We evaluated prospectively the detection of respiratory syncytial virus (RSV) by culture and by direct antigen detection using an indirect immunofluorescence assay (IFA), a direct monoclonal immunofluorescence assay (DFA), and a monoclonal enzyme immunoassay (EIA). Of 221 specimens, 95 (43%) were culture positive for RSV, 4 (1.8%) contained more than one virus, and 17 (7.6%) contained a virus other than RSV. Overall, HEp-2 and Flow 6000 cells grew significantly more RSV isolates (82 and 72%, respectively) than A549 cells, which grew only 29% of the isolates. The mean time for RSV detection with HEp-2 cells was 2.9 days. This was significantly less than the mean time for RSV detection with either Flow 6000 cells (6.1 days) or A549 cells (6.4 days). Of 221 specimens, 129 were tested simultaneously by culture, IFA, and DFA. Of these 129 specimens, 62 (48%) were positive by culture, 69 (53%) were positive by IFA, and 70 (54%) were positive by DFA. For 92 specimens screened simultaneously by culture, IFA, and EIA, positive results were obtained for 33 (36%) of the specimens by both culture and IFA and for 29 (32%) of the specimens by EIA. Of 126 culture-negative specimens, 21 (17%) were positive for RSV when determined by IFA. Conversely, 14 (15%) of 95 RSV culture-positive specimens were negative by IFA, whereas DFA missed 19% of the culture-positive specimens. Compared with culture, the Kallestad EIA kit had a sensitivity and specificity of 73 and 92% respectively, but missed 9 (27%) of 33 culture-positive specimens. These data demonstrate that isolation by culture continues to be important for viral diagnosis of RSV infections and that for valid comparative studies between viral isolations and rapid detection methods, both sensitive host cells and appropriate test conditions are required.