Cardiac involvement is the most important component of acute rheumatic fever. The role of echocardiography in the evaluation of rheumatic carditis has not been adequately defined. We used echocardiography in a large sample of patients with acute rheumatic fever to describe morphological abnormalities associated with rheumatic carditis and to assess its role in the diagnosis of rheumatic carditis. Cross-sectional and color Doppler echocardiographic examination was performed in 108 consecutive patients with acute rheumatic fever within 24 to 48 hours of diagnosis. Twenty-eight patients had acute rheumatic fever without clinical evidence of carditis (group 1). Thirty-five patients had a presumed first episode of rheumatic carditis (group 2), and 45 patients had a recurrence of carditis (group 3). Patients in group 1 did not demonstrate any evidence of valvular regurgitation. Mitral regurgitation was the most common Doppler echocardiographic feature in groups 2 (94%) and 3 (84%). Valvular thickening with or without restriction of leaflet mobility was frequently seen in rheumatic carditis. One of every 4 patients with rheumatic carditis demonstrated echocardiographic presence of focal valvular nodules. These nodules were found on the body and the tips of the mitral valve leaflets and disappeared on follow-up. Ventricular dilatation (group 2, 54%; group 3, 74%) and restriction of leaflet mobility (group 3, 37%) were common mechanisms of mitral regurgitation in rheumatic carditis; valve prolapse (group 2, 9%; group 3, 16%) and annular dilatation (group 2, 12%; group 3, 21%) were infrequent. The majority of patients with rheumatic carditis had normal left ventricular systolic function. Congestive heart failure (group 2, 17%; group 3, 40%) was invariably associated with the presence of hemodynamically significant valve lesions. On follow-up, no patient in group 1 developed valvular regurgitation. In group 2 patients, a progressive decrease in left ventricular dimensions was observed without any change in ventricular fractional shortening. Valvular regurgitation remained unchanged in 69% of patients, decreased in 22%, and disappeared in 9%. In patients with rheumatic carditis, the mitral valve is most often involved and mitral regurgitation is the most common finding on color flow imaging. Mitral regurgitation in rheumatic carditis is related to ventricular dilatation and/or restriction of leaflet mobility. Rheumatic carditis does not result in congestive heart failure in the absence of hemodynamically significant valve lesions. In a quarter of patients with rheumatic carditis, we observed valve nodules that may represent echocardiographic equivalents of rheumatic verrucae. Our study failed to reveal any incremental diagnostic utility of echocardiography and Doppler color flow imaging in rheumatic fever without clinical evidence of carditis.