OBJECTIVE When patients with implantable cardioverter defibrillators (ICD) develop symptomatic atrial fibrillation (AF), external direct current cardioversion (EDCCV), as well as internal cardioversion using their ICD, are the options available. It is currently unknown which of these two methods are more effective. We compared the effectiveness of EDCCV versus internal cardioversion to terminate AF in patients with a single-coil ICD. METHODS This randomized controlled trial (clinicaltrial.gov NCT03164395) enrolled consecutive patients with a single-coil ICD that presented with symptomatic AF of less than 1-year duration. They received either the maximum energy internal shock through the ICD or an EDCCV using transcutaneous pads of 200 J. The primary endpoint was a successful conversion to sinus rhythm after one shock. Crossover was permitted if the first shock was unsuccessful. RESULTS Thirty-one patients were enrolled in the study, including 16 in the internal ICD cardioversion group. The study included patients with a mean age of 59.5 ± 16.0 years, 41.9% females, median AF duration 1 month (interquartile range 1-3), 45.2% non-ischemic cardiomyopathies, mean EF 28.6 ± 16.0%, and 45.2% biventricular ICD. There were no significant differences in baseline clinical characteristics between the two groups. In the internal cardioversion group, 5/16 patients (31.3%) met the primary endpoint versus 14/15 (93.3%) in the EDCCV group, p < 0.001. All patients that failed the first shock were subsequently cardioverted externally. CONCLUSIONS Among patients with a single-coil ICD and symptomatic AF of less than 1 year, external direct current cardioversion is much more effective than internal shock through the ICD.