A 77-year-old woman presents to the emergency department complaining of dark-colored stools and lethargy. She was in her usual state of health until approximately one week ago when she began feeling fatigued and lightheaded. Since that time, she has had up to four to five black, tarry stools per day. She has never experienced anything like this in the past. Today, she felt too weak to stand, prompting her family to bring her to the emergency department for urgent evaluation.
Her past medical history is notable for heart failure with preserved ejection fraction, permanent atrial fibrillation (AF), hypertension, arthritis, and type 2 diabetes mellitus. Her home medications include warfarin (target international normalized ratio [INR] 2.0-3.0), furosemide, metoprolol, metformin, and naproxen as needed for joint pain.
Vital signs are notable for an irregularly irregular heart rate of 126 beats per minute (BPM) and a blood pressure of 78/46 mm Hg. Her height and weight are 62 inches and 68 kg, respectively. Laboratory values are notable for a creatinine of 0.58, an INR of 4.8, and a hemoglobin of 6.8 mg/dL. A review of prior laboratory values reveals a baseline hemoglobin of approximately 11 mg/dL and a long history of labile INRs, with frequent supratherapeutic levels. Her electrocardiogram shows AF with a rapid ventricular rate. Rectal exam reveal dark, guaiac positive stool.
She is urgently stabilized with intravenous fluids, packed red blood cells, and fresh frozen plasma. She later undergoes an upper endoscopy, which reveals a duodenal ulcer that is treated with endoscopic clipping.
She is discharged several days later off anticoagulation and follows up with her physician in clinic. Her hemoglobin has returned to her baseline and she remains hemodynamically stable without further evidence of gastrointestinal (GI) bleeding. She and her physician engage in a discussion regarding the complexity of restarting anticoagulation. Her physician explains how anticoagulants that reduce the risk of stroke will increase the risk of having another GI bleed. The patient expresses an understanding of the competing risks and explains that she is most concerned about having a stroke. She requests an anticoagulation strategy that will reduce her risk of stroke, while minimizing as much as possible her risk of experiencing another bleeding event.
The correct answer is: D. Apixaban 5 mg twice-daily.
This patient has permanent AF and a CHA2DS2-VASc score of 6, corresponding to a 19.7% annual stroke risk. She has now suffered a GI bleed secondary to a duodenal ulcer. While endoscopic clipping and discontinuing her non-steroidal anti-inflammatories may reduce her bleeding risk, she remains at risk for recurrent bleeding events. Patients with elevated risk of stroke and recurrent GI bleeding represent a challenging clinical dilemma regarding if and how to restart anticoagulation.
In order to choose a treatment strategy that appropriately balances competing risks, it is essential to understand patients' preferences and priorities. The process begins by actively engaging patients in their treatment decisions and explaining the risks and benefits of treating each disease state. The optimal strategy for patients with identical risk profiles can vary dramatically. An elderly patient who prefers to avoid any hospitalization, a young athlete who values optimal function, and a Jehovah's Witness patient may all have different optimal therapies based on their values and priorities. In this case, the patient's priority is to reduce her risk of having a stroke. She understands that anticoagulation will increase her risk of bleeding and is willing to accept this risk. Her physician is now challenged to choose an anticoagulation strategy that can minimize the risk of GI bleeding among those therapies that provide maximal stroke reduction.
Warfarin vs. NOACs
The first question is whether or not warfarin should be continued versus switching to a non-vitamin K inhibiting oral anticoagulant (NOAC)? In a large meta-analysis of four NOACs (dabigatran, rivaroxaban, apixaban, and edoxaban) in patients with AF, NOACs were associated with a significant increase in GI bleeding (relative risk [RR], 1.25; 95% confidence interval [CI], 1.01-1.55; P = 0.043).1 In this case, the patient had a history of labile INRs and suffered her bleeding event while her INR was supratherapeutic. As such, it is reasonable to switch to a NOAC with more reliable dosing, as the prevention of supratherapeutic anticoagulation alone may help her avoid a future bleed.
Choosing Among the NOACs
The next question is: which NOAC will minimize the risk of GI bleeding? There are four NOACs currently approved by the U.S. Food and Drug Administration (FDA) for stroke prevention in AF: a direct thrombin inhibitor (dabigatran) and direct factor Xa inhibitors (rivaroxaban, apixaban, and edoxaban).
In the Randomized Evaluation of Long-Term Anticoagulation Therapy (RE-LY), Rivaroxaban Once Daily Oral Direct Factor Xa Inhibition Compared With Vitamin K Antagonism for Prevention of Stroke and Embolism Trial in Atrial Fibrillation (ROCKET-AF), and Effective Anticoagulation With Factor Xa Next Generation in Atrial Fibrillation–Thrombolysis in Myocardial Infarction 48 (ENGAGE AF-TIMI 48) trials, dabigatran 150 mg twice-daily, rivaroxaban 20 mg once-daily, and edoxaban 60 mg once-daily, respectively, were all found to be associated with significantly more GI bleeding when compared to warfarin (for dabigatran, 1.51% vs. 1.02%; RR, 1.36; 95% CI, 1.09-1.70; P <0.001; for rivaroxaban, 3.2% vs. 2.2%; P <0.001; for edoxaban, hazard ratio [HR], 1.23; 95% CI, 1.02-1.50; P = 0.03).2-4 Apixaban is the only full-dose NOAC not associated with a significant increase in GI bleeding when compared to warfarin. In the Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation (ARISTOTLE) trial, apixaban 5 mg twice-daily was associated with a non-significant reduction in GI bleeding when compared to warfarin (HR, 0.89; 95% CI, 0.70-1.15; P = 0.37).5
A large meta-analysis analyzed the GI bleeding risk of dabigatran, rivaroxaban, and apixaban compared with conventional anticoagulants among 91,933 patients across 17 randomized clinical trials (RCTs). The preliminary results were presented at the 2014 American College of Cardiology (ACC) Scientific Sessions.6 There was an approximately 20% relative risk increase of GI bleeding with NOACs when evaluating all three agents in aggregate. Individually, the risk of GI bleeding was significantly higher with dabigatran (Peto Odds Ratio [POR], 1.31; 95% CI, 1.10-1.57) and rivaroxaban (POR, 1.35; 95% CI, 1.16-1.57). However, apixaban showed a non-significant reduction in GI bleeding (POR, 0.85; 95% CI, 0.67-1.08).
The available data suggest that the risk of GI bleeding with apixaban 5 mg twice-daily is at least not significantly higher than with warfarin, and may be lower. Among the four NOACs approved for stroke prevention in AF, apixaban appears to have the most favorable safety profile with respect to GI bleeding.
Each of the four NOACs discussed above have options for reduced dosing. Dabigatran is dosed at 75 mg twice-daily for a creatinine clearance of 15-30 mL/min. Outside the U.S., many countries have approved a 110 mg twice-daily dose, which was non-inferior to warfarin for stroke prevention with an improved safety profile in the RE-LY trial. However, this dose is not approved by the FDA.2 Edoxaban is dosed at 30 mg once-daily for a creatinine clearance of 15-50 mL/min. Rivaroxaban is dosed at 15 mg once-daily for a creatinine clearance of 15-50 mL/min. Apixaban is dosed at 2.5 mg twice-daily for patients with two of the following factors: age ≥80 years, body weight ≤60 kg, or serum creatinine ≥1.5 mg/dL.
There is limited data regarding the efficacy and safety of reduced-dose NOACs in patients without indications for reduced dosing. In the RE-LY trial,2 dabigatran 110 mg twice-daily was found to be non-inferior to warfarin with a statistically similar risk of GI bleeding (RR, 1.10; 95% CI, 0.86-1.41; P = 0.43). Despite the improved safety profile of the 110 mg dose, the FDA opted against approval due to an inability to identify any particular subgroup that would benefit from its use versus the 150 mg dose.7 In the ENGAGE AF-TIMI 48 trial, the lower dose regimen of edoxaban (30/15 mg once-daily) was associated with significantly less GI bleeding when compared to warfarin (HR, 0.67; 95% CI, 0.53-0.83; P <0.001), however this lower dose regimen was not as effective as warfarin to prevent ischemic stroke.3
Despite the limited data, patients at high risk for bleeding without indications for dose reduction are sometimes managed with NOACs at reduced doses. In patients at high risk for bleeding who desire an anticoagulation strategy that offers some reduction in stroke risk with less bleeding risk than full dose anticoagulation, reduced-dose NOACs are very reasonable options, with the most promising data supporting edoxaban 30 mg once-daily or apixaban 2.5 mg twice-daily.
Back to the Case
At 77 years old, with a creatinine clearance of approximately 87 mL/min. and a body weight of 68 kg, the patient in this case does not have an indication for dose reduction of any of the four NOACs. Given her expressed desire to not compromise stroke prevention despite her GI bleeding risk, as well her prior bleed occurring on supratherapeutic anticoagulation, the most appropriate choice is full-dose anticoagulation with apixaban 5 mg twice-daily, the NOAC with the most favorable safety profile at full dose.
Should she develop recurrent bleeding on apixaban 5 mg twice-daily or desire a strategy that provides a better balance between some reduction in stroke and a more favorable safety profile, it would be reasonable to switch to a reduced-dose NOAC, such as edoxaban 30 mg once-daily or apixaban 2.5 mg twice-daily, although this strategy of reducing the dose of a NOAC after a bleeding event has not been rigorously evaluated in clinical trials.
- Ruff CT, Giugliano RP, Braunwald E, et al. Comparison of the efficacy and safety of new oral anticoagulants with warfarin in patients with atrial fibrillation: a meta-analysis of randomised trials. Lancet 2014;383:955-62.
- Connolly SJ, Ezekowitz MD, Yusuf S, et al. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med 2009;361:1139-51.
- Giugliano RP, Ruff CT, Braunwald E, et al. Edoxaban versus warfarin in patients with atrial fibrillation. N Engl J Med 2013;369:2093-104.
- Patel MR, Mahaffey KW, Garg J, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med 2011;365:883-91.
- Granger CB, Alexander JH, McMurray JJ, et al. Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med 2011;365:981-92.
- Sardar P, Chatterjee S, DiNicolantonio J, Wetterslev Jr, Gluud C, Bangalore S. New oral anticoagulants and gastrointestinal bleeding: insights from meta-analyses and trial sequential analyses of randomized clinical trials. J Am Coll Cardiol 2014;63(12_S):A333.
- Beasley BN, Unger EF, Temple R. Anticoagulant options--why the FDA approved a higher but not a lower dose of dabigatran. N Engl J Med 2011;364:1788-90.