Two-Year Follow-Up of Patients With Atrial Fibrillation Receiving Edoxaban in Routine Clinical Practice: Results From the Global ETNA-AF Program

Abstract

Background: Randomized clinical trials demonstrated similar efficacy and improved safety of direct oral anticoagulants versus warfarin in patients with atrial fibrillation (AF). Long-term data in routine clinical practice are needed.

Hypothesis: Patients with AF receiving edoxaban at baseline continue to have low annualized effectiveness and safety event rates in the second year of follow-up, with regional variations observed.

Methods: The Global ETNA-AF program is a prospective, noninterventional study of patients with AF receiving edoxaban. Patient characteristics and annualized clinical event rates were assessed overall and by region across the 2-year follow-up. Annualized event rates of bleeding and thromboembolic events were assessed within the first year and conditionally in patients who were event-free up to 12 months in the second year.

Results: This analysis comprised 26 805 patients from Europe (n = 13 164), Japan (n = 10 342), and non-Japanese Asian regions (n = 3299). Patients from Europe had the highest burden of comorbidities. The annualized event rates for major bleeding, any stroke, all-cause death, and cardiovascular death varied by region. The global annualized event rates in the first and second year were 1.31%/year and 0.86%/year for major bleeding, 1.06%/year and 0.65%/year for any stroke, 0.84%/year and 0.73%/year for cardiovascular death, and 3.05%/year and 3.18%/year for all-cause death.

Conclusion: Annualized event rates for any stroke and major bleeding remained low through 2-year follow-up for patients with AF receiving edoxaban at baseline. Differences in annualized event rates for all-cause and cardiovascular mortality between Europe, Japan, and non-Japanese Asian regions may reflect variations in baseline characteristics.

Trial registration: Europe, NCT02944019; Japan, UMIN000017011; Korea/Taiwan, NCT02951039; Hong Kong, NCT03247582; and Thailand, NCT03247569.

Keywords: anticoagulation; atrial fibrillation; direct oral anticoagulant (DOAC); edoxaban; major bleeding; oral anticoagulants; stroke prevention.

Figure 1

Global effectiveness and safety annualized clinical event rates overall as well as in the first year (months 1–12) and second year (months 13–24). ^aCV mortality is defined as deaths due to CV‐related reasons plus deaths where there was a bleeding event with fatal outcome or where any stroke, TIA, SEE, PE, MI, VTE, or major bleeding occurred within 30 days before death and the death reason was missing or unknown. For all regions, it is censored by 730 days, study discontinuation, or last follow‐up, whichever comes first. ^bIncludes patients (n = 15) with unknown stroke type. CI, confidence interval; CV, cardiovascular; GI, gastrointestinal; ISTH, International Society on Thrombosis and Haemostasis; MI, myocardial infarction; PE, pulmonary embolism; SEE, systemic embolic event; TIA, transient ischemic attack; VTE, venous thromboembolism.

Figure 2

Cumulative incidence function curves from 0 to 24 months for (A) all‐cause mortality and CV mortality, (B) any stroke and ischemic stroke, (C) major bleeding and major GI bleeding, and (D) intracranial hemorrhage and hemorrhagic stroke. ^aCV mortality is defined as deaths due to CV‐related reasons plus deaths where there was a bleeding event with fatal outcome or where any stroke, TIA, SEE, PE, MI, VTE, or major bleeding occurred within 30 days before death and the death reason was missing or unknown. For all regions, it is censored by 730 days, study discontinuation, or last follow‐up, whichever comes first. CV, cardiovascular; GI, gastrointestinal; ISTH, International Society on Thrombosis and Haemostasis; MI, myocardial infarction; PE, pulmonary embolism; SEE, systemic embolic event; TIA, transient ischemic attack; VTE, venous thromboembolism.

Figure 3

Annualized clinical event rates^a for (A) major bleeding, (B) any stroke, (C) all‐cause mortality, and (D) CV mortality in Europe, Japan, and non‐Japanese Asian^b regions over the 2‐year follow‐up (months 1–24). ^aBar graphs show the annualized rate ± 95% CI. Clinical event counts (with the annualized rate shown in parentheses below them) are shown above each bar for each region. ^bNon‐Japanese Asian regions include Hong Kong, South Korea, Taiwan, and Thailand. ^cCV mortality is defined as deaths due to CV‐related reasons plus deaths where there was a bleeding event with fatal outcome or where a stroke, TIA, SEE, PE, MI, VTE, or major bleeding occurred within 30 days before death and the death reason was missing or unknown. For all regions, it is censored by 730 days, study discontinuation, or last follow‐up, whichever comes first. CI, confidence interval; CV, cardiovascular; GI, gastrointestinal; ISTH, International Society on Thrombosis and Haemostasis; MI, myocardial infarction; PE, pulmonary embolism; SEE, systemic embolic event; TIA, transient ischemic attack; VTE, venous thromboembolism.

Figure 4

Cumulative incidence function curves from months 0 to 24 by region for (A) all‐cause mortality, (B) CV mortality, (C) any stroke, (D) ischemic stroke, (E) major bleeding, (F) major GI bleeding, (G) intracranial hemorrhage, and (H) hemorrhagic stroke. ^aCV mortality is defined as death due to CV‐related reasons plus deaths where there was a bleeding event with fatal outcome or where any stroke, TIA, SEE, PE, MI, VTE, or major bleeding occurred within 30 days before death and the death reason was missing or unknown. For all regions, it is censored by 730 days, study discontinuation, or last follow‐up, whichever comes first. CV, cardiovascular; GI, gastrointestinal; ISTH, International Society on Thrombosis and Haemostasis; MI, myocardial infarction; PE, pulmonary embolism; SEE, systemic embolic event; TIA, transient ischemic attack; VTE, venous thromboembolism.