Background:
In the AUGUSTUS trial, apixaban resulted in less bleeding and fewer hospitalizations than vitamin K antagonists (VKA), and aspirin caused more bleeding than placebo in patients with atrial fibrillation and acute coronary syndrome or percutaneous coronary intervention treated with a P2Y
12
inhibitor. We evaluated the risk-benefit balance of antithrombotic therapy according to kidney function.
Methods:
In 4456 patients, the CKD-EPI formula was used to calculate baseline estimated glomerular filtration rate (eGFR). The effect of apixaban vs. VKA and aspirin vs. placebo was assessed across kidney function categories using Cox models. The primary outcome was ISTH major or clinically relevant non-major bleeding. Secondary outcomes included death or hospitalization and ischemic events (death, stroke, myocardial infarction, stent thrombosis [definite or probable], or urgent revascularization). Creatinine clearance below 30 mL/min was an exclusion criterion in the AUGUSTUS trial.
Results:
Overall, 30%, 52%, and 19% had an eGFR of >80, >50 to 80, and 30-50 mL/min/1.73m
2
, respectively. During 6-months follow-up a total of 543 primary outcomes of bleeding, 1125 death or hospitalizations, and 282 ischemic events occurred. Compared with VKA, patients assigned apixaban had lower rates for all 3 outcomes across most eGFR categories without significant interaction. The absolute risk reduction with apixaban was most pronounced in those with an eGFR of 30-50 mL/min/1.73m
2
for bleeding events with rates of 13.1% vs. 21.3%; HR (95% CI) 0.59 (0.41-0.84). Patients assigned aspirin had a higher risk of bleeding in all eGFR categories with an even greater increase among those with eGFR >80 mL/min/1.73m
2
: 16.6% vs. 5.6%; HR 3.22 (2.19-4.74); p for interaction=0.007). The risk of death or hospitalization and ischemic events were comparable with aspirin and placebo across eGFR categories with HR ranging from 0.97 (0.76-1.23) to 1.28 (1.02-1.59) and from 0.75 (0.48-1.17) to 1.34 (0.81-2.22), respectively.
Conclusions:
The safety and efficacy of apixaban was consistent irrespective of kidney function, as compared with warfarin, and in accordance with the overall trial results. The risk of bleeding with aspirin was consistently higher across all kidney function categories.
Clinical Trial Registration:
URL: https://www.clinicaltrials.gov Unique Identifier: NCT02415400
Comparison of the Short-Term Risk of Bleeding and Arterial Thromboembolic Events in Nonvalvular Atrial Fibrillation Patients Newly Treated With Dabigatran or Rivaroxaban Versus Vitamin K Antagonists
