Antinuclear antibodies (ANA) can be induced by some drugs used in the treatment of cardiovascular disease. The reported frequency with which these antibodies are detected in patients varies widely. This variation results from a number of factors. The sensitivity of the ANA assay is influenced by the selection of substrates, the concentration of antisera and characteristics of the detection systems such as ultraviolet microscopes or electrophoretic apparatus. The incidence of ANA also varies with age and sex of the patient, being more common in older people and in females. Identification of a drug suspected of producing ANA demands a careful evaluation of the data with precisely standardised laboratory procedures and comparison of data with appropriate control groups of untreated and treated patients. Cardiovascular drugs associated with increased ANA incidence can be considered in two categories: A) A few drugs induce ANA in most patients if therapy is continued for long enough at high enough dosage. Many of these patients develop systemic lupus erythematosus like-syndromes. This group includes procainamide, hydrallazine at high doses and practolol. B) A further group of drugs produces ANA in 20 to 30% of patients, few if any, of whom develop SLE. Methyldopa and acebutolol are clearly in this category, while there is some evidence that labetolol, guanethidine and hydrallazine at low doses may also be implicated. Some very preliminary evidence suggests those patients on the beta-adrenoceptor blocking drugs atenolol, metoprolol and exprenolol exhibit a mildly increased incidence of ANA, but there is no evidence to suggest associated SLE. Only patients who develop ANA while on treatment with category A drugs require careful monitoring for SLE.