A prospective study of the prevalence of bronchiectasis in rheumatoid arthritis was conducted over an 18-month period in 46 patients (34 women and 12 men with a mean age of 60.1 years) meeting 1987 American College of Rheumatology criteria for rheumatoid arthritis. All patients underwent high-resolution computed tomography of the chest, whose results were confronted with a number of clinical, laboratory test and lung function testing parameters. Bronchiectasis or bronchiolectasis was found in 23 patients (50%) and was the most common abnormality detected by high-resolution computed tomography. When the four patients with isolated bronchiolectasis were excluded, the prevalence was 41%. Eighteen of the 23 patients had not been diagnosed with bronchiectasis before the study and 13 were free of respiratory symptoms. No significant differences were found between the 23 patients with and the 23 patients without bronchiectasis for age at onset or duration of the rheumatoid arthritis, extraarticular involvement, positive rheumatoid factors, bony erosions, use of corticosteroids or immunosuppressives, respiratory manifestations, smoking, or spirometry parameters. Patients without bronchiectasis were more likely to have impaired diffusion of carbon monoxide across the alveolar-capillary membrane. Among the patients with bronchiectasis, those with respiratory symptoms (n = 10) were more likely to have a history of lung disease and those without respiratory symptoms (n = 13) were more likely to have a diagnosis of bronchiectasis secondary to rheumatoid arthritis; no other differences were found between these two subgroups. Routine use of high-resolution computed tomography, a technique capable of demonstrating silent bronchiectasis, showed that bronchiectasis was the most common lung change in rheumatoid arthritis in our study. At the time of the study, there was no evidence that presence of bronchiectasis was associated with more severe joint or lung symptoms.