Library › Pharmacology › Direct oral anticoagulants (DOACs)

Direct oral anticoagulants: apixaban, rivaroxaban, dabigatran, and edoxaban

The direct factor Xa and thrombin inhibitors from mechanism and pharmacokinetics through dosing by indication, kidney and liver function, body weight, drug interactions, laboratory measurement, contraindications, switching, adherence, and the factor XI inhibitors in development.

The direct oral anticoagulants and their approved uses

Direct oral anticoagulants bind the active site of a single coagulation enzyme without a cofactor. Four are approved in the United States: dabigatran etexilate (approved for atrial fibrillation in 2010), rivaroxaban (2011), apixaban (2012), and edoxaban (2015). Each reached approval through large randomized trials against warfarin in atrial fibrillation (AF) and against warfarin or low-molecular-weight heparin in venous thromboembolism (VTE). Rivaroxaban, alone among them, also carries a low "vascular" dose for chronic coronary and peripheral artery disease.

The 2023 ACC/AHA/ACCP/HRS AF guideline recommends a DOAC over warfarin in eligible patients with AF who do not have a mechanical heart valve or moderate to severe rheumatic mitral stenosis (Class 1, Level A). The 2026 AHA/ACC acute pulmonary embolism guideline recommends a DOAC over a vitamin K antagonist when oral therapy is appropriate (Class 1). The CHEST and American Society of Hematology (ASH) VTE guidelines make the same choice. DOACs have a rapid onset, a fixed dose without routine laboratory monitoring, fewer food and drug interactions than warfarin, and, across the pivotal trials, about half the rate of intracranial hemorrhage.

Their limits follow from the same pharmacology. Short half-lives make adherence decisive, because a missed day leaves the patient unprotected. Renal clearance makes kidney function part of every dose decision. Standard coagulation tests do not quantify their effect. DOACs failed, or caused harm, where thrombosis is driven by a mechanical surface, by severe rheumatic valve disease, or by antiphospholipid antibodies.

US labeled adult indications from current prescribing information. Rivaroxaban, dabigatran, and apixaban also carry pediatric VTE indications with weight-based formulations; this page addresses adults only.

Mechanism of action: direct inhibition of factor Xa and thrombin

Coagulation converges on two enzymes. Factor Xa, assembled with factor Va on phospholipid surfaces as the prothrombinase complex, converts prothrombin (factor II) to thrombin. Thrombin cleaves fibrinogen to fibrin, activates platelets through protease-activated receptors, and amplifies its own generation by activating factors V, VIII, and XI. Because one molecule of factor Xa generates many molecules of thrombin, inhibition upstream reduces the thrombin burst, and inhibition of thrombin itself blocks the final effector.

Apixaban, rivaroxaban, and edoxaban are small molecules that bind reversibly to the active site of factor Xa. They inhibit free factor Xa and factor Xa within the prothrombinase complex and in clot, which antithrombin-dependent heparins reach poorly. Dabigatran is a reversible, competitive inhibitor of the active site of thrombin, both free and fibrin-bound. Dabigatran etexilate, the oral prodrug, has no anticoagulant activity until esterases convert it.

DOACs differ from the older anticoagulants in mechanism, onset, and offset. Heparin and low-molecular-weight heparin act indirectly through antithrombin. Warfarin lowers the functional concentrations of factors II, VII, IX, and X, and of proteins C and S, over several days by blocking vitamin K recycling. A DOAC acts within hours of the first dose, and its effect resolves within one to two days of the last. Most settings need no overlap at initiation, and there is no rebound procoagulant state from protein C depletion.

Pharmacokinetics of apixaban, rivaroxaban, dabigatran, and edoxaban

The four drugs differ in absorption, protein binding, metabolism, and renal clearance, and those differences explain nearly every dosing rule. Peak effect arrives 1 to 4 hours after a dose, so the time of the last dose determines the residual effect at the moment of bleeding, surgery, or a laboratory draw.

From current US prescribing information (section 12.3). P-gp, P-glycoprotein; BCRP, breast cancer resistance protein; CYP, cytochrome P450. Dabigatran removal depends on dialyzer flux and blood flow: about 49% at 200 mL/min and 57% at 300 mL/min in the label study.

Absorption. Apixaban, dabigatran, and edoxaban are absorbed equally well fasting or fed. Rivaroxaban 15 mg and 20 mg are not: food raises the exposure from the 20-mg tablet by 39% (maximum concentration by 76%), and these doses are taken with a meal, the evening meal for AF. The 2.5-mg and 10-mg tablets are fully absorbed without food. Dabigatran etexilate is a poorly soluble prodrug. The tartaric acid core of the capsule pellets creates an acidic microenvironment that aids its dissolution and absorption and contributes to dyspepsia. Its absolute bioavailability is still only 3% to 7%, because the prodrug undergoes P-glycoprotein efflux in the gut wall and must be hydrolyzed to active dabigatran. All four are absorbed in the stomach and proximal small bowel, so gastric bypass and extensive small-bowel resection create uncertainty about absorption.

Distribution and clearance. High protein binding makes apixaban and rivaroxaban essentially nondialyzable. Low protein binding and a moderate volume of distribution allow hemodialysis to remove dabigatran. Clearance falls with age and with kidney function, and dabigatran shows the steepest rise in exposure as creatinine clearance falls. Exposure rises in low body weight, and peak concentrations fall modestly in severe obesity.

Onset and offset. Therapeutic anticoagulation begins within hours of the first dose, which is why apixaban and rivaroxaban can start VTE treatment without a parenteral lead-in. After the last dose, anticoagulant effect falls by half every half-life; two to three half-lives remove most of it in patients with normal kidney function, which underlies the 1-day and 2-day interruption schedules before surgery.

Administration: food, crushed tablets, feeding tubes, and dabigatran capsules

Each label gives instructions for patients who cannot swallow tablets. Apixaban 2.5-mg and 5-mg tablets may be crushed and suspended in water, 5% dextrose, or apple juice, or mixed with applesauce, and given by mouth. They may also be crushed in 60 mL of water or dextrose and given through a nasogastric tube; the suspension is stable for 4 hours. Rivaroxaban tablets of every strength may be crushed and mixed with applesauce, or suspended in 50 mL of water for a nasogastric or gastric tube. The 15-mg and 20-mg doses are followed immediately by food or enteral feeding, and administration distal to the stomach is avoided because absorption falls. Edoxaban tablets may be crushed and mixed with 2 to 3 ounces of water and given by mouth or gastric tube, or mixed into applesauce and given by mouth, with exposure similar to the intact tablet.

Dabigatran capsules are swallowed whole with a full glass of water and are never opened, chewed, or emptied. The pellets taken without the capsule shell have about 75% higher bioavailability, which raises bleeding risk. Capsules are kept in the original blister or bottle to protect them from moisture. Dabigatran is therefore unsuitable for patients who depend on a feeding tube.

Gastrointestinal adverse reactions are most common with dabigatran. In RE-LY, they occurred in 35% of patients taking dabigatran 150 mg twice daily and in 24% taking warfarin, mostly dyspepsia and gastritis-like symptoms; in the VTE trials, dyspepsia occurred in 7.5% versus 5.5%. Taking the capsule with food lessens symptoms. Dyspepsia is a frequent reason for discontinuing dabigatran and switching to a factor Xa inhibitor.

Dosing for stroke prevention in atrial fibrillation

The AF doses are those tested in the pivotal trials, with reductions defined by each drug's own criteria. The criteria differ: apixaban uses a combination of age, weight, and serum creatinine; rivaroxaban, dabigatran, and edoxaban use creatinine clearance; edoxaban adds an upper clearance limit. Stroke risk scores determine whether to anticoagulate; they do not change the dose. CHA₂DS₂-VASc, CHA₂DS₂-VA, and HAS-BLED are described on the atrial fibrillation page.

US prescribing information and 2023 ACC/AHA/ACCP/HRS AF guideline, Table 19. Creatinine clearance by Cockcroft-Gault with actual body weight. The trials used broader reduction criteria for edoxaban (weight 60 kg or less and potent P-glycoprotein inhibitors), but the US AF label reduces the dose for kidney function only.

The apixaban criteria are often misapplied. A single criterion, such as a creatinine of 1.6 mg/dL in an otherwise robust 70-year-old, does not justify 2.5 mg twice daily. Conversely, an 82-year-old weighing 58 kg meets two criteria even with a normal creatinine. Apixaban dose reduction is an AF rule; it does not apply to acute VTE treatment, where 2.5 mg is used only for extended secondary prevention.

Edoxaban at high creatinine clearance. In ENGAGE AF-TIMI 48, patients with creatinine clearance above 95 mL/min had a higher rate of ischemic stroke with edoxaban 60 mg than with warfarin, because higher clearance lowers edoxaban concentrations. The US label carries a boxed warning against its use for AF above 95 mL/min, and another anticoagulant is used. The European label sets no fixed upper limit and advises use at high clearance only after careful evaluation of thromboembolic and bleeding risk.

Dose reduction for bleeding. The 2023 guideline rates doses not supported by trial evidence, including empiric reduction to lower bleeding risk, as harmful (Class 3: Harm). Bleeding risk is reduced by correcting modifiable factors: uncontrolled blood pressure, concomitant antiplatelet or nonsteroidal anti-inflammatory drugs, alcohol excess, and anemia.

Kidney function monitoring. Creatinine clearance is calculated before the first dose and reassessed at least yearly. The European Heart Rhythm Association (EHRA) practical guide sets the minimum recheck interval in months at creatinine clearance divided by 10 when clearance is 60 mL/min or less (for example, every 3 months at 30 mL/min). Acute illness, dehydration, contrast exposure, and new nephrotoxic drugs prompt an unscheduled check, because a falling clearance raises exposure without any change in the prescription.

Landmark trials of DOACs in atrial fibrillation

Four phase 3 trials enrolling more than 71,000 patients compared a DOAC with dose-adjusted warfarin. All four reduced intracranial hemorrhage, by about one-third with rivaroxaban and by about half or more with the other three. They differed in design (RE-LY was open-label for warfarin), in population risk (ROCKET AF enrolled the highest-risk patients), and in the quality of warfarin control.

CrCl, creatinine clearance. Rates are per year in the primary analyses. Across the four trials, every DOAC reduced intracranial hemorrhage compared with warfarin, and dabigatran 150 mg, rivaroxaban, and high-dose edoxaban increased gastrointestinal bleeding.

No adequately powered randomized trial has compared one DOAC with another. Observational comparisons in US claims data have generally associated apixaban with less major and gastrointestinal bleeding than rivaroxaban or dabigatran, which is consistent with the trial results against warfarin. These comparisons carry residual confounding, and drug choice depends on kidney function, interacting drugs, bleeding history, dosing frequency, cost, and patient preference. The full comparison of DOAC and warfarin strategies in AF, including device-detected AF, is on the atrial fibrillation page.

Dosing for venous thromboembolism treatment and extended secondary prevention

VTE treatment has three phases: an initial phase of up to 21 days with higher-intensity therapy, a primary treatment phase that brings total treatment to 3 to 6 months, and an extended phase without a planned stop date. Apixaban and rivaroxaban were tested as single-drug regimens that begin with an oral loading dose. Dabigatran and edoxaban were tested after at least 5 days of parenteral anticoagulation and are started only after that lead-in.

US prescribing information. The P-glycoprotein inhibitors that trigger the edoxaban 30-mg VTE dose are verapamil, quinidine, and short-term azithromycin, clarithromycin, erythromycin, oral itraconazole, or oral ketoconazole.

Extended-phase dose. After at least 6 months of full-dose treatment, the 2026 AHA/ACC pulmonary embolism guideline recommends reduced-dose apixaban (2.5 mg twice daily) or rivaroxaban (10 mg daily) over full doses to reduce bleeding in patients who continue anticoagulation (Class 1). The reduced doses were tested in patients who had completed treatment without recurrence; recurrence on a reduced dose is treated with the full dose. The reduced doses were not tested in patients at high recurrence risk, such as those with recurrent events or active cancer, except in the API-CAT trial in cancer. The decision to extend therapy is covered on the pulmonary embolism page.

Choice of agent. A DOAC is preferred over a vitamin K antagonist for VTE treatment (2026 AHA/ACC guideline, Class 1; CHEST 2021 and ASH 2020 guidelines). Apixaban and rivaroxaban suit outpatient treatment because they need no injections. Dabigatran and edoxaban suit patients who start on heparin in hospital. Warfarin remains the agent for thrombotic antiphospholipid syndrome and for patients in whom DOACs are contraindicated or unaffordable.

Landmark trials of DOACs in venous thromboembolism

The VTE trials compared each DOAC with parenteral heparin followed by warfarin. All were noninferior for recurrent VTE. Rivaroxaban in EINSTEIN-PE and apixaban in AMPLIFY reduced major bleeding. Dabigatran in RE-COVER did not significantly change major bleeding but reduced any bleeding, and edoxaban in Hokusai-VTE reduced the composite of major and clinically relevant nonmajor bleeding. A second generation of trials tested extended treatment against placebo or aspirin.

Comparator in the treatment trials was low-molecular-weight or unfractionated heparin followed by warfarin (international normalized ratio [INR] 2.0–3.0); EINSTEIN-PE and AMPLIFY used enoxaparin. In AMPLIFY-EXT, recurrent VTE or VTE-related death over 12 months occurred in 8.8% with placebo and in 1.7% with apixaban 2.5 mg or 5 mg twice daily, without an increase in major bleeding.

Cancer-associated venous thromboembolism

Low-molecular-weight heparin replaced warfarin for cancer-associated VTE after the CLOT trial, and DOACs have since been compared with dalteparin in randomized trials. In Hokusai VTE Cancer, edoxaban after 5 days of heparin was noninferior for the composite of recurrence and major bleeding, with more major bleeding, concentrated in gastrointestinal cancer. In SELECT-D, rivaroxaban reduced recurrence with more clinically relevant nonmajor bleeding. In Caravaggio, apixaban was noninferior to dalteparin for recurrent VTE (5.6% versus 7.9%) with similar major bleeding (3.8% versus 4.0%), including in patients with gastrointestinal cancer.

The 2026 AHA/ACC guideline recommends a DOAC or low-molecular-weight heparin over a vitamin K antagonist for cancer-associated pulmonary embolism (Class 1). Agent choice depends on tumor site (luminal gastrointestinal and genitourinary cancers carry more mucosal bleeding with edoxaban and rivaroxaban), nausea and oral intake, platelet count, kidney function, planned procedures, and interactions with anticancer drugs. Many kinase inhibitors, antiandrogens, and supportive drugs inhibit or induce CYP3A4 and P-glycoprotein. Apixaban is often selected in gastrointestinal cancer.

For extended treatment in active cancer, API-CAT randomized 1,766 patients who had completed at least 6 months of anticoagulation to apixaban 2.5 mg or 5 mg twice daily for 12 months. The reduced dose was noninferior for recurrent VTE (cumulative incidence over 12 months, 2.1% versus 2.8%) and caused less clinically relevant bleeding (12.1% versus 15.6%), with similar mortality (17.7% versus 19.6%). Detailed management is on the pulmonary embolism page.

Thromboprophylaxis after hip and knee replacement and in medical illness

Prophylactic doses are a fraction of treatment doses and start after surgical hemostasis. Apixaban 2.5 mg twice daily begins 12 to 24 hours after hip or knee replacement and continues for 35 days after hip and 12 days after knee replacement. Rivaroxaban 10 mg once daily begins 6 to 10 hours after surgery, once hemostasis is established, for the same durations. Dabigatran is labeled after hip replacement only: 110 mg 1 to 4 hours after surgery, then 220 mg once daily for 28 to 35 days. The label provides no dosing recommendation at a creatinine clearance of 30 mL/min or less or on dialysis. Edoxaban carries no US prophylaxis indication.

Rivaroxaban 10 mg once daily is also labeled for acutely ill medical patients at risk of VTE and not at high bleeding risk, in hospital and after discharge for a total of 31 to 39 days. Neuraxial anesthesia and epidural catheters require specific timing with any prophylactic anticoagulant. The labels carry a boxed warning about spinal and epidural hematoma, and the American Society of Regional Anesthesia and Pain Medicine intervals apply.

Low-dose rivaroxaban with aspirin in coronary and peripheral artery disease

Rivaroxaban 2.5 mg twice daily, the "vascular dose," combined with aspirin 75 to 100 mg daily, reduces atherothrombotic events by adding low-intensity factor Xa inhibition to platelet inhibition (dual-pathway inhibition). It is not an anticoagulant regimen for AF or VTE and does not prevent cardioembolic stroke.

In COMPASS (27,395 patients with stable coronary or peripheral artery disease), rivaroxaban 2.5 mg twice daily plus aspirin reduced cardiovascular death, stroke, or myocardial infarction compared with aspirin alone (4.1% versus 5.4%; hazard ratio 0.76). Major bleeding was more frequent (3.1% versus 1.9%), without a significant increase in fatal or intracranial bleeding. Rivaroxaban 5 mg twice daily without aspirin did not reduce events significantly. In VOYAGER PAD (6,564 patients within 10 days of lower extremity revascularization), the same regimen reduced the composite of acute limb ischemia, major amputation, myocardial infarction, ischemic stroke, or cardiovascular death at 3 years (17.3% versus 19.9%; hazard ratio 0.85).

The vascular dose is selected for patients with high ischemic risk and acceptable bleeding risk. Both trials excluded patients needing full-dose anticoagulation and patients at high bleeding risk; COMPASS also excluded recent stroke and any prior hemorrhagic or lacunar stroke. The 2020 ACC expert consensus decision pathway lists it as an option in chronic atherosclerotic disease without another anticoagulant indication. It is never combined with a full-dose anticoagulant, and the rivaroxaban label carries no kidney function adjustment for this dose.

Kidney function and the Cockcroft-Gault equation

DOAC labels and the pivotal trials used creatinine clearance estimated by the Cockcroft-Gault equation with actual body weight:

Creatinine clearance (mL/min) = (140 − age in years) × weight in kg ÷ (72 × serum creatinine in mg/dL), multiplied by 0.85 for women.

The estimated glomerular filtration rate (eGFR) reported by laboratories (Chronic Kidney Disease Epidemiology Collaboration [CKD-EPI] equation, indexed to 1.73 m² body surface area) is not interchangeable with Cockcroft-Gault clearance. In older, lighter patients, eGFR is often higher than Cockcroft-Gault clearance and can move a patient out of a dose-reduction band, which leads to overdosing. The 2023 AF guideline specifies Cockcroft-Gault for DOAC dosing, and the calculator below uses it.

Serum creatinine must be at steady state for any equation to be valid. In acute kidney injury, the estimate lags the true clearance, and a DOAC dose set during recovery or deterioration is revisited once creatinine stabilizes. Low muscle mass, amputation, and cirrhosis lower creatinine production and inflate any creatinine-based estimate. Measured clearance from a timed urine collection, or a cystatin C–based estimate, is used when muscle mass is atypical, although the DOAC labels are based on Cockcroft-Gault values.

DOAC dose calculator by indication

The calculator applies US prescribing information to the indication, Cockcroft-Gault creatinine clearance, age, weight, dialysis status, and one interacting drug. It reports the labeled dose for each of the four drugs, or the reason a drug is avoided.

Cockcroft-Gault creatinine clearance with actual body weight, × 0.85 for women. The calculator does not address liver disease, pregnancy, mechanical valves, rheumatic mitral stenosis, antiphospholipid syndrome, or combinations with antiplatelet therapy, all of which change the choice of agent.

Dose adjustment by creatinine clearance across indications

Labeled adjustments are listed for each band of creatinine clearance. Bands below 30 mL/min rest largely on pharmacokinetic modeling, because ARISTOTLE excluded clearance below 25 mL/min and the other AF trials excluded clearance below 30 mL/min.

Creatinine clearance in mL/min by Cockcroft-Gault. US prescribing information; 2023 ACC/AHA/ACCP/HRS AF guideline, Table 19. The edoxaban VTE dose is also reduced for weight 60 kg or less or listed P-glycoprotein inhibitors.

The 2023 AF guideline grades anticoagulant choice by stage of chronic kidney disease. In stage 3, DOACs at evidence-based doses are recommended over warfarin (Class 1). In stage 4 (creatinine clearance 15 to 30 mL/min), warfarin or a labeled-dose DOAC is reasonable (Class 2a). In kidney failure (below 15 mL/min) or on dialysis, warfarin with an INR of 2.0 to 3.0 or an evidence-based dose of apixaban might be reasonable (Class 2b). For VTE, the 2026 AHA/ACC pulmonary embolism guideline recommends a DOAC over a vitamin K antagonist in stages 2 and 3 (Class 1). It states that the benefit of apixaban over warfarin in stages 4 and 5 or on dialysis is uncertain (Class 2b).

Kidney disease raises bleeding risk independently of drug exposure, through uremic platelet dysfunction, anemia, and vascular fragility. In ARISTOTLE, the relative reduction in major bleeding with apixaban compared with warfarin was greatest at a Cockcroft-Gault clearance of 50 mL/min or less (hazard ratio 0.50), which is one reason apixaban is favored in advanced chronic kidney disease. The 2023 Beers Criteria advise avoiding dabigatran at creatinine clearance below 30 mL/min and edoxaban below 15 or above 95 mL/min. They removed apixaban from their kidney function table because of evidence of safe use at lower kidney function.

DOACs in end-stage kidney disease and dialysis

Patients receiving dialysis have high rates of both stroke and bleeding, and AF in this population carries a stroke risk that is not well described by CHA₂DS₂-VASc. None of the pivotal trials enrolled patients on dialysis, and no randomized trial has shown that any anticoagulant improves net outcomes compared with none.

Apixaban. US labeling provides an AF dose on dialysis (5 mg twice daily, or 2.5 mg if age 80 or older or weight 60 kg or less) based on a pharmacokinetic study in which 5 mg twice daily produced concentrations similar to those in ARISTOTLE. Hemodialysis removes little apixaban: dialysis clearance is about 18 mL/min, and exposure is 14% lower on dialysis days. In the RENAL-AF pharmacokinetic substudy, the 12-hour exposure with 5 mg twice daily was about twice that with 2.5 mg twice daily.

Randomized trials. RENAL-AF randomized 154 patients on hemodialysis to apixaban or warfarin before stopping for slow enrollment. At 1 year, major or clinically relevant nonmajor bleeding occurred in 32% versus 26% (hazard ratio 1.20, 95% CI 0.63 to 2.30) and stroke or systemic embolism in 3.0% versus 3.3%; time in therapeutic range on warfarin was 44%. The trial was underpowered for any conclusion, and death was the most frequent major event (26% with apixaban and 18% with warfarin). A European trial comparing apixaban 2.5 mg twice daily with phenprocoumon was similarly small and inconclusive.

Other DOACs. Rivaroxaban 15 mg once daily is labeled for AF on dialysis on the basis of exposure data, with no outcome trial. Dabigatran is dialyzable, accumulates between sessions, and has no labeled dose on dialysis. Edoxaban is not recommended below 15 mL/min. For VTE treatment, rivaroxaban is avoided below 15 mL/min, and dabigatran and edoxaban are not used. Apixaban carries no labeled adjustment, but patients on dialysis were not enrolled in the efficacy trials.

The decision to anticoagulate a patient on dialysis is individualized and revisited, because transplantation, access complications, falls, and gastrointestinal angiodysplasia change the balance. Percutaneous left atrial appendage occlusion is an alternative when bleeding risk is prohibitive, although patients on dialysis were excluded from the pivotal device trials.

DOACs in hepatic impairment

Liver disease alters DOAC exposure through reduced metabolism (apixaban, rivaroxaban), reduced protein binding, and coexisting kidney dysfunction, and it independently raises bleeding through thrombocytopenia, low clotting factors, and portal hypertension. The baseline INR in cirrhosis reflects reduced synthesis of procoagulant factors and does not measure bleeding risk, so it cannot guide DOAC use. Patients with active liver disease or transaminases more than twice the upper limit were excluded from the pivotal trials.

US prescribing information. The 2023 AF guideline rates any DOAC as reasonable in Child-Pugh A; apixaban, dabigatran, or edoxaban as reasonable in Child-Pugh B; and rivaroxaban as harmful in Child-Pugh B (Class 3: Harm). DOACs are avoided in Child-Pugh C. The 2026 AHA/ACC pulmonary embolism guideline rates a DOAC as reasonable in Child-Pugh A (Class 2a), possibly reasonable in Child-Pugh B (Class 2b), and harmful in Child-Pugh C (Class 3: Harm).

In mild or moderate liver disease without significant coagulopathy, anticoagulation for AF is reasonable (2023 AF guideline, Class 2a). Observational studies in cirrhosis associate DOACs with similar or lower bleeding than warfarin, but they enroll mainly Child-Pugh A and B patients. Warfarin and low-molecular-weight heparin remain the agents with the longest experience in portal vein thrombosis and decompensated cirrhosis. Drug-induced liver injury with the approved DOACs is rare, in contrast to ximelagatran, an earlier oral thrombin inhibitor withdrawn for hepatotoxicity.

DOACs at extremes of body weight and after bariatric surgery

Obesity. Body weight changes DOAC exposure less than it changes the exposure to weight-based heparins. Few trial participants weighed more than 120 kg or had a body mass index (BMI) of 40 kg/m² or more, so dosing above these thresholds relies on pharmacokinetic and observational data. The 2021 International Society on Thrombosis and Haemostasis (ISTH) guidance on VTE treatment in obesity supports standard-dose rivaroxaban or apixaban regardless of BMI or weight. It advises against dabigatran and edoxaban when BMI exceeds 40 kg/m² or weight exceeds 120 kg, because data are insufficient. It also advises against routine measurement of peak or trough levels, because levels have no established relation to outcomes. Data for BMI of 50 kg/m² or more or weight above 150 kg remain sparse.

For AF, the 2023 guideline rates a DOAC as reasonable over warfarin in class III obesity, BMI of 40 kg/m² or more (Class 2a). For VTE, the 2026 AHA/ACC pulmonary embolism guideline rates a DOAC as reasonable over a vitamin K antagonist when BMI exceeds 30 kg/m², including severe obesity (Class 2a), with apixaban and rivaroxaban having the most data. Cockcroft-Gault clearance with actual body weight overestimates kidney function in obesity, which matters for dabigatran and edoxaban.

Bariatric surgery. Gastric bypass and sleeve gastrectomy shorten the absorptive surface, alter gastric pH, and accelerate transit. The ISTH guidance suggests parenteral anticoagulation in the early postoperative period after bariatric surgery, with a switch to a vitamin K antagonist or a DOAC only after at least 4 weeks. It suggests a trough level to confirm absorption when a DOAC is used. Absorption is least affected after gastric banding and most uncertain after Roux-en-Y gastric bypass and biliopancreatic diversion. For AF, the 2023 guideline rates warfarin as reasonable after bariatric surgery because of uncertain DOAC absorption (Class 2b).

Low body weight. Weight of 60 kg or less is a dose-reduction criterion for apixaban in AF (with age or creatinine) and for edoxaban in VTE. Patients below 50 kg were few in the trials and have higher DOAC concentrations. Labeled doses are used, and the decision is individualized in frail, very light patients, in whom both stroke and bleeding risks are high.

DOACs in older adults and frailty

The absolute benefit of anticoagulation for AF rises with age, because stroke risk rises faster than bleeding risk. In each pivotal AF trial, the relative effects of the DOAC versus warfarin were consistent in patients 75 years or older. Older adults also have the lowest creatinine clearance, the most interacting drugs, and the highest rates of off-label dosing.

Two trials address the frailest patients. ELDERCARE-AF randomized 984 Japanese patients 80 years or older with AF who were not candidates for standard-dose anticoagulation to edoxaban 15 mg daily or placebo. Stroke or systemic embolism fell from 6.7% to 2.3% per year (hazard ratio 0.34), and major bleeding was not significantly increased. The 15-mg dose is not approved in the United States. FRAIL-AF randomized 1,330 frail patients 75 years or older (mean age 83) who were stable on a vitamin K antagonist to continue it or switch to a DOAC. Switching increased major or clinically relevant nonmajor bleeding (hazard ratio 1.69, 95% CI 1.23 to 2.32) without reducing thromboembolism, and the trial stopped for futility. The result applies to frail patients already stable on warfarin; it does not apply to the choice of a first anticoagulant.

The 2023 Beers Criteria advise against starting warfarin for AF or VTE unless DOACs are contraindicated or impractical. They advise avoiding rivaroxaban for long-term treatment of AF or VTE in older adults in favor of safer alternatives, and caution with dabigatran because of higher gastrointestinal bleeding than apixaban. Falls alone are not a reason to withhold anticoagulation; the number of falls needed to offset the benefit is very high. Cognitive impairment without supervision of dosing, a medication burden that cannot be managed, and life expectancy shorter than the time to benefit are reasons for reassessment.

Drug interactions through P-glycoprotein and CYP3A4

All four DOACs are substrates of the efflux transporter P-glycoprotein, which limits intestinal absorption and promotes renal and biliary excretion. Apixaban and rivaroxaban are also metabolized by CYP3A4; edoxaban only minimally, and dabigatran not at all. Drugs that inhibit both pathways raise DOAC concentrations and bleeding, and drugs that induce both lower concentrations and the protection against thrombosis. Kidney dysfunction amplifies inhibitor interactions, most strongly for dabigatran and edoxaban.

US prescribing information. CrCl, Cockcroft-Gault creatinine clearance in mL/min; NSAIDs, nonsteroidal anti-inflammatory drugs; SSRIs, selective serotonin reuptake inhibitors; SNRIs, serotonin-norepinephrine reuptake inhibitors. The 2023 AF guideline recommends active management of these interactions (Class 1).

Antiepileptic drugs. Carbamazepine, phenytoin, phenobarbital, and primidone induce both CYP3A4 and P-glycoprotein and are avoided with all DOACs. Levetiracetam, lamotrigine, and lacosamide have no relevant pharmacokinetic interaction.

Antimicrobials and antivirals. Rifampin lowers rivaroxaban and apixaban exposure by about half and dabigatran exposure by about two-thirds, and a parenteral anticoagulant or warfarin is used during rifampin therapy. Azole antifungals vary: ketoconazole, itraconazole, voriconazole, and posaconazole are strong CYP3A4 inhibitors, and fluconazole is a moderate one. HIV protease inhibitors boosted with ritonavir or cobicistat are strong dual inhibitors. Nirmatrelvir-ritonavir, given for 5 days, is a strong dual inhibitor and is managed by the same rules as other ritonavir-containing regimens.

Cardiovascular drugs. Amiodarone, verapamil, diltiazem, dronedarone, and quinidine inhibit P-glycoprotein. Their effects on factor Xa inhibitors are modest, and US labels require no AF dose change except for dabigatran with dronedarone. Ticagrelor modestly raises dabigatran exposure without a labeled adjustment. Statins and digoxin do not require DOAC dose changes. Cyclosporine and tacrolimus inhibit P-glycoprotein and raise DOAC exposure, most for dabigatran; transplant recipients are managed with specialist pharmacy input.

Interactions across the wider cardiovascular formulary, including warfarin, statins, and antiarrhythmic drugs, are covered under cardiovascular drug interactions and dosing.

Measuring DOAC effect: drug-specific assays and routine coagulation tests

Routine monitoring of DOAC levels is not recommended, because no level has been validated as a dosing target and the trials used fixed doses. Measurement matters in a defined set of situations: life-threatening bleeding, urgent surgery or an invasive procedure, suspected overdose or accumulation in acute kidney injury, thrombolysis decisions in acute ischemic stroke, suspected nonadherence, thrombosis on treatment, and, in specialized centers, extremes of body weight or multiple interacting drugs.

Adapted from the 2020 ACC expert consensus decision pathway on bleeding (Tables 2 and 3) and the 2021 EHRA practical guide. A heparin-calibrated anti-Xa activity below the lower limit of quantitation makes a clinically relevant factor Xa inhibitor level unlikely; a result above the range indicates that drug is present but cannot be converted to a concentration.

Expected concentrations. The EHRA guide lists the ranges of plasma levels measured in the AF trials: dabigatran peak 52 to 383 and trough 28 to 215 ng/mL; apixaban peak 69 to 321 and trough 34 to 230 ng/mL; edoxaban peak 101 to 288 and trough 12 to 43 ng/mL; rivaroxaban peak 178 to 343 and trough 12 to 137 ng/mL. The ranges are wide and overlap between doses, which is why a single level cannot identify the correct dose. Peak samples are drawn 2 to 4 hours after a dose and trough samples just before the next dose, and the timing of the last dose is recorded with every result.

Decision thresholds. Consensus thresholds from the ISTH (a level above 50 ng/mL in serious bleeding and above 30 ng/mL before an urgent high-bleeding-risk procedure) are used to select patients for reversal; they are not trial-derived targets. Interpretation of tests in bleeding and before procedures, including viscoelastic testing, is covered under anticoagulant bleeding, reversal, and periprocedural management.

Interference with other tests. DOACs prolong clot-based assays used in thrombophilia testing. Factor Xa inhibitors and dabigatran can both cause false-positive lupus anticoagulant results, and each falsely raises antithrombin activity in assays based on the enzyme it inhibits. Dabigatran can also lower fibrinogen results with some Clauss reagents. Testing is timed at trough or after a short interruption, or DOAC-adsorbent agents are used in the laboratory. During overlap with warfarin, the INR is drawn just before the next DOAC dose, when interference is lowest.

Contraindications: mechanical valves, rheumatic mitral stenosis, and antiphospholipid syndrome

Mechanical heart valves. DOACs are contraindicated with mechanical prostheses. In RE-ALIGN, dabigatran in patients with mechanical aortic or mitral valves caused ischemic or unspecified stroke in 5% versus none with warfarin and major bleeding in 4% versus 2%, all of it pericardial. The trial stopped after 252 patients. In PROACT Xa, apixaban 5 mg twice daily in patients with an On-X aortic valve, the prosthesis with the lowest thrombogenicity, caused valve thrombosis or valve-related thromboembolism at 4.2% versus 1.3% per patient-year. The trial stopped after 863 patients. Thrombosis on a mechanical valve is driven by contact activation on the prosthetic surface, which generates thrombin locally and overwhelms fixed-dose direct inhibition. Valve-specific warfarin management is covered on the warfarin topic page.

Rheumatic mitral stenosis. INVICTUS randomized 4,565 patients with rheumatic heart disease and AF (mean age 50.5 years; 72% women) to rivaroxaban or a dose-adjusted vitamin K antagonist. The vitamin K antagonist led to fewer events of the composite of stroke, systemic embolism, myocardial infarction, or vascular or unknown death (446 versus 560 events; restricted mean survival time 1,675 versus 1,599 days). The difference arose mainly through fewer deaths, without more major bleeding. The 2023 AF guideline restricts DOAC use to patients without moderate to severe rheumatic mitral stenosis; warfarin is used in those with it.

Antiphospholipid syndrome. In TRAPS, which enrolled 120 triple-positive patients (lupus anticoagulant, anticardiolipin, and anti-β2-glycoprotein I antibodies), rivaroxaban caused thromboembolic events in 12% (ischemic stroke and myocardial infarction) versus none with warfarin, and the trial stopped early. Other trials with rivaroxaban and apixaban showed excess arterial events. All four US labels state that DOACs are not recommended in triple-positive antiphospholipid syndrome. The ISTH guidance advises against DOACs in patients with arterial thrombosis or triple positivity and favors a vitamin K antagonist. The 2026 AHA/ACC guideline rates a DOAC as a possible alternative after VTE when only a single anticardiolipin or anti-β2-glycoprotein I antibody is present (Class 2b). Testing for antiphospholipid antibodies is affected by DOAC interference, as described above.

Other contraindications. Active pathologic bleeding and hypersensitivity to the drug are labeled contraindications for all four. The edoxaban label also lists moderate to severe mitral stenosis as a setting where safety and efficacy have not been studied.

Other cardiac settings: bioprosthetic valves, TAVR, left ventricular thrombus, cardioversion, and device-detected AF

Bioprosthetic valves and other native valve disease. Valve disease other than a mechanical valve or moderate to severe rheumatic mitral stenosis does not exclude DOACs. In patients with AF and aortic stenosis, mitral regurgitation, prior mitral repair, or a bioprosthetic valve, the 2023 AF guideline recommends a DOAC over a vitamin K antagonist (Class 1). In RIVER, rivaroxaban was noninferior to warfarin in 1,005 patients with AF and a bioprosthetic mitral valve. The dabigatran label states that use with bioprosthetic valves was not studied and is not recommended.

Transcatheter aortic valve replacement (TAVR). A DOAC is used after TAVR when AF or another indication requires anticoagulation, generally without an added antiplatelet agent. In ENVISAGE-TAVI AF, edoxaban was noninferior to a vitamin K antagonist for net adverse events but caused more major bleeding (9.7 versus 7.0 per 100 person-years), mostly gastrointestinal. Without an independent indication, routine anticoagulation after TAVR is harmful. In GALILEO, low-dose rivaroxaban with aspirin increased death, thromboembolism, and bleeding compared with antiplatelet therapy. The rivaroxaban label states that use after TAVR is not recommended.

Left ventricular thrombus. The 2022 AHA scientific statement describes DOACs as a reasonable alternative to warfarin for left ventricular thrombus after myocardial infarction and in cardiomyopathy, based on observational studies and small randomized trials that showed similar thrombus resolution and less bleeding. Treatment usually lasts about 3 months with repeat imaging. When antiplatelet therapy after an acute coronary syndrome is also required, the combination rules for AF after percutaneous coronary intervention (PCI) apply.

Cardioversion. For AF of 48 hours or longer, 3 weeks of uninterrupted therapeutic anticoagulation (or imaging to exclude left atrial thrombus) precedes elective cardioversion, and anticoagulation continues for at least 4 weeks after. DOACs were tested for cardioversion in X-VeRT, ENSURE-AF, and EMANATE with low event rates similar to warfarin. Unlike warfarin, a DOAC leaves no INR record of the preceding 3 weeks, so adherence is confirmed by the patient's account of every dose. Details are on the atrial fibrillation page.

Device-detected AF. ARTESIA (apixaban versus aspirin) and NOAH-AFNET 6 (edoxaban versus placebo) tested anticoagulation in subclinical AF lasting 6 minutes to 24 hours; apixaban reduced stroke at the cost of more major bleeding, and edoxaban did not significantly reduce the primary outcome. The interpretation is on the atrial fibrillation page.

Switching between warfarin, heparin, and DOACs

Transitions are timed so that anticoagulation never lapses and never stacks. The onset of a DOAC within hours, and its offset within a day or two, allow most transitions to occur at the time the next dose is due. Warfarin is the exception in both directions, because its onset is slow and the INR is unreliable while a DOAC is present.

From current US prescribing information. DOAC-to-DOAC switching follows the labels and the 2021 EHRA practical guide. CrCl, Cockcroft-Gault creatinine clearance in mL/min. The apixaban and rivaroxaban labels describe starting a parenteral anticoagulant with warfarin when the next DOAC dose is due. The dabigatran label overlaps dabigatran with warfarin. The edoxaban label offers either a half-dose overlap or a parenteral option, continued until the INR is stable at 2.0 or more. The ASH 2018 guideline on optimal management of anticoagulation therapy suggests overlapping the DOAC with warfarin until the INR is therapeutic, rather than bridging with heparin (conditional recommendation).

The transition from a DOAC to warfarin carries the highest risk of an unprotected interval. DOACs raise the INR, so an INR drawn shortly after a DOAC dose overstates the warfarin effect. INR values are drawn at trough, just before the next DOAC dose, and the dabigatran label states that the INR reflects warfarin only after dabigatran has been stopped for at least 2 days. A new mechanical valve, newly diagnosed moderate or severe rheumatic mitral stenosis, or newly confirmed triple-positive antiphospholipid syndrome requires this switch, with continuous parenteral coverage until the INR is therapeutic.

Switching from warfarin to a DOAC is indicated for poor INR control (time in therapeutic range below 60% to 65%), patient preference, or drug and dietary interactions. FRAIL-AF showed that in frail patients 75 years or older who are stable on warfarin, routine switching increases bleeding. INR management on warfarin is covered on the warfarin topic page.

Missed doses, double doses, adherence, and persistence

The short half-lives that allow quick interruption also leave little residual protection after a missed dose. Within one to two missed days, a patient taking a DOAC has little anticoagulant effect, whereas a missed warfarin dose barely changes the INR. Adherence is therefore part of the decision to prescribe a DOAC, and a once-daily drug is chosen for some patients because it simplifies the schedule.

US prescribing information. The EHRA practical guide gives a general rule: a forgotten dose may be taken until half the dosing interval has passed (6 h for twice-daily and 12 h for once-daily drugs), after which it is skipped.

Double doses. When a dose is taken twice, the EHRA guide advises skipping the next planned dose of a twice-daily drug and restarting 24 hours after the double dose, and continuing the usual schedule for a once-daily drug. Larger overdoses, especially with impaired kidney function, are assessed with a drug-calibrated level and the time of ingestion. Activated charcoal reduces absorption when given within 2 to 4 hours of ingestion, and hemodialysis removes dabigatran.

Adherence and persistence. Adherence (taking prescribed doses) and persistence (remaining on therapy) are higher with DOACs than with warfarin in clinical practice but remain incomplete, with substantial discontinuation within the first year in AF cohorts. Nonadherence is associated with higher stroke risk. Contributors include cost, dyspepsia with dabigatran, minor bleeding, misunderstanding of the absence of monitoring, and polypharmacy. These factors are addressed by education at initiation, a written plan for missed doses and procedures, pharmacy refill data, and blister packs. Kidney function, hemoglobin, and interacting drugs are reviewed at each visit. Thrombosis on a DOAC is evaluated first for nonadherence, underdosing, interacting inducers, and absorption problems before it is labeled treatment failure.

Cost and coverage. Out-of-pocket cost is a recognized cause of nonadherence and of switching to warfarin. Apixaban and rivaroxaban were among the first ten drugs selected for Medicare drug price negotiation under the Inflation Reduction Act (CMS). The FDA Orange Book lists approved generic versions, where they exist, for each DOAC. Formulary coverage often determines which factor Xa inhibitor a patient takes, and an interruption in coverage is a predictable point of nonadherence.

Dosing errors and off-label dose reduction

Off-label dosing is common and harmful. In a US claims analysis of 14,865 patients starting a DOAC for AF, 43% of patients with a renal indication for dose reduction received the full dose. In apixaban-treated patients this overdosing was associated with more major bleeding (hazard ratio 2.19), while stroke did not differ in the pooled analysis of the three drugs. Of patients without a renal indication, 13.3% received a reduced dose, and in apixaban-treated patients this underdosing was associated with almost five times the stroke risk (hazard ratio 4.87) without less bleeding.

Common dosing errors include:

• Apixaban 2.5 mg twice daily prescribed for a single criterion, for age alone, or because of a minor bleed.
• eGFR from the laboratory report, or a creatinine drawn during acute kidney injury, used in place of current Cockcroft-Gault clearance, which moves older, light patients out of a reduction band.
• The AF apixaban reduction rule applied to acute VTE treatment, or the 7-day 10-mg apixaban or 21-day 15-mg rivaroxaban loading phase omitted or extended.
• Rivaroxaban 15 mg or 20 mg taken without food, which lowers absorption.
• Edoxaban prescribed for AF at a creatinine clearance above 95 mL/min.
• Dabigatran capsules opened for a feeding tube or stored outside the original packaging.
• Duplicate anticoagulation at transitions (a DOAC started while therapeutic heparin continues, or taken after discharge together with enoxaparin).
• A kidney function decline during intercurrent illness not followed by a dose change.
• Rivaroxaban 2.5 mg twice daily mistaken for an anticoagulant dose in a patient with AF.

The 2023 AF guideline rates doses not supported by trial evidence as harmful (Class 3: Harm), and dose appropriateness is reviewed whenever age, weight, kidney function, or interacting drugs change.

Bleeding on DOACs: rates, gastrointestinal bleeding, and intracranial hemorrhage

Major bleeding with the standard DOAC regimens in the AF trials occurred at about 2.1% to 3.6% per year, depending on drug, dose, and population, against 3.1% to 3.4% per year with warfarin. The pattern differs by site. Intracranial hemorrhage, the bleeding complication with the highest mortality, fell with every DOAC: by about half across the four trials combined and by about one-third with rivaroxaban in ROCKET AF. Gastrointestinal bleeding rose with dabigatran 150 mg, rivaroxaban, and high-dose edoxaban and was not increased with apixaban.

Two mechanisms explain the gastrointestinal signal. Unabsorbed active drug in the gut lumen exerts a local anticoagulant effect on mucosal lesions, more so for drugs with low or incomplete bioavailability. Once-daily dosing produces higher peak concentrations. Risk factors include age, prior gastrointestinal bleeding, concomitant antiplatelet drugs and NSAIDs, kidney dysfunction, and Helicobacter pylori. Proton pump inhibitors reduce upper gastrointestinal bleeding in patients taking combined antithrombotic therapy and are commonly given to patients at high gastrointestinal risk.

Bleeding risk scores, including HAS-BLED, identify modifiable factors and set the frequency of review; the 2023 AF guideline states they are not used alone to withhold anticoagulation in patients at high stroke risk. Scores are described on the atrial fibrillation page. Management of active bleeding, transfusion, and the timing of resumption after gastrointestinal or intracranial hemorrhage are covered under anticoagulant bleeding, reversal, and periprocedural management.

Reversal agents and the status of andexanet alfa

Most bleeding on a DOAC is managed by stopping the drug, local hemostasis, and supportive care, because anticoagulant effect falls by half within about 12 hours with normal kidney function. Reversal is reserved for life-threatening bleeding, bleeding at a critical site, and urgent surgery that cannot be delayed.

Dabigatran. Idarucizumab, a humanized antibody fragment, binds dabigatran with an affinity about 350 times that of dabigatran for thrombin. The dose is 5 g intravenously, given as two consecutive 2.5-g infusions or bolus injections, and reversal is complete within minutes. The 2023 AF guideline recommends idarucizumab for life-threatening bleeding on dabigatran (Class 1), with activated prothrombin complex concentrate (aPCC) reasonable when it is unavailable (Class 2a). Hemodialysis removes dabigatran and is used when levels are very high or kidney function is poor.

Factor Xa inhibitors. Andexanet alfa, a recombinant inactive factor Xa decoy, received accelerated US approval in 2018 for life-threatening bleeding on apixaban or rivaroxaban. In ANNEXA-I, which randomized 530 patients with factor Xa inhibitor–associated intracerebral hemorrhage, andexanet improved hemostatic efficacy compared with usual care (67.0% versus 53.1%). Usual care was mostly four-factor prothrombin complex concentrate (PCC). Andexanet increased thrombotic events (10.3% versus 5.6%) and ischemic stroke (6.5% versus 1.5%), without a difference in functional outcome or 30-day death. The US Food and Drug Administration concluded that the risks of andexanet outweighed its benefits, and the manufacturer withdrew it from the US market; it remains available in some other countries. Guideline statements that preferred andexanet predate ANNEXA-I and the withdrawal. In the United States, four-factor PCC is the agent used for life-threatening bleeding on apixaban, rivaroxaban, and edoxaban. The dose is a fixed 2,000 units or 25 to 50 units/kg, with activated PCC 50 units/kg as an institutional alternative.

Every reversal exposes the patient to the thrombotic risk that prompted anticoagulation, and the anticoagulant is resumed once hemostasis is secure and the indication persists. Agent doses, the evidence for PCC, and resumption timing are covered in the bleeding management topic.

Interruption of DOACs for procedures

Predictable half-lives allow DOAC interruption to be timed by pharmacokinetics and procedural bleeding risk, without heparin bridging and without routine testing. In PAUSE (3,007 patients with AF taking apixaban, dabigatran, or rivaroxaban), the DOAC was omitted for 1 day before a low-to-moderate bleeding-risk procedure and 2 days before a high-bleeding-risk procedure, with longer omission of dabigatran when creatinine clearance was below 50 mL/min. Major bleeding at 30 days was 0.90% to 1.85%, and arterial thromboembolism was 0.16% to 0.60%.

PAUSE schedule as adopted by the 2022 CHEST perioperative guideline and the 2023 AF guideline (Table 18). No dose is taken on the day of the procedure. Resumption is about 24 hours after low-to-moderate and 48 to 72 hours after high-bleeding-risk procedures, once hemostasis is secure. Neuraxial anesthesia follows longer intervals.

Many minimal-risk procedures, including dental extraction, cataract surgery, minor skin procedures, and, in BRUISE CONTROL-2, cardiac device implantation, can proceed with continued or briefly interrupted DOAC therapy. Catheter ablation of AF is commonly performed on uninterrupted or minimally interrupted DOACs. Detailed schedules for lower clearance, neuraxial anesthesia, and specific procedures are covered under anticoagulant bleeding, reversal, and periprocedural management.

DOACs combined with antiplatelet therapy and alternatives to anticoagulation

AF with PCI or acute coronary syndrome. In patients taking an anticoagulant who undergo PCI, the 2023 AF guideline recommends a DOAC over a vitamin K antagonist in combination with antiplatelet therapy (Class 1). Dual therapy (DOAC plus a P2Y12 inhibitor, usually clopidogrel) follows 1 to 4 weeks of triple therapy. The DOAC is given at its full stroke-prevention dose; the trial exception is rivaroxaban 15 mg daily (10 mg at creatinine clearance 30 to 50 mL/min) from PIONEER AF-PCI while a P2Y12 inhibitor is combined. After 12 months in most patients, anticoagulant monotherapy follows. The 2020 ACC expert consensus decision pathway applies the same sequence to patients anticoagulated for VTE, at VTE treatment doses. The trials, regimens, and timelines are summarized on the atrial fibrillation page and in the separate topic on antithrombotic therapy in AF with PCI.

Chronic coronary disease. In patients with AF and stable coronary disease more than 12 months after PCI, anticoagulant monotherapy is preferred over adding aspirin; in EPIC-CAD, edoxaban alone reduced net adverse events compared with edoxaban plus a single antiplatelet agent. Aspirin without a current indication is an avoidable source of bleeding on a DOAC.

Left atrial appendage occlusion. In patients with AF and a contraindication to long-term anticoagulation from a nonreversible cause, percutaneous appendage occlusion is reasonable when CHA₂DS₂-VASc is 2 or more (2023 AF guideline, Class 2a), and it may be reasonable in patients at high bleeding risk who prefer it (Class 2b). The device requires short-term antithrombotic therapy after implantation. Selection and outcomes are on the atrial fibrillation page.

DOACs in pregnancy, lactation, and contraception

DOACs cross the placenta, human safety data are insufficient, and pregnancy changes their absorption, distribution, and clearance. They are not used in pregnancy. The 2026 AHA/ACC pulmonary embolism guideline recommends low-molecular-weight or unfractionated heparin in pregnancy and classifies DOACs and warfarin as potentially harmful (Class 3: Harm). Women of childbearing potential who take a DOAC are counseled about contraception and about switching to heparin as soon as pregnancy is planned or recognized. Inadvertent exposure in early pregnancy is managed by stopping the DOAC and starting heparin.

The apixaban, dabigatran, and edoxaban labels state that breastfeeding is not recommended during treatment, because human milk data are lacking and the drugs appear in rat milk. Rivaroxaban has been detected in human milk, and its label directs that the benefits of breastfeeding be weighed against the mother's need for the drug and the possible effects on the infant. The 2026 AHA/ACC guideline recommends low-molecular-weight heparin, unfractionated heparin, or warfarin over DOACs during breastfeeding (Class 1). Heavy menstrual bleeding is a recognized adverse effect of oral anticoagulants. In US Sentinel data, severe abnormal uterine bleeding requiring surgery was more frequent with rivaroxaban than with apixaban, dabigatran, or warfarin. Heavy menstrual bleeding on a DOAC is addressed with gynecologic input, hormonal or tranexamic acid therapy, or a change of agent.

DOACs in heparin-induced thrombocytopenia

DOACs do not cross-react with heparin-induced thrombocytopenia (HIT) antibodies and are increasingly used after the acute phase of HIT. The ASH 2018 HIT guideline lists argatroban, bivalirudin, danaparoid, fondaparinux, or a DOAC as options for acute HIT (conditional recommendation, very low certainty of evidence), with the choice determined by thrombotic burden, bleeding risk, kidney and liver function, and the need for procedures. DOACs are used mainly in clinically stable patients without life-threatening thrombosis, and rivaroxaban at VTE treatment doses has the most reported experience. Warfarin is avoided in acute HIT until the platelet count recovers, because of the risk of venous limb gangrene. Diagnosis and parenteral therapy are covered under parenteral anticoagulants and HIT.

Factor XI and XIa inhibitors in development

Factor XI sits in the amplification loop of thrombin generation. People with congenital factor XI deficiency have fewer ischemic strokes and venous thromboses and rarely bleed spontaneously, which suggests that inhibiting factor XI could prevent thrombosis with less bleeding than inhibiting factor Xa or thrombin. Three classes are in development: small-molecule oral inhibitors of factor XIa (asundexian, milvexian), monoclonal antibodies given monthly by injection (abelacimab, osocimab), and antisense oligonucleotides that reduce factor XI synthesis.

Results of peer-reviewed phase 2 and phase 3 trials. TIA, transient ischemic attack.

The completed trials separate two questions. Factor XI inhibition reduces bleeding consistently, but OCEANIC-AF showed that the asundexian dose tested did not give stroke protection comparable to apixaban in AF, possibly because factor XIa inhibition at that dose was incomplete. OCEANIC-STROKE showed benefit on top of antiplatelet therapy for noncardioembolic stroke, which involves a different mechanism and comparator. Whether more complete factor XI inhibition can match DOAC efficacy in AF is being tested in LIBREXIA AF (milvexian versus apixaban) and LILAC-TIMI 76 (abelacimab versus placebo in patients with AF unsuitable for anticoagulation). No factor XI inhibitor is approved for AF or VTE, and none has a specific reversal agent.

Differences between US and European DOAC guidance

European guidance differs in a few practical respects. Dabigatran 110 mg twice daily is approved in Europe and recommended for patients 80 years or older or at high bleeding risk, a dose not available for AF in the United States. The European edoxaban label carries no upper creatinine clearance limit, and in Europe the 30-mg AF dose is also used for weight 60 kg or less and for selected P-glycoprotein inhibitors. The EHRA practical guide and European labeling advise against combining factor Xa inhibitors with strong dual CYP3A4 and P-glycoprotein inhibitors, whereas US labeling allows apixaban at half dose. Rivaroxaban 2.5 mg twice daily after acute coronary syndrome, based on ATLAS ACS 2-TIMI 51, is approved in Europe but not in the United States. Andexanet alfa remains available in Europe.

Sources of error in DOAC prescribing

• Treating the DOACs as interchangeable: dose rules, renal limits, food requirements, and interactions differ for each drug and each indication.
• Prescribing any DOAC with a mechanical valve, moderate to severe rheumatic mitral stenosis, or triple-positive antiphospholipid syndrome.
• Combining a DOAC with rifampin, carbamazepine, phenytoin, or St. John's wort, or with ketoconazole, itraconazole, or ritonavir without the labeled adjustment.
• Interpreting a normal PT or aPTT as proof that a factor Xa inhibitor has cleared, or a prolonged INR as a measure of DOAC intensity.
• Drawing the INR at peak DOAC concentration during transition to warfarin, which overstates the warfarin effect.
• Continuing aspirin without a current indication alongside a DOAC.
• Bridging with heparin for routine DOAC interruption, which increases bleeding without reducing thromboembolism.
• Using a DOAC in pregnancy or during breastfeeding.
• Expecting andexanet alfa to be available for factor Xa inhibitor reversal in the United States.

References

1. Joglar JA, Chung MK, Armbruster AL, et al. 2023 ACC/AHA/ACCP/HRS guideline for the diagnosis and management of atrial fibrillation: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2024;149:e1–e156. Corrections: Circulation. 2024;149:e167, e936, and e1413. doi:10.1161/CIR.0000000000001193
2. Creager MA, Barnes GD, Giri J, et al. 2026 AHA/ACC/ACCP/ACEP/CHEST/SCAI/SHM/SIR/SVM/SVN guideline for the evaluation and management of acute pulmonary embolism in adults: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2026;153:e977–e1051 (correction: Circulation. 2026;154:e24). doi:10.1161/CIR.0000000000001415
3. Stevens SM, Woller SC, Baumann Kreuziger L, et al. Antithrombotic therapy for VTE disease: second update of the CHEST guideline and expert panel report. Chest. 2021;160:e545–e608. doi:10.1016/j.chest.2021.07.055
4. Ortel TL, Neumann I, Ageno W, et al. American Society of Hematology 2020 guidelines for management of venous thromboembolism: treatment of deep vein thrombosis and pulmonary embolism. Blood Adv. 2020;4:4693–4738. doi:10.1182/bloodadvances.2020001830
5. Kumbhani DJ, Cannon CP, Beavers CJ, et al. 2020 ACC expert consensus decision pathway for anticoagulant and antiplatelet therapy in patients with atrial fibrillation or venous thromboembolism undergoing percutaneous coronary intervention or with atherosclerotic cardiovascular disease: a report of the American College of Cardiology Solution Set Oversight Committee. J Am Coll Cardiol. 2021;77:629–658. doi:10.1016/j.jacc.2020.09.011
6. Tomaselli GF, Mahaffey KW, Cuker A, et al. 2020 ACC expert consensus decision pathway on management of bleeding in patients on oral anticoagulants: a report of the American College of Cardiology Solution Set Oversight Committee. J Am Coll Cardiol. 2020;76:594–622. doi:10.1016/j.jacc.2020.04.053
7. Douketis JD, Spyropoulos AC, Murad MH, et al. Perioperative management of antithrombotic therapy: an American College of Chest Physicians clinical practice guideline. Chest. 2022;162:e207–e243. doi:10.1016/j.chest.2022.07.025
8. Steffel J, Collins R, Antz M, et al. 2021 European Heart Rhythm Association practical guide on the use of non-vitamin K antagonist oral anticoagulants in patients with atrial fibrillation. Europace. 2021;23:1612–1676. doi:10.1093/europace/euab065
9. Martin KA, Beyer-Westendorf J, Davidson BL, et al. Use of direct oral anticoagulants in patients with obesity for treatment and prevention of venous thromboembolism: updated communication from the ISTH SSC Subcommittee on Control of Anticoagulation. J Thromb Haemost. 2021;19:1874–1882. doi:10.1111/jth.15358
10. Zuily S, Cohen H, Isenberg D, et al. Use of direct oral anticoagulants in patients with thrombotic antiphospholipid syndrome: guidance from the Scientific and Standardization Committee of the International Society on Thrombosis and Haemostasis. J Thromb Haemost. 2020;18:2126–2137. doi:10.1111/jth.14935
11. Cuker A, Arepally GM, Chong BH, et al. American Society of Hematology 2018 guidelines for management of venous thromboembolism: heparin-induced thrombocytopenia. Blood Adv. 2018;2:3360–3392. doi:10.1182/bloodadvances.2018024489
12. Levine GN, McEvoy JW, Fang JC, et al. Management of patients at risk for and with left ventricular thrombus: a scientific statement from the American Heart Association. Circulation. 2022;146:e205–e223. doi:10.1161/CIR.0000000000001092
13. 2023 American Geriatrics Society Beers Criteria Update Expert Panel. American Geriatrics Society 2023 updated AGS Beers Criteria for potentially inappropriate medication use in older adults. J Am Geriatr Soc. 2023;71:2052–2081. doi:10.1111/jgs.18372
14. Bristol-Myers Squibb Company. ELIQUIS (apixaban) tablets: US prescribing information. Revised April 2025. DailyMed
15. Janssen Pharmaceuticals. XARELTO (rivaroxaban) tablets and oral suspension: US prescribing information. Revised June 2025. DailyMed
16. Boehringer Ingelheim Pharmaceuticals. PRADAXA (dabigatran etexilate) capsules: US prescribing information. Revised June 2025. DailyMed
17. Daiichi Sankyo. SAVAYSA (edoxaban) tablets: US prescribing information. Revised October 2023. DailyMed
18. Connolly SJ, Ezekowitz MD, Yusuf S, et al. Dabigatran versus warfarin in patients with atrial fibrillation (RE-LY). N Engl J Med. 2009;361:1139–1151. doi:10.1056/NEJMoa0905561
19. Patel MR, Mahaffey KW, Garg J, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation (ROCKET AF). N Engl J Med. 2011;365:883–891. doi:10.1056/NEJMoa1009638
20. Granger CB, Alexander JH, McMurray JJV, et al. Apixaban versus warfarin in patients with atrial fibrillation (ARISTOTLE). N Engl J Med. 2011;365:981–992. doi:10.1056/NEJMoa1107039
21. Giugliano RP, Ruff CT, Braunwald E, et al. Edoxaban versus warfarin in patients with atrial fibrillation (ENGAGE AF-TIMI 48). N Engl J Med. 2013;369:2093–2104. doi:10.1056/NEJMoa1310907
22. Agnelli G, Buller HR, Cohen A, et al. Oral apixaban for the treatment of acute venous thromboembolism (AMPLIFY). N Engl J Med. 2013;369:799–808. doi:10.1056/NEJMoa1302507
23. EINSTEIN-PE Investigators; Büller HR, Prins MH, Lensing AW, et al. Oral rivaroxaban for the treatment of symptomatic pulmonary embolism. N Engl J Med. 2012;366:1287–1297. doi:10.1056/NEJMoa1113572
24. Hokusai-VTE Investigators; Büller HR, Décousus H, Grosso MA, et al. Edoxaban versus warfarin for the treatment of symptomatic venous thromboembolism. N Engl J Med. 2013;369:1406–1415. doi:10.1056/NEJMoa1306638
25. Schulman S, Kearon C, Kakkar AK, et al. Dabigatran versus warfarin in the treatment of acute venous thromboembolism (RE-COVER). N Engl J Med. 2009;361:2342–2352. doi:10.1056/NEJMoa0906598
26. Weitz JI, Lensing AWA, Prins MH, et al. Rivaroxaban or aspirin for extended treatment of venous thromboembolism (EINSTEIN CHOICE). N Engl J Med. 2017;376:1211–1222. doi:10.1056/NEJMoa1700518
27. Agnelli G, Becattini C, Meyer G, et al. Apixaban for the treatment of venous thromboembolism associated with cancer (Caravaggio). N Engl J Med. 2020;382:1599–1607. doi:10.1056/NEJMoa1915103
28. Mahé I, Carrier M, Mayeur D, et al. Extended reduced-dose apixaban for cancer-associated venous thromboembolism (API-CAT). N Engl J Med. 2025;392:1363–1373. doi:10.1056/NEJMoa2416112
29. Eikelboom JW, Connolly SJ, Bosch J, et al. Rivaroxaban with or without aspirin in stable cardiovascular disease (COMPASS). N Engl J Med. 2017;377:1319–1330. doi:10.1056/NEJMoa1709118
30. Bonaca MP, Bauersachs RM, Anand SS, et al. Rivaroxaban in peripheral artery disease after revascularization (VOYAGER PAD). N Engl J Med. 2020;382:1994–2004. doi:10.1056/NEJMoa2000052
31. Eikelboom JW, Connolly SJ, Brueckmann M, et al. Dabigatran versus warfarin in patients with mechanical heart valves (RE-ALIGN). N Engl J Med. 2013;369:1206–1214. doi:10.1056/NEJMoa1300615
32. Wang TY, Svensson LG, Wen J, et al. Apixaban or warfarin in patients with an On-X mechanical aortic valve (PROACT Xa). NEJM Evid. 2023;2(7):EVIDoa2300067. doi:10.1056/EVIDoa2300067
33. Connolly SJ, Karthikeyan G, Ntsekhe M, et al. Rivaroxaban in rheumatic heart disease-associated atrial fibrillation (INVICTUS). N Engl J Med. 2022;387:978–988. doi:10.1056/NEJMoa2209051
34. Pengo V, Denas G, Zoppellaro G, et al. Rivaroxaban vs warfarin in high-risk patients with antiphospholipid syndrome (TRAPS). Blood. 2018;132:1365–1371. doi:10.1182/blood-2018-04-848333
35. Pokorney SD, Chertow GM, Al-Khalidi HR, et al. Apixaban for patients with atrial fibrillation on hemodialysis: a multicenter randomized controlled trial (RENAL-AF). Circulation. 2022;146:1735–1745. doi:10.1161/CIRCULATIONAHA.121.054990
36. Okumura K, Akao M, Yoshida T, et al. Low-dose edoxaban in very elderly patients with atrial fibrillation (ELDERCARE-AF). N Engl J Med. 2020;383:1735–1745. doi:10.1056/NEJMoa2012883
37. Joosten LPT, van Doorn S, van de Ven PM, et al. Safety of switching from a vitamin K antagonist to a non-vitamin K antagonist oral anticoagulant in frail older patients with atrial fibrillation: results of the FRAIL-AF randomized controlled trial. Circulation. 2024;149:279–289. doi:10.1161/CIRCULATIONAHA.123.066485
38. Yao X, Shah ND, Sangaralingham LR, Gersh BJ, Noseworthy PA. Non-vitamin K antagonist oral anticoagulant dosing in patients with atrial fibrillation and renal dysfunction. J Am Coll Cardiol. 2017;69:2779–2790. doi:10.1016/j.jacc.2017.03.600
39. Douketis JD, Spyropoulos AC, Duncan J, et al. Perioperative management of patients with atrial fibrillation receiving a direct oral anticoagulant (PAUSE). JAMA Intern Med. 2019;179:1469–1478. doi:10.1001/jamainternmed.2019.2431
40. Connolly SJ, Sharma M, Cohen AT, et al. Andexanet for factor Xa inhibitor–associated acute intracerebral hemorrhage (ANNEXA-I). N Engl J Med. 2024;390:1745–1755. doi:10.1056/NEJMoa2313040
41. US Food and Drug Administration. Update on the safety of Andexxa (coagulation factor Xa [recombinant], inactivated-zhzo). FDA safety communication. 2025. fda.gov
42. Piccini JP, Patel MR, Steffel J, et al. Asundexian versus apixaban in patients with atrial fibrillation (OCEANIC-AF). N Engl J Med. 2025;392:23–32. doi:10.1056/NEJMoa2407105
43. Ruff CT, Patel SM, Giugliano RP, et al. Abelacimab versus rivaroxaban in patients with atrial fibrillation (AZALEA-TIMI 71). N Engl J Med. 2025;392:361–371. doi:10.1056/NEJMoa2406674
44. Sharma M, Dong Q, Hirano T, et al. Asundexian for secondary stroke prevention (OCEANIC-STROKE). N Engl J Med. 2026;394:1467–1479. doi:10.1056/NEJMoa2513880
45. Gibson CM, Steg PG, Bahit MC, et al. Milvexian with antiplatelet therapy after acute coronary syndrome event (LIBREXIA ACS). N Engl J Med. 2026;395:1575–1584. doi:10.1056/NEJMoa2608717
46. Agnelli G, Buller HR, Cohen A, et al. Apixaban for extended treatment of venous thromboembolism (AMPLIFY-EXT). N Engl J Med. 2013;368:699–708. doi:10.1056/NEJMoa1207541
47. Cho MS, Kang DY, Ahn JM, et al. Edoxaban antithrombotic therapy for atrial fibrillation and stable coronary artery disease (EPIC-CAD). N Engl J Med. 2024;391:2075–2086. doi:10.1056/NEJMoa2407362
48. Van Mieghem NM, Unverdorben M, Hengstenberg C, et al. Edoxaban versus vitamin K antagonist for atrial fibrillation after TAVR (ENVISAGE-TAVI AF). N Engl J Med. 2021;385:2150–2160. doi:10.1056/NEJMoa2111016
49. Guimarães HP, Lopes RD, de Barros e Silva PGM, et al. Rivaroxaban in patients with atrial fibrillation and a bioprosthetic mitral valve (RIVER). N Engl J Med. 2020;383:2117–2126. doi:10.1056/NEJMoa2029603
50. Hohnloser SH, Hijazi Z, Thomas L, et al. Efficacy of apixaban when compared with warfarin in relation to renal function in patients with atrial fibrillation: insights from the ARISTOTLE trial. Eur Heart J. 2012;33:2821–2830. doi:10.1093/eurheartj/ehs274
51. Ray WA, Chung CP, Stein CM, et al. Association of rivaroxaban vs apixaban with major ischemic or hemorrhagic events in patients with atrial fibrillation. JAMA. 2021;326:2395–2404. doi:10.1001/jama.2021.21222
52. Eworuke E, Hou L, Zhang R, et al. Risk of severe abnormal uterine bleeding associated with rivaroxaban compared with apixaban, dabigatran and warfarin. Drug Saf. 2021;44:753–763. doi:10.1007/s40264-021-01072-0
53. Schiele F, van Ryn J, Canada K, et al. A specific antidote for dabigatran: functional and structural characterization. Blood. 2013;121:3554–3562. doi:10.1182/blood-2012-11-468207
54. Witt DM, Nieuwlaat R, Clark NP, et al. American Society of Hematology 2018 guidelines for management of venous thromboembolism: optimal management of anticoagulation therapy. Blood Adv. 2018;2:3257–3291. doi:10.1182/bloodadvances.2018024893