Paroxysmal Atrial Fibrillation Complicating Acute Myocardial Infarction


Source: Eldar M, et. al. Circulation 1998;97:965-970.

Synopsis: Paroxysmal atrial fibrillation remains a significant marker for increased risk in the thrombolytic era.

The incidence and significance of paroxysmal atrial fibrillation in patients with acute myocardial infarction, in the time periods before and after the common use of thrombolysis, is of interest. In this study, the pre-thrombolytic era patients were extracted from a large multicenter trial of nifedipine in acute myocardial infarction that was conducted in Israel between 1981 and 1983. Surveys were done in January and February of 1992, 1994, and 1996 in 25 coronary care units in Israel, and these patients comprised the thrombolytic era group. Patients with chronic atrial fibrillation were excluded from both groups. Paroxysmal atrial fibrillation had to occur during the CCU portion of the hospital stay, but the duration of the episodes is not specified in the paper.

Paroxysmal atrial fibrillation occurred in 255 of 2936 patients (8.9%) during the thrombolytic era. Approximately 50% of all patients in the more recent surveys received thrombolytic therapy. The incidence of paroxysmal atrial fibrillation was lower among those who received thrombolytic therapy (6.7% vs 10.8%). Patients with paroxysmal atrial fibrillation were older, more likely to be female, and had more co-morbidities than patients without paroxysmal atrial fibrillation. Interestingly, the incidence of paroxysmal atrial fibrillation declined during the three periods analyzed. In 1992, the incidence was 10.2%; in 1994, the incidence was 8.9%; and, in 1996, the incidence was 7.6%. Paroxysmal atrial fibrillation was associated with an increased incidence of stroke-3.9% in those with paroxysmal atrial fibrillation vs. 0.6% in those without. Overall mortality was also increased among the patients with paroxysmal atrial fibrillation. The mortality rates for patients with and without paroxysmal atrial fibrillation at 30 days and one year were 25.1% vs. 10.5% and 38.4% vs. 15.2%, respectively. The relative risk for mortality in the thrombolytic era for patients with paroxysmal atrial fibrillation after adjustment for other significant predictors of mortality was 1.32 at 30 days and 1.33 at one year.

The incidence of paroxysmal atrial fibrillation was similar in the pre-thrombolytic era group. Patients with paroxysmal atrial fibrillation had a similar incidence of complications including other arrhythmias, congestive heart failure, cardiogenic shock, and cerebrovascular accidents during the two time periods.

Eldar and associates conclude that paroxysmal atrial fibrillation remains a significant marker for increased risk in the thrombolytic era. Although the adjusted contemporary mortality of paroxysmal atrial fibrillation is lower than previously observed, its relative importance as a risk factor remains unchanged.


This interesting paper shows that paroxysmal atrial fibrillation in the acute phase of myocardial infarction remains a significant problem. It is interesting that the overall incidence of paroxysmal atrial fibrillation was similar in both the pre-thrombolytic and thrombolytic eras. However, myocardial infarctions in the more recent surveys occurred in older and sicker patients. These factors were also associated with the occurrence of paroxysmal atrial fibrillation. Although thrombolysis seemed to reduce the incidence of atrial fibrillation, the differences in baseline characteristics resulted in a similar incidence in the two time periods.

It is also interesting that paroxysmal atrial fibrillation continues to have prognostic significance. Once again, much of the negative prognosis is associated with other risk factors that are linked to atrial fibrillation, but atrial fibrillation still had the same independent negative prognostic value in both periods. This is probably true since atrial fibrillation is likely to occur when the same level of electrophysiologic insult occurs and the adverse consequences of the arrhythmia are likely to be constant.

On the basis of this paper, it seems wise to take an aggressive approach to the evaluation and management of patients with paroxysmal atrial fibrillation. Concentrating therapeutic efforts in these highest risk patients will have the greatest possible yield. Unfortunately, the data presented don't let us know whether any specific antiarrhythmic strategy directed toward atrial fibrillation is of value. Eldar et al do not mention whether the patients were in atrial fibrillation at the time they left the coronary care unit, so it is impossible to make any conclusions about effects, either beneficial or harmful, of antiarrhythmic interventions.