Survival in Patients Presenting with Atrial Fibrillation: The Atrial Fibrillation Follow-Up Investigation of Rhythm Management (AFFIRM) Study
D. George Wyse
University of Calgary
Calgary, AB, Canada

The AFFIRM trial compares mortality risk among approximately 4,000 atrial fibrillation patients randomized to the popular rhythm control approach or to rate control. Primary results of this large trial show fewer deaths in the rate control arm. This result did not quite achieve statistical significance. However, the results suggest rate control is at least as beneficial as rhythm control in these patients. In addition, the rate control approach may be more cost-effective and may provide better quality of life.

The 2 treatment strategies for patients with recurrent atrial fibrillation that has not yet become permanent are sinus rhythm control or ventricular rate control.

Most North American physicians prefer to start with rhythm control. Presumed benefits include lower risk of stoke, improved survival, and fewer patient symptoms. Also, some physicians feel they can discontinue anticoagulation if treatment results in maintenance of sinus rhythm. These presumed benefits remain largely unproven.

Atrial Fibrillation Follow-Up Investigation of Rhythm Management (AFFIRM) is the largest investigation to date of rate versus rhythm control in atrial fibrillation. It is also among the first using mortality as a primary endpoint.

Patients in the AFFIRM trial had ECG documented atrial fibrillation. They had to have at least 1 risk factor for stroke or death, such as advanced age, hypertension or diabetes. At ACC, investigators presented follow up available for 98% of the 4,060 patients they enrolled. The mean follow up was 3.5 years. There were no differences in patient characteristics between the rate and rhythm control arms.

After randomization to rate or rhythm control, investigators chose the therapy they felt was appropriate. They used non-pharmacologic therapies if pharmacologic therapies were unsuccessful. Patients could cross over from rate control to rhythm control, and vice versa.

Initial therapy in the rate arm included digoxin, beta-blockers and calcium channel blockers. The most common initial therapy in the rhythm control arm was amiodarone or sotalol. During the study, there were frequent changes in therapy.

Primary results of AFFIRM showed a late trend toward increased all cause mortality in the rhythm control arm. Beginning at 18 to 24 months, mortality curves began to separate in favor of the rate control arm. At the end of follow-up, there were 306 deaths in the rate control arm and 356 deaths in the rhythm control arm (P = 0.058).

Adverse Events
-
Rate Control
Rhythm Control
P-Value
Death
306 (26%)
356 (27%)
0.058
Torsades de pointes VT
2 (0.2%)
13 (0.8%)
0.004
Sustained VT/VF cardiac arrest
24 (1.7%)
18 (1.2%)
0.355
Bradycardia cardiac arrest
2 (0.1%)
13 (0.8%)
0.004
Intracranial bleed
31 (2.0%)
29 (2.1%)
0.799
Major hemorrhage
106 (7.6%)
96 (6.9%)
0.473
Patients hospitalized after baseline
1,218 (70%)
1,375 (78%)
<0.001
Ischemic stroke
INR 2.0
INR < 2.0
Not taking warfarin
AF at time of event
79 (5.7%)
24 (30%)
28 (35%)
26 (33%)
45 (69%)
84 (7.3%)
18 (22%)
17 (20%)
48 (58%)
25 (36%)
0.680
Event rates derived from Kaplan-Meier analysis

Interestingly, strokes were most common in patients who stopped anticoagulation or received a sub-therapeutic dose. This suggests it may be important to maintain anticoagulation therapy in patients with atrial fibrillation and stroke risk factors.

Hospitalizations after baseline were significantly less frequent in the rate control arm. Because hospitalization can be a surrogate for cost, rate control may be more cost effective than rhythm control.

There were significantly fewer cases of bradycardia cardiac arrest and torsades de pointes in the rate control arm. There were no significant differences in intracranial bleeding, major hemorrhage, or sustained ventricular tachycardia/ventricular fibrillation cardiac arrest.

The AFFIRM study demonstrates that rate control is an effective primary therapy. The results did not support any of the presumed benefits of maintaining sinus rhythm. The trends in mortality and hospitalization may actually be benefits of the rate control approach.


Reporter: Andrew Bowser