Conversion of persistent atrial fibrillation (AF) to sinus rhythm is frequently seen during the 3-day in-hospital loading period required during dofetilide initiation, but it is not known whether pharmacologic conversion (PC) without the need for electrical cardioversion (EC) is a predictor of long-term maintenance of sinus rhythm during continued therapy with dofetilide. We sought to test the hypothesis that PC predicts durable maintenance of sinus rhythm and determine additional predictors of long-term maintenance of sinus rhythm on dofetilide. We retrospectively reviewed all elective inpatient admissions for dofetilide loading from 2003 to 2011 at the University of Virginia. A multivariate Cox proportional hazards model was used to assess predictors of maintenance of sinus rhythm after in-hospital dofetilide loading. In all, 101 patients with a current duration of AF lasting for a median of 1.86 months (interquartile range 0.47 to 6.03) were included in the analysis. Forty-seven patients were in the PC group, whereas 54 patients were in the EC group. Patients in the PC group remained longer in sinus rhythm compared with the patients in the EC group (log-rank p = 0.032). The seventy-fifth percentile for the current episode duration in the PC group was 5.77 months, indicating that even long-standing persistent AF frequently converted pharmacologically. Hypertension and a longer duration of the current AF episode were also predictors of recurrence in the multivariate model. In conclusion, PC during in-hospital dofetilide loading is an important predictor of durable response even in long-standing persistent patients, which has important public health implications for choice of therapy.
Dofetilide is a class III antiarrhythmic drug first approved by the Food and Drug Administration in 2000 for the cardioversion of patients in atrial fibrillation (AF) or atrial flutter and subsequent maintenance of sinus rhythm. Spontaneous conversion of persistent AF to sinus rhythm is frequently seen during the 3-day in-hospital loading period required during dofetilide initiation, but it is controversial whether pharmacologic conversion (PC) is a predictor of long-term maintenance of sinus rhythm during continued therapy with dofetilide. In addition, there are only limited data on patient characteristics that might predict conversion and/or long-term efficacy. In this retrospective review of our patient population with persistent AF treated with dofetilide, we sought to determine whether patients with PC to sinus rhythm after initiation of dofetilide without the need for electrical cardioversion (EC) have a more durable response to therapy compared with patients who require EC after dofetilide initiation to achieve normal sinus rhythm.
Methods
We retrospectively reviewed all elective inpatient admissions for dofetilide loading from 2003 to 2011 at the University of Virginia. Patients who were not in AF at the time of admission or began taking dofetilide immediately after undergoing pulmonary vein isolation procedure were excluded from the analysis. Before administration of the first dofetilide dose, renal function was evaluated and the estimated glomerular filtration rate calculated. The initial dose was selected according to package insert guidelines. Patients were admitted to the hospital for monitoring on day 1 and given the first dose in the evening. If AF persisted after 4 doses of dofetilide, patients were electrically cardioverted. All patients were monitored on an inpatient basis for 6 doses, as per guidelines. Renal function was assessed each day, and corrected QT intervals were assessed 2 hours after each dose. Guidelines for dosage adjustment during drug loading were followed. For those patients who pharmacologically converted, we calculated the mean number of doses to cardioversion. Guideline-based definitions for AF were used.
After discharge from the initial hospitalization, patients underwent regular clinical follow-up with monitoring done per physician preference. Because patients had persistent AF before commencement of therapy, continuation of dofetilide without the need for additional cardioversion was considered treatment success, even if brief self-terminating recurrences might have occurred.
Two patients were treated with dofetilide on 2 separate occasions because of drug interruptions. If they were treated twice, the most recent treatment was included in the analysis. Echocardiographic data were obtained from a 2-dimensional echocardiography performed before dofetilide therapy or immediately after. We also recorded whether patients were in AF or normal rhythm at the time of the echocardiogram. Medical history and previous therapies were recorded from the medical record. Providers defined patients as having hypertension, obesity, and hyperlipidemia.
A univariate Cox regression analysis was performed to determine predictors of dofetilide success. A multivariate Cox analysis was then performed to determine predictors of acute failure. The continuous variables assessed were heart rate, corrected QT interval during AF, time between initial diagnosis of AF and initiation of dofetilide, and duration of the current persistent episode of AF. Categorical variables were recognized risk factors for AF such as hypertension, heart failure, obesity, type 2 diabetes mellitus, degree of left atrial enlargement (none, mild, moderate, or severe enlargement ), historical use of amiodarone or class IC drugs, previous ablation, gender, and final dofetilide dose. Finally, a multivariate logistic analysis was performed to identify predictors of the need for EC.
Statistical analysis was performed using SAS, version 9.3 (The SAS Institute, Cary, North Carolina). The Institutional Review Board for Human Subjects Research at the University of Virginia Health System approved this study.
Results
There were 101 patients who met entry criteria for the study (patient characteristics listed in Table 1 ). One patient received 2 doses of dofetilide, but because of QT prolongation, the medication was discontinued. This patient was not included in the analysis. No patient had torsades de pointes during loading. The mean age of the total population was 61 years. The group was 40% men with type 2 diabetes mellitus in 9%, hypertension in 63%, hyperlipidemia in 30%, and obesity in 17%; 23% of patients had used amiodarone previously, and 3% had previous use of flecainide or propafenone. Catheter ablation for AF had been attempted in 10% of patients.
| Variable | All, n = 101 (%) | PC Group, n = 47 (%) | EC Group, n = 54 (%) | p Value |
|---|---|---|---|---|
| Age, yrs, mean ± SD | 61.4 ± 11.5 | 61.3 ± 9.9 | 61.6 ± 12.9 | 0.89 |
| Women | 41 (40.6) | 18 (38.3) | 23 (42.6) | 0.66 |
| Time since AF Dx (yrs), median (IQR) | 3.82 (1.33–7.38) | 4.36 (1.33–7.29) | 3.47 (1.27–7.83) | 0.96 |
| Duration current AF (mo), median (IQR) | 1.86 (0.47–6.03) | 1.87 (0.45–5.77) | 1.37 (0.47–7.35) | 0.82 |
| Dofetilide dose (μg BID) | ||||
| 125 | 3 (3.0) | 1 (2.1) | 2 (3.7) | 0.71 |
| 250 | 45 (44.5) | 21 (44.7) | 24 (44.4) | |
| 500 | 53 (52.5) | 25 (53.1) | 28 (51.9) | |
| QTc, BL (ms), mean ± SD | 442 ± 33.4 | 439.4 ± 40.0 | 443.8 ± 26.7 | 0.53 |
| HR, BL (beats/min), mean ± SD | 92.9 ± 22.4 | 95.0 ± 24.2 | 91.1 ± 21.0 | 0.39 |
| Previous class IC use | 22 (21.8) | 14 (29.8) | 8 (14.8) | 0.07 |
| Previous amiodarone | 23 (22.8) | 11 (23.4) | 12 (22.2) | 0.89 |
| Previous AF ablation | 10 (9.9) | 7 (14.9) | 3 (5.6) | 0.12 |
| LA size | ||||
| Normal | 13 (12.9) | 6 (12.8) | 7 (13.0) | 0.35 |
| Mildly enlarged | 20 (19.8) | 16 (34.0) | 14 (25.9) | |
| Moderately enlarged | 20 (19.8) | 11 (23.4) | 9 (16.7) | |
| Severely enlarged | 20 (19.8) | 6 (12.8) | 14 (26.0) | |
| Not available | 18 (17.8) | 8 (17.0) | 10 (18.5) | |
| Diabetes mellitus | 9 (8.9) | 6 (12.8) | 3 (5.6) | 0.24 |
| Hypertension | 64 (63.4) | 28 (59.6) | 36 (66.7) | 0.46 |
| Hyperlipidemia | 30 (29.7) | 15 (31.9) | 15 (27.8) | 0.65 |
| Heart failure | 20 (19.8) | 8 (17.0) | 8 (17.0) | 0.51 |
| Obesity | 17 (16.8) | 8 (17.0) | 8 (17.0) | 0.96 |
In the total cohort, the mean ejection fraction was 48% by echocardiography. Mean left atrial size was 4.7 cm. AF had been initially diagnosed a median of 3.82 years (interquartile range 1.33 to 7.38) before dofetilide initiation, and the current episode had begun a median of 1.86 months (interquartile range 0.47 to 6.03) before drug initiation: 47 (46%) of the total cohort of 101 patients converted pharmacologically after initiation of dofetilide (PC group), whereas 54 (54%) underwent planned EC after initiation of dofetilide (EC group) to restore sinus rhythm. Those who had successful PC did so after 2.5 doses. In each group, 8 patients had dose reductions because of QT prolongation, and the average dose did not differ between the groups. There were no significant differences between patients in the EC group and those in the PC group, although there was a trend for greater previous use of class IC drugs and a greater number of patients with previous AF ablation in the PC group. Previous amiodarone use did not differ significantly between these groups.
Patients in the PC group were more likely to remain longer in sinus rhythm compared with the patients in the EC group ( Table 2 and Figure 1 ) based on Kaplan-Meier analysis. The interquartile range upper bound (seventy-fifth percentile) of duration of the current episode of AF was >5 months in the PC group, indicating that a number of patients with longer durations of persistent AF converted with dofetilide; 2.7 years after loading with dofetilide, 40% of patients remained in sinus rhythm. Patients were followed until dofetilide was stopped, and the longest follow-up period was 6.3 years. The mean times to recurrence in the PC and EC groups were 521 and 397 days, respectively.
| Covariate | HR (95% CI) | p Value |
|---|---|---|
| Electrical cardioversion | 1.72 (1.04–2.84) | 0.0354 |
| Time since first AF diagnosis (yrs) | 0.98 (0.92–1.04) | 0.501 |
| Hypertension | 1.75 (1.04–2.95) | 0.036 |
| Duration of current AF episode (yrs) | 1.06 (1.01–1.12) | 0.018 |
| Amiodarone | 1.16 (0.66–2.05) | 0.602 |
| Class IC | 1.26 (0.72–2.20) | 0.416 |
| Previous AF ablation | 1.12 (0.53–2.36) | 0.764 |
| Age | 1.000 (0.980–1.021) | 0.989 |
| Gender | 0.96 (0.58–1.58) | 0.867 |
| Final dofetilide dose | 1.000 (0.998–1.002) | 0.982 |
| Baseline QTc | 1.000 (0.993–1.007) | 0.972 |
| Type 2 diabetes mellitus | 0.62 (0.19–1.98) | 0.419 |
| Dyslipidemia | 1.10 (0.65–1.87) | 0.717 |
| Obesity | 1.22 (0.64–2.35) | 0.542 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
