Chronic Chlamydia pneumoniae infection, characterized by elevated levels of C. pneumoniae IgG and IgA antibodies and immunocomplexes, is associated with myocardial infarction and angiographically verified coronary heart disease. C. pneumoniae organisms have also been found in coronary atheromas, but not in healthy vessels. We investigated the effect of 4 months' doxycycline therapy on serological markers of C. pneumoniae infection and coronary risk factors. Thirty-four non-smoking men, aged 57.9 (+/- 5.2) years, who had mild hypertension or moderate hypercholesterolaemia and a previous coronary bypass, were randomly assigned to receive doxycycline or matching placebo for 4 months. Acetylsalicylic acid and beta-blocker were the only other medications allowed. Patients were examined physically and by laboratory tests; their basal nitric oxide production was determined by blood flow responses to intra-arterial monomethyl-L-arginine at baseline and at 2 and 4 months. The tests were also taken at 6 months after medication. At entry, the demographic, clinical, blood flow and laboratory measurements were similar in both groups, with the exception of fibrinogen and triglyceride levels, which were higher in the placebo group. No significant changes were found in any of the parameters during the treatment. Thus extended doxycycline therapy did not affect C. pneumoniae antibodies or coronary heart disease risk factors. We conclude that doxycycline monotherapy may not be sufficient to eradicate chronic C. pneumoniae infection.