Retrospective examination of lung tissue specimens for the presence of Legionella organisms: comparison of an indirect fluorescent-antibody system with direct fluorescent-antibody testing.

Although direct fluorescent-antibody (DFA) testing has been used successfully for a number of years to detect legionellae in clinical specimens, the number of known species and serogroups of Legionella has now increased to such an extent that the performance of DFA testing for all serological variants is impractical. Lung homogenates that were submitted to the Centers for Disease Control, Atlanta, Ga., from patients with suspected legionellosis, from November 1977 through May 1982, were originally screened by DFA testing. In our study 498 of these lung homogenates were screened by indirect fluorescent-antibody (IFA) testing, using a panvalent antiserum pool containing antibodies to 25 serological variants of Legionella spp. Fluorescein isothiocyanate-labeled goat anti-rabbit immunoglobulin was used as the second antibody of the sandwich system. For positive homogenates, i.e., those containing Legionella organisms, species and serogroup identification was made by IFA staining with polyvalent serum pools and then with monovalent antiserum. Of the 498 homogenates screened, 39 (7.8%) were positive by IFA testing. Four (0.8% of total; 10.3% of positive homogenates) of these had previously been negative by DFA testing, but subsequent testing showed that they contained Legionella organisms for which DFA reagents were not available at the initial screening. All specimens that were positive by DFA testing were also positive by IFA testing. IFA testing with polyvalent antisera is a simple, efficient method which is at least as sensitive as DFA testing and which can be used by clinical laboratories to cope with the increasing number of known serological variants of Legionella spp.