Introduction: Dofetilide is a class III anti-arrhythmic agent approved for achieving and maintaining sinus rhythm in patients with symptomatic atrial fibrillation or flutter (AF). Due to a risk of QT prolongation and ventricular tachycardia (VT) patients initiated on Dofetilide need to be hospitalized for at least 3 days to closely monitor ECG. In large clinical trials 1.8-2.5% of the patients had to discontinue Dofetilide due to QT prolongation. We aimed to evaluate the frequency and correlates of discontinuation of Dofetilide due to QT prolongation (QTc >500 msec) or VT in a real-world, inpatient setting. Methods: We included 100 consecutive patients, hospitalized for Dofetilide initiation at the Minneapolis VA Medical Center from 2011 to 2014. Data was collected retrospectively from electronic medical records. Results: The mean age was 63.8 ± 9.1 years. The indication for Dofetilide was AF (88%) and VT (12%). Dofetilide had to be discontinued in 21% of the patients due to QTc prolongation (17%) or VT (2%) or lack of efficacy (2%). One patient died in the hospital. The mean QTc in patients in whom Dofetilide had to be discontinued was 512±53 msec. In 29% of the cases Dofetilide cardioverted patients to sinus rhythm. Patients who had to discontinue Dofetilide were less likely to have coronary heart disease (CHD) than those who were discharged on Dofetilide (14% vs. 44%, respectively; p= 0.012) (Table). The average number of doses received by patients before discharge was 5.5±0.9 in patients who were discharged on Dofetilde vs. 3.2±1.5 in whom Dofetilide was discontinued because of QTc prolongation (55% discontinued after 2 nd dose). Concomitant QTc prolonging drugs were used in 29% of patients who had to discontinue Dofetilide vs. 24% of those who were discharged on Dofetilide (p=0.78) (Table). Conclusion: The incidence of Dofetilide discontinuation due to high-risk ECG features in the real world setting is remarkably higher than that reported in prior clinical trials.