Symptoms Burden as a Clinical Outcomes Assessment in Heart Failure Patients With Atrial Fibrillation

Figures

• 

  FIGURE 1. Cohorts Analyzed From the GENETIC-AF Parent Population AF = atrial fibrillation; AFB = atrial fibrillation burden; AFSQ = Atrial Fibrillation Symptoms Questionnaire; Dx = diagnosis; GENETIC-AF = Genotype-Directed Comparative Effectiveness Trial of Bucindolol and Toprol-XL for Prevention of Atrial Fibrillation/Atrial Flutter in Patients with Heart Failure; HF = heart failure; HFmprEF = heart failure with both mildly reduced and preserved left ventricular ejection fraction; LVEF = left ventricular ejection fraction; PTP = precision phenotyped.

• 

  FIGURE 2. Structural Equation Model Using a Priori Specified Factor Structure of AF and HF Symptomatology in the Psychometric Cohort The table insert contains regression coefficients and P values for AF or HF symptoms and 3 AF clinical scenarios, as described in the Methods. Indices for model fit at the bottom on the figure are the comparative fit index (CFI), Tucker-Lewis index (TLI), root mean square error of approximation (RMSEA), and standardized root mean square residual (SRMR). CPP = chest pain or pressure; CTC = confusion or trouble concentrating; FOT = fatigue or tiredness; LDF = lightheadedness, dizziness, or fainting; PAL = heart palpitations; SLF= swelling of both legs and/or feet; SOB = shortness of breath; SUA = sweating unrelated to physical activity; WG2 = weight gain ≥2 pounds; WPE = weakness or problems exercising; other abbreviations as in Figure 1.

• 

  FIGURE 3. Genetic-AF Cohort Patients (A) Covariate-adjusted time to first symptomatic AF/atrial flutter (AFL) event or all-cause mortality (ACM) curves. (B) Atrial fibrillation symptoms burden (SxBAF) plots, with the same subjects as in panel A. (C) SxBAF in GENETIC-AF cohort patients plotting only the 5 AFSQ AF symptoms. (D) SxBAF in the GENETIC-AF cohort, plotting only the 5 AFSQ HF symptoms. In A to D, there was 1 ACM in the metoprolol group, on day 54. The incidence rate ratio (IRR) is reported for bucindolol/metoprolol analyses. SxBAF curves plotted here are cubic b-spline estimates, knotted monthly, of average daily SxBAF by treatment over time with accompanying 95% CIs. F/U = follow-up; other abbreviations as in Figure 1.

• 

  Figure 4:. PTP Cohort Patients (A) Covariate-adjusted time to first symptomatic AF/AFL event or ACM in the PTP cohort, baseline LVEF 0.10 to 0.55. PTP patients are those meeting the PTP entry criteria that were generated by a predefined multivariate analysis in the GENETIC-AF trial.15 (B) Covariate-adjusted SxBAF plots in the same (PTP) cohort as in panel A. (C) SxBAF in PTP cohort patients plotting only the 5 AFSQ AF symptoms. (D) SxBAF in the PTP cohort, plotting only the 5 AFSQ HF symptoms. In A to D, there was 1 ACM in the metoprolol group, on day 54. The incidence rate ratio (IRR) is reported for bucindolol/metoprolol analyses. Abbreviations as in Figures 1 and 3.

• 

  Figure 5.. HFmprEF/PTP Cohort Patients (A) Covariate-adjusted time to first symptomatic event in PTP patients from GENETIC-AF cohort with LVEF of 0.40 to 0.55 (HFmprEF/PTP). (B) AF symptoms burden (SxBAF) in the same entry criteria as in panel A. Daily average SxBAF scores for each group are plotted using cubic b-splines (knotted monthly starting on day 1) of structural equation modeling 95% CI. In A and B, there was 1 all-cause mortality (ACM) in the metoprolol group, on day 54. (C) SxBAF plotting only the 5 AFSQ AF symptoms, HFmprEF/PTP patients. (D) SxBAF plotting only the 5 AFSQ HF symptoms. In A to D, there was 1 ACM in the metoprolol group, on day 54. The incidence rate ratio (IRR) is reported for bucindolol/metoprolol analyses. Abbreviations as in Figures 1 and 3.

• 

  Figure 6.. Total Symptoms Burden (SxBT) Plots in PTP or HFmprEF Cohort Patients Without or With AF (A) SxBAF plots in PTP cohort patients who did not experience an AF event during the efficacy F/U period, using all AFSQ data. (B) SxBAF plot in HFmprEF/PTP subjects who did not experience an AF episode during the efficacy F/U period, using all AFSQ data. (C) SxBT plots in entire PTP cohort patients using all AFSQ data, including those outside the AF episode window. (D) SxBT plot in HFmprEF/PTP patients. Abbreviations as in Figures 1 and 3.