Estimated atrial fibrillation burden on early rhythm-control and cardiovascular events in the EAST-AFNET 4 trial

Figures

• 

  Fig. 1. CONSORT diagram. Patients enrolled in EAST-AFNET 4 and randomized to early rhythm control were asked to submit telemetric electrocardiograms (ECGs) twice per week. All ECGs were retrieved and analysed employing supervised artificial intelligence in a core lab. The primary analysis population included all patients of which at least two ECGs were available and classifiable for either atrial fibrillation (AF) or sinus rhythm (SR). The landmark analysis population included all of those patients, that had more than one ECG during the first year of follow-up until the landmark timepoint at 12 months. Abbreviations: AF = Atrial fibrillation; ECG = Electrocardiogram; ERC = Early rhythm-control; FU = Follow-up; IQR = Interquartile range; SR = Sinus rhythm; UC = Usual Care.

• 

  Fig. 2. Distribution of atrial fibrillation (AF) burden (A) and days in sinus rhythm (B) on rhythm-control therapy in the first year of follow-up in the EAST-AFNET 4 trial. A: AF burden in patients with early rhythm-control within the first year after randomization. B: Number of days in sinus rhythm in the interpolated dataset within the first year after randomization. AF burden was calculated as the number of ECGs in AF divided by the total number of available ECGs in the monitoring period and is displayed in percent (0–100) on the x axis. Similarly, days in sinus rhythm was calculated as the number of days with an ECG in sinus rhythm divided by the monitored time. The height of each column (y-axis) is proportional to the number of patients with a given AF burden or number of days in sinus rhythm. The box plot on top of the histogram shows the distribution of AF burden (A) and days in sinus rhythm (B) in a box-whisker plot. The thick lines indicate the median, boxes illustrate the interquartile range. Abbreviations: AF = Atrial Fibrillation; ECG = electrocardiogram; FU = follow-up; SR = sinus rhythm. Colors and labels refer to AF quartiles, so that patients in Q4 and displayed in red…

• 

  Fig. 3. Aalen-Johansen cumulative-incidence curves by atrial fibrillation (AF) burden quartile for the primary efficacy and safety outcomes in EAST-AFNET 4. AF burden was estimated by repeated patient-operated ECGs. Numbers give the AF burden percentage boundaries in the four quartiles. The uneven distribution of patients across quartiles is due to the high number of patients with an AF burden of 0%. Starting point of this landmark analysis is the 12-months follow-up. The first year of follow-up was used to determine AF burden. A: Incidence rates of the composite primary efficacy outcome by AF burden quartiles in EAST-AFNET 4 patients randomised to early rhythm-control. The primary outcome was a composite of cardiovascular death, stroke, or unplanned hospitalisation for heart failure or acute coronary syndrome. The estimated annualized event rates in the higher AF burden quartiles (Q3: 4.8/100 patient-years; Q4: 4.2/100 patient-years) are quite close to the estimated event rates in patients randomized to usual care (4.5/100 patient-years), while the estimated annualized event rates in Q1 and Q2 were much lower (2 and 2.6/100 patient-years). B: Incidence rates of the composite primary…