BACKGROUND Transient atrial and appendage dysfunction occurs after cardioversion of atrial fibrillation. It has been suggested that one component of early dysfunction is related to the method of restoration of sinus rhythm and it is less severe in patients undergoing pharmacological than electrical cardioversion. The aim of this study was to compare left atrial chamber and left atrial appendage mechanical function before and after 48 hours from electrical or pharmacological cardioversion in patients with chronic atrial fibrillation. METHODS We studied the effects of the mode of cardioversion on Doppler left atrial and appendage function in 19 patients with persistent atrial fibrillation (> or = 4 weeks), who were randomized to pharmacological (quinidine) or electrical cardioversion (protocol: 200, 300, 360 J) after pre-treatment with verapamil. Transthoracic and transesophageal echocardiography were performed before and 48 hours after the restoration of sinus rhythm. To determine left atrial and appendage mechanical dysfunction, the peak A wave velocities were obtained from transmitral flow velocity profiles recorded in the apical 4-chamber view, and peak emptying and filling appendage velocities were measured by the transesophageal approach with the sample volume placed at the orifice of the left atrial appendage. All the patients were pre-treated with verapamil before cardioversion in order to achieve a satisfactory control of heart rate. RESULTS Mean peak A wave velocities were 0.52 +/- 0.12 m/s in the patients treated electrically and 0.54 +/- 0.08 m/s in those treated pharmacologically (p = NS). Before and after electrical cardioversion, the peak filling velocities of the left atrial appendage were 0.42 +/- 0.17 and 0.43 +/- 0.17 m/s respectively, and the peak emptying velocities 0.30 +/- 0.14 and 0.36 +/- 0.17 m/s respectively; before and after pharmacological treatment, the peak filling velocities were 0.38 +/- 0.1 and 0.43 +/- 0.1 m/s respectively, and the peak emptying velocities were 0.30 +/- 0.13 and 0.43 +/- 0.24 m/s respectively (p = 0.08). CONCLUSIONS Even a long period of atrial fibrillation does not lead to a marked depression of global left atrial and left atrial appendage function 48 hours after the restoration of sinus rhythm by means of electrical or pharmacological cardioversion. There is no evidence that electrical cardioversion causes greater post-cardioversion atrial and/or appendage dysfunction than pharmacological treatment after 48 hours. Pre-treatment with verapamil may have reduced the dysfunction (probably because of a reduction in mechanical remodeling during atrial fibrillation).