Frequent premature atrial complexes and risk of concurrent atrial fibrillation

Abstract

Abstract Background Premature atrial complexes (PACs) predict incident atrial fibrillation (AF) in long-term follow-up studies. It is unclear whether frequent PACs on ambulatory ECG recordings indicate a higher likelihood of concurrent, undiagnosed AF. Furthermore, the reproducibility of a 24 h PAC count is unclear. Objectives To determine if frequent PACs on 24 h ambulatory ECG monitoring predicts concurrent AF during subsequent, prolonged ECG monitoring and to assess the diagnostic reliability of PAC counts on a 24 hour ECG recording. Methods AF was defined as ≥30 seconds of irregular rhythm without P waves, which was detected by a proprietary algorithm and manually verified. The proportion of AF occurrence during the remainder of the monitoring period was calculated for pre-specified levels of PACs during the first 24 h, and a function describing the association was fitted. The diagnostic reproducibility of a 24 h PAC count was assessed by calculating the likelihood of a PAC count ≥1000/day during the entire monitoring duration for prespecified PAC count levels during the first 24h. Results The study population comprised 20,973 patients (41% men, mean age 69.5 years) who had recorded an ambulatory ECG with a monitoring duration of 4–30 days in the United States during the year 2017 (median monitoring duration 16 days). AF was detected in 2,029 (9.7%) of patients and the median time to first occurrence of AF was 5 days. PAC frequency during the first 24 h was associated with AF during the monitoring period beyond the first 24 h, increasing steadily from 4.2% among those with 0–5 PACs, to a plateau around 17% among those with 250–1000 PACs per day and above. (Fig. 1A). The reproducibility of low PAC counts was good. Only 5.5% of patients with 0–5 PACs during the first 24 h of monitoring (31.8% of the population), had ≥1000 PACs on an alternate monitoring day. In contrast, among subjects with 100 PACs the probability of a day with ≥1000 PACs was close to 50% (Fig. 1B). Conclusion In patients undergoing ambulatory ECG monitoring, frequent PACs during the first 24 h indicate a higher likelihood of AF occurrence during subsequent days of monitoring. Less than 5 PACs during the first 24 h indicate a low probability of AF or frequent PACs on a subsequent day of ECG monitoring. Figure 1 Funding Acknowledgement Type of funding source: Foundation. Main funding source(s): The Swedish Heart and Lung Foundation; The Swedish Heart and Lung Association