Laboratory diagnosis of Legionnaires' disease

Legionnaires' disease was diagnosed in 32 patients with nosocomially acquired pneumonia by using direct immunofluorescence examination of respiratory tract secretions, indirect immunofluorescence determinations of antibody titer, and/or culture of Legionella pneumophila from respiratory tract secretions. Cultures of specimens for Legionella pneumophila were positive for 13 of 21 patients for whom appropriate specimens were submitted. Using a positive culture as a definition of Legionnaires' disease, the sensitivity of direct immunofluorescence examination was 62%; that of indirect immunofluorescence serologic studies, 75%. Using any positive test as a definition of Legionnaires' disease, the sensitivity of direct immunofluorescence examination of respiratory tract secretions was 50%, that of indirect immunofluorescence serologic studies (using all 4 serogroups), 80%; that of culture, 65%. A negative control group of 21 patients without Legionnaires' disease was used to estimate the specificity of the tests. The specificity of direct immunofluorescence examination was 94%; that of indirect immunofluorescence serologic studies, 75%. Culture of specimens was necessary to diagnose some cases of Legionnaires' disease because the other 2 tests were negative. All 3 tests should be performed for optimal sensitivity and specificity.