Assessing Heart Rate Fragmentation to Predict Atrial Fibrillation in the General Population Aged 65

Abstract Rationale Screening the general population aged 65 for atrial fibrillation (AF) has been proposed as a preventive measure against potential complications. The dysfunction of the autonomic nervous system (ANS) plays an important role in atrial fibrillation (AF). Short-term HRV is conceptually attributed to physiologic vagal tone modulation. But recent data suggest that a component of short-term HRV, called heart rate fragmentation (HRF), is related to ANS breakdown and may be a predictor of AF occurrence. Metrics derived from heart rate variability (HRV) that depict heart rate fragmentation (HRF) have been suggested to reflect autonomic nervous system (ANS) dysfunction. Objectives To assess the predictive capacity of HRF markers for AF occurrence over an 18-year follow-up in individuals aged 65 at the study’s inspection and to develop a predictive score for AF onset. Methods A cohort study was designed to prospectively assess the predictive value of ANS activity level among general population regarding cardiovascular events and mortality. 1011 subjects aged 65 were enrolled in the study from September 2000 to December 2002. The study population had no history of AF and a low cardiovascular risk. A median follow-up of 17.8 years was managed. HRV data were acquired by 24-h Holter electrcardiogram (ECG) monitoring at baseline and assessed by frequency domain and time domain methods. Heart rate fragmentation using the percentage of inflection points (PIP) and alpha 1-index (using fractal evaluation of NN intervals) were then calculated. The study outcome was the development of AF during follow-up, defined as a standard 12-lead ECG recording or Holter recording with ≥ 30 s of AF. Results AF was diagnosed in 123 patients during the 18-year follow-up, representing a cumulative incidence of 13.4%. Using multivariate analysis, the occurrence of AF was independently predicted by the male gender (HR=1.74, 95%CI:1.16-2.63, p<0.01), a heart rate below 70bpm (HR=1.85, 95%CI:1.18-2.90, p<0.01), an increased percentage of NN intervals that differ by more than 50 ms from the previous interval (pnn50) (HR=1.69, 95%CI:1.06-2.70, p=0.03), an increased percentage of inflection points (HR=2.80, 95%CI:1.66-4.72,p<0.01), and a reduced alpha 1 index (HR=1.61, 95%CI:1.03-2.53, p=0.04). Given these results, a predictive score was built up based on the 4 independent predictors, ranging from 0 to 6 (Graphical Abstract). The high-risk group, defined as having a PROOF-AF score of 4 or higher, exhibited a 3.3-fold higher risk of developing AF (Figure). The PROOF-AF score demonstrated superior predictive performance compared to other available risk scores for incident AF such as CHARGE-AF score, C2HEST score, FHS score, and JMC score. Conclusion HRF parameters, included in the PROOF-AF score, may be used to identify individuals aged 65 without pre-existing health conditions who are at high-risk of developing AF and assist population screening.Graphical abstractFigure