Background: In-hospital atrial fibrillation (AF) often occurs at the acute worsening phase of heart failure (HF). In-hospital AF terminates spontaneously following the recovery of HF in some cases, but administration of anti-arrhythmic agents or electrical defibrillation is performed in other cases. The aim of this study was to investigate the impact of the in-hospital onset AF on the clinical outcome in patients with acute HF.
Methods: Study subjects included 551 new onset acute HF patients referred to our hospital (85 males, 66±14 years old) and classified into 4 groups: group 1 (n = 243), no AF was detected during hospital stay and after discharge from the hospital; group 2 (n = 35), new onset AF was detected during hospital stay without recurrence of AF after discharge; group 3 (n = 47), AF occurrence only after the discharge without AF detection during hospital stay; and group 4 (n = 226), the clinical history of AF was confirmed only before admission. The cardiac death after discharge was compared among the 4 groups. Cox regression analysis was performed for the prediction of cardiac death.
Results: In group 2, new in-hospital onset AF occurred 7 ± 6 days after admission, and two-thirds of new in-hospital onset AF terminated spontaneously. During the mean follow up periods of 1512 days (ranging from 2 to 2667 days), 117 cardiac deaths occurred. In Kaplan-Meyer analysis, group 2 showed higher survival rate than groups 3 and 4 (P = 0.049 and 0.014, respectively). There was no significant difference in survival rate between groups 1 and 2 (P = 0.068). In group 2, the survival rate did not differ between patients who required intervention for in hospital AF or not (P = 0.513). Finally, Cox proportional hazard analysis revealed that group 2 had a lower risk of cardiac death in acute HF patients (hazard ratio 0.205, 95% confidence interval 0.05 to 0.84, P<0.028).
Conclusion: In-hospital onset AF terminated spontaneously in most of cases and showed better clinical outcome with or without intervention, compared with the AF occurring before admission and after discharge. On the other hand, new onset AF after discharge showed worse cardiac survival. Therefore, the intervention for the in-hospital onset AF in patients with acute HF might not be necessary until AF occurrence.