The paper by Cook et al 1 examines the possible association between Chlamydia pneumoniaeinfection and asthma. The authors conclude that their data do not support this association. However, we feel that the serological tests performed give important information on the prevalence of infection, but are not sufficiently complete to make definitive conclusions on the incidence of acute C pneumoniaeinfection in the populations under study. The major pitfall in the study, as pointed out by the authors, is the small proportion of patients from whom a convalescent serum sample was drawn. Moreover, the arbitrary exclusion of IgM positive patients for the diagnosis of acuteC pneumoniae infection may have been misleading since the possibility of cross reactivity with rheumatoid factor could have been effectively ruled out by using IgG absorption prior to IgM microimmunofluorescence determination.2 Notwithstanding these facts, the authors conclude that the study does not support “an association between C pneumoniae antibody titres and the incidence of acute asthma attacks”.

Analysis of table 1 indicates that the acute asthma and control populations appear to be significantly different in terms of age and sex distribution, the control population being significantly older and showing a male predominance. Both these factors are associated with increased C pneumoniae incidence and prevalence. The authors report using a logistic regression modelling method in which the age value is implemented as “± 10 years”, which is roughly equivalent to the difference in mean age between the acute asthma and control populations.

This study is certainly noteworthy in that it underlines an association between C pneumoniae infection and severe chronic asthma, particularly “brittle” asthma, which will require further investigation in the future.