BACKGROUND AND OBJECTIVES: The role of atrial myocardial dysfunction after cardioversion is unclear. In a comparison of patients after successful cardioversion from chronic atrial fibrillation (CAF) and normal controls, we sought to determine whether Doppler-derived atrial strain rate (A-sr) could be used to measure global left atrial function and whether A-sr was reduced in patients with CAF. METHODS: A-sr was measured from the basal septal, lateral, inferior and anterior atrial walls from the apical four-chamber and two-chamber views in 37 patients with CAF who had been cardioverted to sinus rhythm and followed up for 6 months, and in a cohort of 37 healthy people. Conventional measures of atrial function included peak transmitral A-wave velocity, A-wave velocity time integral, atrial fraction and the left atrial ejection fraction. Doppler tissue imaging was used to estimate atrial contraction velocity (A' velocity). In addition to amplitude parameters, the time to peak A-sr was measured from aortic valve closure. RESULTS: Immediately after cardioversion, A-sr in the CAF cohort (baseline) was significantly lower than in controls (mean (SD) -0.53 (0.31) v -1.6 (0.75) s(-1); p<0.001); the A-sr correlated with A' velocity (r = 0.63; p<0.001) in patients. Atrial function improved over time, with maximal change observed in the initial 4 weeks after cardioversion. The time to peak A-sr was increased in the CAF group compared with controls (0.55 (0.15) v 0.46 (0.12) s), but this failed to normalise over time. CONCLUSION: A-sr is a descriptor of atrial function, which is reduced after cardioversion from CAF and subsequently recovers.