Absence of correlation between symptoms and rhythm in "symptomatic" atrial fibrillation.

BACKGROUND: Symptoms are widely used as a means of assessment and follow-up of patients with atrial fibrillation. This study assessed the correlation between symptoms and cardiac rhythm in patients being evaluated for operative therapy for atrial fibrillation. METHODS: Seven days of preoperative continuous outpatient home electrocardiographic monitoring was performed on 50 patients with symptomatic atrial fibrillation. Cardiac rhythm was continuously monitored automatically, while patients recorded their symptoms electronically. Correlations were then drawn between symptomatic events and actual rhythm, and between atrial fibrillation episodes and symptoms. RESULTS: Fifty patients (37 men) with symptomatic atrial fibrillation were monitored for a combined 356 days (mean, 7.1 days). Patients were average age of 69 years old. Intermittent atrial fibrillation was reported by 36 patients, and 14 believed their atrial fibrillation was continuous. During monitoring, all patients had periods of both atrial fibrillation and normal sinus rhythm. Of the 552 documented episodes of atrial fibrillation, 467 (85%) were asymptomatic, and 85 (15%) episodes were symptomatic. Patients indicated that they experienced atrial fibrillation symptoms 163 times. Of the 163 symptomatic events, 85 (52%) were actual atrial fibrillation, 64 (42%) were sinus rhythm, and 14 (6%) were other rhythms. The ability of an individual patient to accurately identify atrial fibrillation ranged from 0% to 100%. CONCLUSIONS: Patient-reported symptoms of atrial fibrillation had poor correlation with actual rhythm. The lack of correlation between symptoms and rhythm underscores the importance of continuous home monitoring for accurately quantifying preoperative atrial fibrillation burden and for postoperative follow-up.