complex Dissociated (ICD) p24 antigen, but HIV-1 RNA was correlated with CD4+ cell counts (P < 0.0001) and the clinical stage (P = 0.0042), with higher HIV-1 RNA concentrations in patients with a more advanced stage of the disease. The significant association of HIV-1 RNA with major markers of HIV infection and the reliability of this sensitive, easy-to-use RT-PCR assay indicate its suitability for use in clinical trials and suggest that this assay is appropriate for routine clinical applications.