Detection of atrial fibrillation recurrences after ablation: long-term intermittent versus short continuous heart rhythm monitoring
Abstract Background Catheter based ablation therapy is an established treatment in patients with symptomatic atrial fibrillation (AF). Currently, the detection of AF recurrences after AF ablation is most frequently performed by short continuous heart rhythm monitoring, such as Holter monitoring. Our aim is to evaluate the effectiveness and accurateness of long-term intermittent rhythm monitoring using a single-lead ECG (AliveCor Kardia®) compared to short continuous monitoring with Holter for the detection of AF recurrences after ablation. Methods We conducted a prospective study of patients after AF ablation between May 2017 and October 2019. As standard of care, patients underwent Holter monitoring (minimum 24 hours) at 3, 6 and 12 months after ablation. At the same time patients were instructed to use an AliveCor Kardia (ACK) monitor to record a 30-second ECG three times a day and in case of symptoms for a period of four weeks. The primary endpoint was the difference in proportion of AF recurrences detected by ACK compared to Holter. Secondary endpoint was the sensitivity and specificity of ACK algorithm. Further, patients were asked to complete a questionnaire on the System Usability Scale to evaluate the usability for both ACK and Holter. Results Out of 126 post-ablation patients, 115 (91%) patients (35 female, age 63±8 years) transmitted their ACK recordings and were included in this analysis. A total of 7838 ECGs were assessed. The mean AKC usage time was 27+11 days, mean number of recordings 68+28 per patient. Our primary endpoint, the proportion of detection of recurrent AF, was almost twice as high in ACK (24%) than in Holter (14%, p<0.05). The ACK algorithm categorized 80% as normal sinus rhythm, 10% as possible AF, 10% as unclassified and 0.6% as unreadable. According to the interpretation of the researchers' team, 38 (0.5%) ECGs were uninterpretable and in 98% of the unclassified ECGs by ACK, the researchers' team was able to establish a diagnosis. The ACK diagnostic algorithm displayed a sensitivity of 95% and specificity of 98% for AF detection. The ACK diagnostic algorithm had a high likelihood of misclassifying premature atrial contractions and sinus rhythm accompanied by artifacts as AF (6.8% and 11% of all recordings assessed by AKC as AF, respectively). Based on the System Usability Scale, patients rated ACK as more acceptable in daily usage than Holter (75.0% versus 58.6% had an overall score above 70%, respectively). Conclusions ACK effectively and accurately detects AF recurrences in patients who underwent AF ablation and has a high patients' acceptability compared to Holter monitoring. Long-term intermittent rhythm monitoring may provide a promising tool for rhythm follow-up after AF ablation procedures. Interpretation of ACK recordings Funding Acknowledgement Type of funding source: None