Direct oral anticoagulants (DOACs)—including dabigatran, rivaroxaban, apixaban, and edoxaban—were developed to overcome many of the limitations associated with warfarin therapy [1]. Initially introduced as alternatives to warfarin for stroke prevention in patients with atrial fibrillation (AF), DOACs have now become the cornerstone of modern anticoagulation management. The standard dosing regimens for each DOAC are as follows: dabigatran is administered at 150 mg twice a day, rivaroxaban at 20 mg once a day, edoxaban at 60 mg once a day, and apixaban at 5 mg twice a day. In specific populations, such as older patients or those with renal dysfunction, reduced doses are recommended—namely, dabigatran 110 mg twice a day, rivaroxaban 15 mg once a day, edoxaban 30 mg once a day, and apixaban 2.5 mg twice a day [2].

For doses below these established reduced regimens, terms like “very-low-dose” or “ultra-low-dose (ULD)” have been used, although the terminology remains inconsistent in the literature. This study aimed to review the indications for ULD DOACs and the current state of research in this area. The concept of ULD DOACs was first introduced in the 2013 European Heart Rhythm Association Practical Guide, which recommended rivaroxaban at doses of 2.5 or 5 mg twice a day [3]. This strategy was intended for patients in whom antiplatelet therapy alone does not sufficiently reduce residual thrombotic risk, given the critical role of factor Xa in thrombosis [4]. However, the guidelines noted that in AF patients who experienced acute coronary syndrome (ACS) within the past year, the use of ULD DOACs alongside dual antiplatelet therapy (DAPT) had not been evaluated and was therefore not recommended.

Despite optimal DAPT following ACS, approximately 10% of patients continue to experience cardiovascular death or ischemic complications [5]. Early attempts to reduce residual thrombotic risk using warfarin improved ACS outcomes as anticipated but significantly increased bleeding events [6]. Moreover, warfarin’s extensive drug and food interactions complicate its routine clinical use. Initial studies using the oral direct thrombin inhibitor ximelagatran in combination with aspirin demonstrated a reduction in ischemic events compared to placebo [7]; however, concerns regarding hepatotoxicity prevented its approval. Following the advent of DOACs, the 2009 ATLAS ACS-TIMI 46 (Anti‐Xa Therapy to Lower Cardiovascular Events in Addition to Aspirin With or Without Thienopyridine Therapy in Subjects With Acute Coronary Syndrome–Thrombolysis in Myocardial Infarction 46) trial assessed whether adding rivaroxaban to aspirin could further reduce ischemic events [4]. As predicted, the trial showed a 31% reduction in the composite endpoint of stroke, myocardial infarction, and death, thereby confirming the persistence of residual thrombotic risk. However, a 20-mg dose of rivaroxaban once a day was associated with an unacceptably high annual bleeding rate of 15.3% compared to 3.3% in the placebo group, and even the 10-mg dose resulted in a threefold increase in bleeding risk. In contrast, a 5-mg dose yielded a relatively acceptable annual bleeding rate of 6.1%. Based on these findings, the subsequent ATLAS ACS-TIMI 51 trial was designed to evaluate dosing regimens of 5 or 2.5 mg twice a day.

The trial enrolled patients within 1 week of an ACS event who were receiving aspirin and a thienopyridine (either clopidogrel or ticlopidine). Participants were randomly assigned to one of three groups—placebo, rivaroxaban 5 mg twice a day, or rivaroxaban 2.5 mg twice a day—and were monitored for both bleeding and ischemic events [8]. Although both rivaroxaban doses (2.5 and 5 mg twice a day) similarly reduced ischemic events, the 5 mg twice a day regimen did not reduce cardiovascular death. Patients receiving 5 mg twice a day experienced more intracranial hemorrhages and non–coronary artery bypass grafting–related bleeding, while fatal bleeding rates were statistically comparable between all patients receiving rivaroxaban (both doses) and those on placebo (0.3% vs. 0.2%, P=0.66). Nonetheless, fatal bleeding was more frequent in the 5 mg twice a day group than in the 2.5 mg twice a day group (0.4% vs. 0.1%, P=0.04), which may explain the lack of mortality benefit. This trial established rivaroxaban 2.5 mg twice a day as the foundation of ULD DOAC therapy.

Although the ATLAS ACS studies provided evidence for the efficacy of adding anticoagulation in the acute ACS setting, the subsequent introduction of potent P2Y12 inhibitors (such as prasugrel and ticagrelor), along with advances in antiplatelet therapy and drug-eluting stents (which have reduced stent thrombosis), has diminished the relative benefit of DOACs during the acute phase [9]. Furthermore, the WOEST (What is the Optimal Antiplatelet and Anticoagulant Therapy in Patients With Oral Anticoagulation and Coronary Stenting) trial demonstrated that triple therapy—combining DAPT with oral anticoagulants—in patients within one year of percutaneous coronary intervention increased bleeding risk without reducing thrombotic events. Consequently, research into the use of stronger anticoagulants in this setting has declined [10]. Nevertheless, secondary prevention remains crucial for patients with persistently high residual risk beyond 1-year post-ACS, particularly for those with polyvascular disease involving peripheral arterial disease or cerebrovascular disease. These patients face a significantly increased risk of recurrence, necessitating additional antithrombotic therapy [11–13]. In response to this need, the COMPASS (Cardiovascular Outcomes for People Using Anticoagulation Strategies) trial was conducted [14], designed as a 2×3 partial factorial study that investigated both the role of proton-pump inhibitors and whether low-dose DOAC therapy could replace or complement aspirin in these high-risk populations [15,16].

The COMPASS trial required patients to have coronary artery disease (CAD), peripheral arterial disease (PAD), or both. Patients under 65 years were also required to have at least two of the following risk factors: current smoking, diabetes mellitus, an estimated glomerular filtration rate (eGFR) below 60 mL/min/1.73 m², heart failure, or a nonlacunar ischemic stroke that had occurred more than 1 month previously. Participants were randomized into three groups: aspirin (100 mg once a day), rivaroxaban (5 mg twice a day), or a combination of aspirin (100 mg once a day) and rivaroxaban (2.5 mg twice a day). An interim analysis demonstrated that the combination therapy was superior to aspirin alone for the composite endpoint of cardiovascular death, stroke, or myocardial infarction (4.1% vs. 5.4%; hazard ratio [HR], 0.76; 95% confidence interval [CI], 0.66–0.86; P<0.001), leading to early termination of the trial [14] (Table 1). In patients with CAD, the combination also reduced all-cause mortality (3.4% vs. 4.1%; HR, 0.82; 95% CI, 0.71–0.96; P<0.01), whereas in patients with PAD it significantly reduced acute limb ischemia and cardiovascular death; all-cause mortality, however, did not reach statistical significance. Based on these findings, current guidelines from the American College of Cardiology [17] and the European Society of Cardiology [18] now endorse the use of aspirin plus rivaroxaban 2.5 mg twice a day in high-risk patients with chronic coronary syndrome. However, it is important to note that ULD rivaroxaban was associated with a higher rate of major bleeding (3.1% vs. 1.9%; HR, 1.70; 95% CI, 1.40–2.05; P<0.001).

The ELDERCARE-AF (Edoxaban Low-Dose for Elder Care Atrial Fibrillation Patients) trial is one of the most representative ULD studies for edoxaban [19]. This trial was based on the findings of the ENGAGE AF-TIMI 48 (Effective Anticoagulation With Factor Xa Next Generation in Atrial Fibrillation–Thrombolysis in Myocardial Infarction 48) study [20], which compared edoxaban at doses of 60 and 30 mg with warfarin to establish a standard dosing regimen. In ENGAGE AF-TIMI 48, dose reductions were applied to patients with impaired renal function (eGFR, 30–50 mL/min/1.73 m2), body weight ≤60 kg, or those taking concomitant verapamil or quinidine (potent P-glycoprotein inhibitors). For these patients, doses were systematically lowered from 60 to 30 mg or from 30 to 15 mg. Patients aged 80 years or older comprised 17% of the study population. Among patients aged ≥75 years, the risk of stroke and systemic embolism increased (1.1% vs. 1.6%), while the risk of major bleeding escalated more steeply with age (1.8% vs. 4.1%), thereby emphasizing the need for dose reduction in elderly populations. Notably, in patients over 80 years, the 30-mg edoxaban group exhibited the lowest rates of major bleeding and intracranial hemorrhage compared to warfarin, thus paving the way for the ELDERCARE-AF trial.

The ELDERCARE-AF trial, conducted in Japan, demonstrated that an edoxaban regimen of 15 mg once a day in patients aged 80 years or older significantly reduced the incidence of stroke and systemic embolism compared to placebo (2.3% vs. 6.7%; HR, 0.34, 95% CI, 0.19–0.61; P<0.001). With regard to bleeding risk, the annual bleeding rate was marginally higher in the edoxaban group (3.3% vs. 1.8%), although the difference was not statistically significant (HR, 1.87; 95% CI, 0.90–3.89; P=0.09) [19] (Table 1). While previous DOAC studies primarily enrolled patients in their early 70s, the ELDERCARE-AF trial was the first to show that ULD edoxaban reduces the risk of stroke and systemic embolism in very elderly patients (aged ≥80 years) with nonvalvular atrial fibrillation. These findings have since supported the use of ULD DOACs in this high-risk population, particularly among those with pronounced frailty or extremely low body weight (<45 kg) [21,22].

Research on dabigatran in ULD regimens is limited, largely because its mechanism as a direct thrombin inhibitor differs from that of factor Xa inhibitors; furthermore, no evidence links it to residual thrombotic risk. Although dabigatran 75 mg twice a day is approved in certain regions—particularly outside Korea—for patients with severe renal impairment (creatinine clearance, 15–30 mL/min) and has been evaluated in pharmacokinetic/pharmacodynamic studies, research in patients aged 65 years and older has demonstrated that while the dose of 75 mg twice a day reduces the incidence of intracranial hemorrhage, its overall efficacy is not significantly different from that of warfarin. Furthermore, a significant reduction in overall mortality was observed only in patients receiving the standard 150 mg regimen, which also effectively reduced ischemic events (150 mg: HR, 0.76 [95% CI, 0.67–0.86]; 75 mg: HR, 0.95 [95% CI, 0.78–1.16]) [23]. In addition, the ULD dabigatran regimen of 75 mg twice a day did not significantly reduce thromboembolic events [24], and no further studies have explored doses lower than 75 mg.

Apixaban has already demonstrated high efficacy along with a low incidence of bleeding at its standard dosage, reducing the need for further exploration of dose reductions [25]. Moreover, it provides noninferior protection against ischemic events compared to other agents. In fact, studies indicate that when apixaban is used off-label at a lower dose (2.5 mg twice a day), it does not confer the same benefits as warfarin [26]. Consequently, there are few investigations exploring doses lower than the standard 2.5 mg twice a day. Although studies have examined a 1.25-mg dose twice daily regimen for preventing venous thromboembolism [27] and a 2.5-mg dose once daily regimen in dialysis patients, which found that the drug concentration was inadequate [28], it is unlikely that further research in this area will be pursued.

The primary objective of DOAC therapy is to reduce ischemic events resulting from systemic thromboembolism in patients with AF. While reducing the dose to mitigate bleeding risk may seem rational, it is only appropriate if adequate ischemic protection is maintained; dose reduction is not inherently beneficial. A DOAC study conducted in Korea found that 36.4% of DOAC users were prescribed off‐label underdosing. These patients were more frequently over 75 years old, female, weighed less than 60 kg, had a creatinine clearance ≤50 mL/min, and had histories of stroke or bleeding, hypertension, or concurrent use of dronedarone or antiplatelet agents [29]. Recent research on apixaban showed that among patients meeting only one of the three dose-reduction criteria (age ≥80 years, body weight ≤60 kg, or serum creatinine ≥1.5 mg/dL), off‐label underdosing occurred in up to 51.3% of cases among 2,000 participants. This practice was especially common among patients who were borderline frail (age, 75–80 years; weight, 60–65 kg; serum creatinine level, 1.2–1.5 mg/dL), whereas physicians tended to prescribe the standard dose to patients with a history of stroke [30]. These findings suggest that subjective judgments regarding frailty or bleeding risk often drive the decision to employ off‐label low doses or ULD, rather than objective criteria. Moreover, in patients who qualify for standard dosing, off‐label underdosing did not confer clinical benefits superior to those observed with warfarin, although outcomes may vary across different DOACs [31]. Therefore, it is crucial to approach the use of ULD DOACs with caution in the absence of robust evidence from large-scale studies.

The clinical application of ULD DOAC therapy should focus on striking a balance between reducing bleeding risk and maintaining effective ischemic protection. Although some studies (such as those investigating rivaroxaban 2.5 mg twice a day in combination with aspirin and edoxaban at 15 mg) have demonstrated the potential efficacy of ULD regimens, off-label underdosing below recommended levels may compromise anticoagulant efficacy and diminish thromboembolism prevention. Therefore, the indiscriminate use of ULD DOACs without robust evidence from large-scale studies should be approached with caution.