AIM: To compare throat swabs with sputum specimens for Chlamydia pneumoniae antigen detection. METHODS: During a one year period, sputum and throat swabs from 50 patients over 15 years of age with acute or persisting lower respiratory tract infection were examined for C pneumoniae antigen by direct immunofluorescence. RESULTS: C pneumoniae antigen was detected in 18/50 patients (36.0%) from sputum, throat swab, or both. Paired sputum and throat swabs were received from 35/50 patients (70.0%). C pneumoniae antigen was detected in either or both specimens from 14/35 patients (40.0%). Of the 14 positive patients, both specimens were positive in nine (64.3%), throat swab only in four (28.6%), and sputum only in one (7.1%). Of the remaining 15 patients from whom only a single specimen was sent, a further three of eight throat swabs and one of seven sputum specimens were positive. There was no statistically significant difference between the results obtained from the two types of specimen. CONCLUSIONS: Throat swabs may be as good as sputum for the detection of C pneumoniae antigen.