In order to determine whether commercial assays presently in use for the quantification of plasma human immunodeficiency virus type 1 (HIV-1) RNA levels detect different genetic viral subtypes with the same reliability, a panel of 38 samples corresponding to 22 HIV-1-infected patients, representing non-B subtypes A-F, was examined. One to three plasma samples belonging to each individual were tested by the second-generation HIV-1 branched DNA (bDNA) assay (Chiron, Spain), the Nuclisens assay (Organon-Teknika, Spain), the Amplicor Monitor reverse transcriptase polymerase chain reaction assay (Roche, Spain), and the Ultradirect Monitor (Roche) using primers specifically designed to amplify non-B HIV-1 subtypes. Each of the different methods for measuring viral load showed a distinct sensitivity for non-B HIV-1 subtypes. Values higher than the assay detection limit were obtained in 22 (57.9%), 33 (86.8%), and 37 (93.3%) samples using the bDNA, Nuclisens, and Monitor assays, respectively. Significantly different values (>0.5 logs) were found in 55.3%, 81.6%, and 71.1% of specimens comparing results provided by bDNA and Nuclisens, bDNA and Monitor, and Nuclisens and Monitor, respectively. Quantitative values provided by the Ultradirect Monitor test using non-B primers were particularly discordant with the other tests. Overall, 44.7% of samples yielded higher viral load values with this assay than with the regular Monitor assay, reflecting its enhanced sensitivity for non-B subtypes; however, the reproducibility of this test was low. These results support the recommendation of always using the same assay when monitoring plasma viraemia.