BACKGROUND: Dofetilide gained Food and Drug Administration approval for persistent atrial fibrillation/flutter (AFF) based on 2 randomized, placebo-controlled, dose-ranging studies. Concerns of proarrhythmia have prompted the manufacturer to develop specific treatment guidelines. OBJECTIVE: To determine the effectiveness and safety of dofetilide in clinical practice as well as to ascertain whether clinicians are following established dosing guidelines. METHODS: This retrospective analysis evaluated guideline adherence and safety in patients who received dofetilide at a tertiary care medical center. Safety assessment included monitoring for the occurrence of excessive QTc interval prolongation and torsade de pointes. Excessive QTc interval prolongation was defined as >15% above baseline after the first dose or >500 msec following any dose (>550 msec in patients with ventricular conduction abnormalities). Patients were included in the effectiveness assessment if they received at least 36 hours of dofetilide for persistent AFF, received an appropriate dose per guidelines, and did not receive direct current cardioversion during the evaluation period. We compared the 36-hour conversion rate with dofetilide in this study with that observed in the EMERALD and SAFIRE-D trials using the Z test, and we evaluated the incidence of excessive QTc interval prolongation in high-risk subgroups by chi(2) analysis. RESULTS: Investigators identified 107 patients. The primary indication for dofetilide was AFF, with 58.9% receiving the drug for paroxysmal disease. Prescribing followed established guidelines, except that it was used intermittently by nonconfirmed prescribers (5.6%) and/or at inconsistent doses (114%). Excessive prolongation of the QTc interval occurred in 17.8% of patients after the first dose and 26.2% during subsequent doses; prolongation was more common in those with structural heart disease (p < 0.01). No patients developed torsade de pointes. In the effectiveness assessment (n = 25), the conversion of persistent AFF at 36 hours was higher than in previous studies (48% vs 27.2%; p = 0.05). CONCLUSIONS: In clinical practice, the conversion of persistent AFF with dofetilide is at least comparable to premarketing studies, with a similar safety profile. Institutions should continue to emphasize adherence with established treatment guidelines.