{"product_id":"blood-thinners-for-atrial-fibrillation-in-patients-at-intermediate-stroke-risk-what-a-major-randomized-trial-found","title":"Blood Thinners for Atrial Fibrillation in Patients at Intermediate Stroke Risk: What a Major Randomized Trial Found","description":"\u003cp\u003eA large randomized trial was conducted at 18 centers in South Korea. The trial enrolled patients with atrial fibrillation, an irregular heartbeat that raises stroke risk. These patients had an intermediate risk of stroke. They were randomly assigned to take a direct oral anticoagulant (DOAC, a blood-thinning pill) or to take no anticoagulant. After 24 months, the primary combined endpoint was stroke, systemic embolism, major bleeding, or death from cardiovascular causes. This endpoint occurred in 4 of 902 patients (0.5%) in the DOAC group. It occurred in 13 of 901 patients (1.5%) in the no-anticoagulant group. This difference of −1.0 percentage points was statistically significant (P = 0.03), with a hazard ratio of 0.31. The authors concluded that DOAC therapy lowered the risk of this combined endpoint compared with no anticoagulation in this patient group.\u003c\/p\u003e\n\n\u003ch1\u003eBlood Thinners for Atrial Fibrillation in Patients at Intermediate Stroke Risk: What a Major Randomized Trial Found\u003c\/h1\u003e\n\n\u003ch2\u003eTable of Contents\u003c\/h2\u003e\n\u003cul\u003e\n\u003cli\u003e\u003ca href=\"#ddn-key-points\"\u003eKey Points\u003c\/a\u003e\u003c\/li\u003e\n\n\u003cli\u003e\u003ca href=\"#background\"\u003eBackground: Why This Question Matters\u003c\/a\u003e\u003c\/li\u003e\n\u003cli\u003e\u003ca href=\"#methods\"\u003eHow the Study Was Conducted\u003c\/a\u003e\u003c\/li\u003e\n\u003cli\u003e\u003ca href=\"#population\"\u003eWho Took Part in the Study\u003c\/a\u003e\u003c\/li\u003e\n\u003cli\u003e\u003ca href=\"#treatments\"\u003eTreatments and Follow-Up\u003c\/a\u003e\u003c\/li\u003e\n\u003cli\u003e\u003ca href=\"#endpoints\"\u003eWhat the Study Measured\u003c\/a\u003e\u003c\/li\u003e\n\u003cli\u003e\u003ca href=\"#statistics\"\u003eHow the Results Were Analyzed\u003c\/a\u003e\u003c\/li\u003e\n\u003cli\u003e\u003ca href=\"#findings\"\u003eKey Findings\u003c\/a\u003e\u003c\/li\u003e\n\u003cli\u003e\u003ca href=\"#subgroups\"\u003eFindings in Subgroups and Sensitivity Analyses\u003c\/a\u003e\u003c\/li\u003e\n\u003cli\u003e\u003ca href=\"#implications\"\u003eWhat These Findings Mean for Patients\u003c\/a\u003e\u003c\/li\u003e\n\u003cli\u003e\u003ca href=\"#limitations\"\u003eLimitations of the Study\u003c\/a\u003e\u003c\/li\u003e\n\u003cli\u003e\u003ca href=\"#recommendations\"\u003ePractical Points for Patients and Families\u003c\/a\u003e\u003c\/li\u003e\n\u003cli\u003e\u003ca href=\"#ddn-faq\"\u003eFrequently Asked Questions\u003c\/a\u003e\u003c\/li\u003e\n\u003cli\u003e\u003ca href=\"#source\"\u003eSource Information\u003c\/a\u003e\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003c!-- ddn:keypoints:start --\u003e\n\u003ch2 id=\"ddn-key-points\"\u003eKey Points\u003c\/h2\u003e\n\u003cul\u003e\n\u003cli\u003eIn a randomized trial of 1803 patients with atrial fibrillation at intermediate stroke risk, a DOAC reduced the combined endpoint of stroke, systemic embolism, major bleeding, or cardiovascular death at 24 months.\u003c\/li\u003e\n\u003cli\u003eThe combined endpoint occurred in 0.5% of the DOAC group versus 1.5% of the no-anticoagulant group, a statistically significant difference of 1.0 percentage point.\u003c\/li\u003e\n\u003cli\u003eStroke alone occurred in 0.3% of the DOAC group and 1.1% of the no-anticoagulant group, but this result did not reach conventional statistical significance.\u003c\/li\u003e\n\u003cli\u003eMajor bleeding rates were similar between groups, and serious adverse events occurred in 8.9% of the DOAC group and 9.3% of the no-anticoagulant group.\u003c\/li\u003e\n\u003cli\u003eThe trial was open-label, enrolled patients in South Korea, and had few primary endpoint events, so results should be interpreted with these limitations in mind.\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c!-- ddn:keypoints:end --\u003e\n\n\n\u003ch2 id=\"background\"\u003eBackground: Why This Question Matters\u003c\/h2\u003e\n\n\u003cp\u003eAtrial fibrillation is a common heart rhythm disorder. It increases the risk of stroke. This fact is well established in the medical literature.\u003c\/p\u003e\n\n\u003cp\u003eDoctors use a scoring system called the CHA\u003csub\u003e2\u003c\/sub\u003eDS\u003csub\u003e2\u003c\/sub\u003e-VASc scale to estimate a patient's stroke risk. The scale ranges from 0 to 9. Higher scores mean a greater risk of stroke. The score is weighted on the basis of several factors. These factors include congestive heart failure, high blood pressure, diabetes, and vascular disease. They also include a prior stroke or transient ischemic attack, a temporary stroke-like episode. Other factors are age 65 to 74 years or 75 years or older, and female sex.\u003c\/p\u003e\n\n\u003cp\u003eFor patients with higher scores on this scale, the expected reduction in the risk of thromboembolism generally outweighs the risk of bleeding. Thromboembolism means blood clots that travel and block vessels. This benefit applies with oral anticoagulant therapy. Current U.S. and European guidelines strongly recommend oral anticoagulants for men with a score of 2 or more and for women with a score of 3 or more.\u003c\/p\u003e\n\n\u003cp\u003eThe picture is less clear for patients at intermediate risk of stroke. Intermediate risk is defined as a CHA\u003csub\u003e2\u003c\/sub\u003eDS\u003csub\u003e2\u003c\/sub\u003e-VASc score of 1 in men and 2 in women. For these patients, current guidelines provide only a class IIa recommendation for considering oral anticoagulants. A class IIa recommendation means the treatment is reasonable to consider, but the evidence is not as strong as for a class I recommendation.\u003c\/p\u003e\n\n\u003cp\u003eEvidence from randomized trials supporting this approach has been limited. Observational studies have provided conflicting evidence about the net benefit of oral anticoagulants in patients at intermediate risk. This gap in knowledge is exactly what the SINGLE-AF trial was designed to address.\u003c\/p\u003e\n\n\u003cp\u003eDirect oral anticoagulants offer a better balance of risks and benefits than vitamin K antagonists (older blood thinners such as warfarin) in patients with atrial fibrillation. The main reason is that DOACs confer a lower risk of intracranial hemorrhage (bleeding inside the skull). Some phase 3 trials of DOACs included patients at low risk for stroke. However, those trials did not directly compare anticoagulation with no treatment in intermediate-risk patients.\u003c\/p\u003e\n\n\u003cp\u003eThe investigators designed the Efficacy and Safety of Non–Vitamin K Antagonist Oral Anticoagulants for Intermediate Stroke Risk in Patients with Atrial Fibrillation (SINGLE-AF) trial. The goal was to assess whether the risk of adverse clinical events would be lower with oral anticoagulation than with no anticoagulation. The goal applied among patients with atrial fibrillation who are at intermediate risk for stroke.\u003c\/p\u003e\n\n\u003ch2 id=\"methods\"\u003eHow the Study Was Conducted\u003c\/h2\u003e\n\n\u003cp\u003eThe SINGLE-AF trial was an investigator-initiated, multicenter, open-label, randomized superiority trial. \"Open-label\" means that patients and their doctors knew which treatment was assigned. \"Randomized\" means that patients were assigned to treatment groups by chance. \"Superiority trial\" means the researchers tested whether one treatment was better than the other.\u003c\/p\u003e\n\n\u003cp\u003eOutcome events were adjudicated by reviewers who were masked to treatment assignment. In other words, the people who judged whether an event had occurred did not know which group the patient was in. This design reduces the chance of bias.\u003c\/p\u003e\n\n\u003cp\u003eThe trial took place at 18 centers in South Korea. The rationale and design have been described previously. The trial protocol and all subsequent amendments were approved by the institutional review board or ethics committee at each participating site. All patients provided written informed consent before randomization.\u003c\/p\u003e\n\n\u003cp\u003eThe trial was funded by the Ministry of Health and Welfare in South Korea, Samjin Pharmaceutical, and Hanmi Pharmaceutical. The funders had no role in the design or conduct of the trial. The funders also had no role in the collection, management, analysis, or interpretation of the data. They had no role in the preparation, review, or approval of the manuscript.\u003c\/p\u003e\n\n\u003cp\u003eTrial data were reviewed by an independent data and safety monitoring board. The first four authors and the last author wrote the first and subsequent drafts of the manuscript. These authors had full access to the data and participated in the analysis and interpretation of the data. The authors made the decision to submit the manuscript for publication. They vouch for the completeness and accuracy of the data and for the adherence of the trial to the protocol.\u003c\/p\u003e\n\n\u003ch2 id=\"population\"\u003eWho Took Part in the Study\u003c\/h2\u003e\n\n\u003cp\u003ePatients were eligible to participate if they were at least 19 years of age, had atrial fibrillation, and were at intermediate risk for stroke. Intermediate risk was defined as a CHA\u003csub\u003e2\u003c\/sub\u003eDS\u003csub\u003e2\u003c\/sub\u003e-VASc score of 1 in men or 2 in women.\u003c\/p\u003e\n\n\u003cp\u003eThe trial excluded patients with certain conditions. Key exclusion criteria were as follows:\u003c\/p\u003e\n\n\u003cul\u003e\n\u003cli\u003eClinically significant kidney disease, defined as a serum creatinine level of 3.5 mg per deciliter or higher, or a creatinine clearance of less than 30 ml per minute\u003c\/li\u003e\n\u003cli\u003eClinically significant liver disease was defined as an aspartate aminotransferase or alanine aminotransferase level more than 3 times the upper limit of the normal range. Clinically significant liver disease was also defined as Child–Pugh class B or C disease.\u003c\/li\u003e\n\u003cli\u003eUse of anticoagulation because of a mechanical prosthetic valve, moderate-to-severe mitral stenosis, or deep-vein thrombosis\u003c\/li\u003e\n\u003cli\u003eClinically significant structural heart disease, such as dilated or hypertrophic cardiomyopathy and cardiac amyloidosis\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003cp\u003eFrom July 27, 2020, through June 29, 2023, a total of 1803 patients underwent randomization. Of these, 902 were assigned to the DOAC group and 901 to the no-anticoagulant group.\u003c\/p\u003e\n\n\u003cp\u003eThe characteristics of the patients at baseline appeared to be well balanced between the two trial groups. The mean age of the patients was 60.4 years, with a standard deviation of 7.9 years. Women made up 23.7% of the patients. Paroxysmal atrial fibrillation (atrial fibrillation that comes and goes) was present in 71.8% of patients. The median time from the diagnosis of atrial fibrillation to randomization was 21 months, with an interquartile range of 2 to 59 months.\u003c\/p\u003e\n\n\u003cp\u003eBaseline characteristics were similar between the two groups. In the DOAC group, the mean age was 60.7 years, and 75.6% were men. In the no-anticoagulant group, the mean age was 60.0 years, and 77.0% were men. Paroxysmal atrial fibrillation was present in 70.5% of the DOAC group and 73.0% of the no-anticoagulant group.\u003c\/p\u003e\n\n\u003cp\u003eOther baseline features were also balanced. High blood pressure was present in 48.2% of the DOAC group and 53.5% of the no-anticoagulant group. Diabetes mellitus was present in 7.4% and 7.8%, respectively. Heart failure was present in 10.3% and 8.0%. The mean CHA\u003csub\u003e2\u003c\/sub\u003eDS\u003csub\u003e2\u003c\/sub\u003e-VASc score was 1.3 in both groups. The mean HAS-BLED score, which estimates bleeding risk, was 0.5 in both groups.\u003c\/p\u003e\n\n\u003ch2 id=\"treatments\"\u003eTreatments and Follow-Up\u003c\/h2\u003e\n\n\u003cp\u003eEligible patients were randomly assigned in a 1:1 ratio to receive DOAC therapy or no anticoagulant therapy. Randomization was performed with a Web-based system. The system used a permuted-block design with block sizes of 4 or 6, stratified according to trial site.\u003c\/p\u003e\n\n\u003cp\u003ePatients assigned to the DOAC group received either apixaban (5 mg twice daily) or rivaroxaban (20 mg once daily). The choice was made at the treating physician's discretion. Appropriate dose reductions were applied for patients who met prespecified criteria. If patients had unacceptable side effects with apixaban or rivaroxaban, another DOAC could be prescribed at the investigator's discretion.\u003c\/p\u003e\n\n\u003cp\u003ePatients assigned to the no-anticoagulant group did not receive routine oral anticoagulation as part of their assigned trial regimen. Temporary administration of a DOAC was permitted before and after rhythm-control interventions, such as cardioversion or catheter ablation for atrial fibrillation.\u003c\/p\u003e\n\n\u003cp\u003eIn both groups, the use of antiplatelet therapy was permitted if it was clinically indicated. Examples include after coronary intervention or an acute coronary syndrome. Antiplatelet therapy was also permitted at the physician's discretion. However, routine use was not recommended.\u003c\/p\u003e\n\n\u003cp\u003eFollow-up assessments were performed at baseline and at 3, 6, 12, 18, and 24 months after randomization. At each visit, the researchers assessed the patients' general health status, the use of cardiovascular medication, and the occurrence of trial end-point events or adverse events.\u003c\/p\u003e\n\n\u003cp\u003eDuring the trial period, all 902 patients (100%) in the DOAC group received a DOAC. In the no-anticoagulant group, 50 of 901 patients (5.5%) received a DOAC. The most commonly prescribed therapy in the DOAC group was apixaban at a dose of 5 mg, used in 67.4% of the patients. Rivaroxaban at a dose of 20 mg was used in 14.1%, and rivaroxaban at 15 mg was used in 13.9%. DOACs were administered at full doses in 81.6% of these patients and at reduced doses in 18.4%.\u003c\/p\u003e\n\n\u003cp\u003eAntiplatelet therapy was used in 0.4% of the patients in the DOAC group and in 36.5% of those in the no-anticoagulant group. Antiarrhythmic drug therapy was used in 36.0% of the DOAC group and 38.6% of the no-anticoagulant group. Cardioversion was performed in 2.1% and 1.3%, respectively. Catheter ablation was performed in 7.8% and 6.2%.\u003c\/p\u003e\n\n\u003cp\u003eDuring follow-up, 151 patients (16.7%) in the DOAC group and 120 patients (13.3%) in the no-anticoagulant group had an increase in CHA\u003csub\u003e2\u003c\/sub\u003eDS\u003csub\u003e2\u003c\/sub\u003e-VASc score to 2 or higher in men or 3 or higher in women. The median follow-up was 730 days. At 24 months, 94.6% of the patients in both groups had completed follow-up.\u003c\/p\u003e\n\n\u003ch2 id=\"endpoints\"\u003eWhat the Study Measured\u003c\/h2\u003e\n\n\u003cp\u003eThe primary end point was a composite of four events at 24 months after randomization. The four events were stroke, systemic embolism, major bleeding, and death from cardiovascular causes. Major bleeding was defined by the International Society on Thrombosis and Haemostasis.\u003c\/p\u003e\n\n\u003cp\u003eThe primary end point originally included death from any cause. In January 2025, the members of the independent data and safety monitoring board noted something important. The board members were unaware of the trial-group assignments. They observed that all the deaths reported up to that point were due to noncardiovascular causes and were unlikely to be related to treatment.\u003c\/p\u003e\n\n\u003cp\u003eTo prevent these unrelated events from diluting the ability to assess the treatment effect, the primary end point was amended. The amended primary end point included only death from cardiovascular causes. Death from any cause was retained as a secondary outcome. This change did not alter the original sample-size or power assumptions.\u003c\/p\u003e\n\n\u003cp\u003eSecondary end points included the individual components of the primary outcome. These components were stroke, systemic embolism, major bleeding, and death from cardiovascular causes. Other secondary end points were death from any cause, transient ischemic attack, clinically relevant nonmajor bleeding, myocardial infarction (heart attack), and hospitalization for any cause.\u003c\/p\u003e\n\n\u003cp\u003eIn post hoc analyses, the researchers evaluated a composite of ischemic stroke or systemic embolism. They also evaluated a composite of major or clinically relevant nonmajor bleeding. Major or clinically relevant nonmajor bleeding was defined primarily in accordance with the criteria of the International Society on Thrombosis and Haemostasis. Bleeding events were additionally classified according to the Bleeding Academic Research Consortium definitions.\u003c\/p\u003e\n\n\u003cp\u003eAll suspected end-point events were adjudicated by an independent clinical-events committee. The members of this committee were unaware of the trial-group assignments.\u003c\/p\u003e\n\n\u003ch2 id=\"statistics\"\u003eHow the Results Were Analyzed\u003c\/h2\u003e\n\n\u003cp\u003eThe researchers hypothesized that DOAC therapy would be superior to no anticoagulant therapy with respect to the primary end point. They based their sample-size calculation on previously reported outcomes. These included a subgroup analysis of the ARISTOTLE (Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation) trial and observational data.\u003c\/p\u003e\n\n\u003cp\u003eThe researchers assumed that the incidence of the primary end-point events at 24 months would be 2.0% in the DOAC group and 4.5% in the no-anticoagulant group. They calculated that enrollment of 1800 patients, with 900 per group, would provide the trial with 80% power. This level of power would allow detection of a between-group difference of 2.5 percentage points at a two-sided alpha level of 0.05. The calculation assumed an attrition of 12%.\u003c\/p\u003e\n\n\u003cp\u003eThe primary analyses were conducted in the intention-to-treat population. This means that patients were analyzed according to the group to which they were assigned, regardless of whether they followed the assigned treatment.\u003c\/p\u003e\n\n\u003cp\u003eThe cumulative incidence of the primary end point at 24 months was estimated with the Aalen–Johansen method. Death from noncardiovascular causes was treated as a competing event. The cumulative incidence of death from any cause was estimated with the Kaplan–Meier method. Treatment effects were assessed with Fine–Gray subdistribution hazard models.\u003c\/p\u003e\n\n\u003cp\u003eThe absolute between-group differences in cumulative incidence at 24 months were calculated from the Aalen–Johansen estimates. Standard errors were derived from the corresponding estimated variances. Subgroup analyses of the primary end point were performed according to prespecified clinical characteristics. In addition, the researchers performed a post hoc subgroup analysis according to the type of atrial fibrillation, either paroxysmal or nonparoxysmal.\u003c\/p\u003e\n\n\u003cp\u003eIn a sensitivity analysis, follow-up data were censored when a patient's CHA\u003csub\u003e2\u003c\/sub\u003eDS\u003csub\u003e2\u003c\/sub\u003e-VASc score increased to 2 or higher in men or 3 or higher in women. At that point, oral anticoagulant therapy was allowed in accordance with class I guideline recommendations. In an additional sensitivity analysis, the researchers calculated the win ratio using an unmatched-pairs approach.\u003c\/p\u003e\n\n\u003cp\u003eAll possible pairs consisting of one patient from the DOAC group and one patient from the no-anticoagulant group were compared sequentially. The comparison followed a prespecified hierarchical sequence: death from cardiovascular causes, intracranial hemorrhage, ischemic stroke or systemic embolism, and major bleeding excluding intracranial hemorrhage.\u003c\/p\u003e\n\n\u003cp\u003eNo imputation was performed for missing baseline variables. The statistical analysis plan did not include a provision for correcting for multiplicity when conducting tests for secondary end points and in subgroup analyses. Therefore, the widths of the confidence intervals have not been adjusted for multiplicity. The researchers state that these intervals should not be used to infer definitive treatment effects.\u003c\/p\u003e\n\n\u003cp\u003eAll statistical analyses were performed with the use of R software, version 4.2.2.\u003c\/p\u003e\n\n\u003ch2 id=\"findings\"\u003eKey Findings\u003c\/h2\u003e\n\n\u003cp\u003eAt 24 months after randomization, a primary end-point event had occurred in 4 of 902 patients in the DOAC group. This represents a cumulative incidence of 0.5%. In the no-anticoagulant group, a primary end-point event had occurred in 13 of 901 patients, a cumulative incidence of 1.5%.\u003c\/p\u003e\n\n\u003cp\u003eThe difference between the groups was −1.0 percentage points, with a 95% confidence interval of −2.0 to −0.1. The P value was 0.03. The hazard ratio was 0.31, with a 95% confidence interval of 0.10 to 0.94.\u003c\/p\u003e\n\n\u003cp\u003eIn plain language, the hazard ratio of 0.31 means that patients in the DOAC group had about a 69% lower risk of the combined primary endpoint over 24 months. This risk was compared with patients in the no-anticoagulant group. The P value of 0.03 means there is about a 3% probability that a difference this large or larger would have appeared by chance alone if the treatments were truly equivalent. The result was statistically significant.\u003c\/p\u003e\n\n\u003cp\u003eThe researchers tested the proportional subdistribution hazards and proportional-hazards assumptions for the primary end point. They used a time-by-treatment interaction term and Schoenfeld residuals. These tests showed no evidence of violation, with a P value of 0.18 for the Cox proportional-hazards model.\u003c\/p\u003e\n\n\u003cp\u003eStroke occurred in 3 patients in the DOAC group, a cumulative incidence of 0.3%. Stroke occurred in 10 patients in the no-anticoagulant group, a cumulative incidence of 1.1%. The hazard ratio was 0.30, with a 95% confidence interval of 0.08 to 1.08.\u003c\/p\u003e\n\n\u003cp\u003eSystemic embolism occurred in no patients in the DOAC group. It occurred in 1 patient in the no-anticoagulant group, a cumulative incidence of 0.1%.\u003c\/p\u003e\n\n\u003cp\u003eMajor bleeding occurred in 3 patients in the DOAC group, a cumulative incidence of 0.3%. It occurred in 4 patients in the no-anticoagulant group, a cumulative incidence of 0.5%. The hazard ratio was 0.75, with a 95% confidence interval of 0.17 to 3.35.\u003c\/p\u003e\n\n\u003cp\u003eClinically relevant nonmajor bleeding occurred in 22 patients in the DOAC group, a cumulative incidence of 2.5%. It occurred in 14 patients in the no-anticoagulant group, a cumulative incidence of 1.6%. The hazard ratio was 1.58, with a 95% confidence interval of 0.81 to 3.08.\u003c\/p\u003e\n\n\u003cp\u003eDeath from any cause occurred in 3 patients in the DOAC group, a cumulative incidence of 0.3%. It occurred in 5 patients in the no-anticoagulant group, a cumulative incidence of 0.6%. The hazard ratio was 0.60, with a 95% confidence interval of 0.14 to 2.51.\u003c\/p\u003e\n\n\u003cp\u003eNo myocardial infarction or deaths from cardiovascular causes occurred in either group. A transient ischemic attack occurred in no patients in the DOAC group and in 3 patients in the no-anticoagulant group, a cumulative incidence of 0.3%.\u003c\/p\u003e\n\n\u003cp\u003eIn post hoc analyses, ischemic stroke or systemic embolism had occurred by 24 months in 1 patient in the DOAC group, a cumulative incidence of 0.1%. It had occurred in 10 patients in the no-anticoagulant group, a cumulative incidence of 1.1%. The hazard ratio was 0.10, with a 95% confidence interval of 0.01 to 0.77.\u003c\/p\u003e\n\n\u003cp\u003eSerious adverse events occurred in 80 patients (8.9%) in the DOAC group and in 84 patients (9.3%) in the no-anticoagulant group.\u003c\/p\u003e\n\n\u003ch2 id=\"subgroups\"\u003eFindings in Subgroups and Sensitivity Analyses\u003c\/h2\u003e\n\n\u003cp\u003eThe results of a sensitivity analysis were consistent with the primary analysis. In this analysis, follow-up was censored when the CHA\u003csub\u003e2\u003c\/sub\u003eDS\u003csub\u003e2\u003c\/sub\u003e-VASc score increased to a guideline-based class I indication for anticoagulation.\u003c\/p\u003e\n\n\u003cp\u003eThe win-ratio comparison of DOAC therapy to no anticoagulant therapy was 3.31, with a 95% confidence interval of 1.15 to 9.54. The win ratio is a statistical method that compares pairs of patients across a ranked list of outcomes.\u003c\/p\u003e\n\n\u003cp\u003eWith regard to the primary end point, the effect of DOAC therapy as compared with no anticoagulant therapy appeared to be consistent across the subgroups. This means the benefit did not appear to be limited to one particular type of patient.\u003c\/p\u003e\n\n\u003ch2 id=\"implications\"\u003eWhat These Findings Mean for Patients\u003c\/h2\u003e\n\n\u003cp\u003eThis trial provides randomized evidence for a question that guidelines had previously answered only with a weaker recommendation. Before this trial, the recommendation to consider oral anticoagulants in intermediate-risk patients rested mainly on observational studies with conflicting results.\u003c\/p\u003e\n\n\u003cp\u003eThe trial found that DOAC therapy led to a lower risk of the combined endpoint at 24 months than no anticoagulation. The combined endpoint was stroke, systemic embolism, major bleeding, or death from cardiovascular causes. The absolute difference was 1.0 percentage point. In other words, about 1 in 100 patients treated with a DOAC avoided one of these events over two years compared with patients who took no anticoagulant.\u003c\/p\u003e\n\n\u003cp\u003eThe stroke results deserve careful reading. Stroke occurred in 0.3% of the DOAC group and 1.1% of the no-anticoagulant group. The hazard ratio was 0.30, which suggests a large relative reduction. However, the 95% confidence interval ranged from 0.08 to 1.08 and crossed 1.0. This means the stroke result alone did not reach conventional statistical significance. The primary endpoint, which combined several events, did reach statistical significance.\u003c\/p\u003e\n\n\u003cp\u003eBleeding results were also important. Major bleeding occurred in 0.3% of the DOAC group and 0.5% of the no-anticoagulant group. The incidence of systemic embolism and major bleeding appeared to be similar in the two trial groups. Clinically relevant nonmajor bleeding was numerically more common in the DOAC group, at 2.5% versus 1.6%, but the confidence interval crossed 1.0.\u003c\/p\u003e\n\n\u003cp\u003eNo deaths from cardiovascular causes occurred in either group. This is a notable finding. It means the primary endpoint difference was driven mainly by strokes and systemic embolic events, not by cardiovascular deaths.\u003c\/p\u003e\n\n\u003cp\u003eSerious adverse events were similar between groups, at 8.9% in the DOAC group and 9.3% in the no-anticoagulant group. This suggests that adding a DOAC did not produce a large overall increase in serious safety problems in this population.\u003c\/p\u003e\n\n\u003ch2 id=\"limitations\"\u003eLimitations of the Study\u003c\/h2\u003e\n\n\u003cp\u003eThe trial had several features that patients and clinicians should keep in mind when interpreting the results.\u003c\/p\u003e\n\n\u003cp\u003eFirst, the trial was open-label. Patients and their doctors knew which treatment was assigned. The researchers reduced the risk of bias by masking the adjudication of outcome events, but open-label design can still influence how care is delivered.\u003c\/p\u003e\n\n\u003cp\u003eSecond, the statistical analysis plan did not include a provision for correcting for multiplicity when testing secondary end points and in subgroup analyses. This means the confidence intervals for those results have not been adjusted. The researchers explicitly state that these intervals should not be used to infer definitive treatment effects. The stroke result, the bleeding results, and the subgroup results should therefore be viewed as supportive rather than conclusive on their own.\u003c\/p\u003e\n\n\u003cp\u003eThird, the trial enrolled patients in South Korea. The results may not apply in exactly the same way to other populations with different backgrounds and different patterns of care.\u003c\/p\u003e\n\n\u003cp\u003eFourth, the number of primary endpoint events was small. Only 4 events occurred in the DOAC group and 13 in the no-anticoagulant group. Small event numbers make the estimates less precise, which is reflected in the wide confidence intervals.\u003c\/p\u003e\n\n\u003cp\u003eFifth, the trial compared DOAC therapy with no anticoagulation. It did not compare DOACs with vitamin K antagonists such as warfarin in this intermediate-risk population.\u003c\/p\u003e\n\n\u003cp\u003eSixth, the primary end point was amended during the trial. The original primary end point included death from any cause. The amended primary end point included only death from cardiovascular causes. The researchers state that this change did not alter the original sample-size or power assumptions, and the change has been described previously.\u003c\/p\u003e\n\n\u003cp\u003eSeventh, in the no-anticoagulant group, 50 patients (5.5%) received a DOAC during the trial. In addition, temporary DOAC therapy was permitted before and after cardioversion or catheter ablation. This crossover could reduce the apparent difference between the two groups.\u003c\/p\u003e\n\n\u003ch2 id=\"recommendations\"\u003ePractical Points for Patients and Families\u003c\/h2\u003e\n\n\u003cp\u003eIf you have atrial fibrillation, your doctor uses a scoring system to estimate your stroke risk. The score is called the CHA\u003csub\u003e2\u003c\/sub\u003eDS\u003csub\u003e2\u003c\/sub\u003e-VASc score. Ask your doctor what your score is and what it means for you.\u003c\/p\u003e\n\n\u003cp\u003eThis trial studied patients at intermediate risk. Intermediate risk was defined as a score of 1 in men and 2 in women. If your score is in this range, this trial provides randomized evidence that may be relevant to your care.\u003c\/p\u003e\n\n\u003cp\u003eThe trial tested two specific DOACs. Patients received either apixaban at 5 mg twice daily or rivaroxaban at 20 mg once daily. Some patients received reduced doses based on prespecified criteria. If you are prescribed a DOAC, your dose should be based on your own kidney function and other clinical factors.\u003c\/p\u003e\n\n\u003cp\u003eKey points to discuss with your doctor include the following:\u003c\/p\u003e\n\n\u003col\u003e\n\u003cli\u003eWhat is my CHA\u003csub\u003e2\u003c\/sub\u003eDS\u003csub\u003e2\u003c\/sub\u003e-VASc score, and am I in the intermediate-risk group studied in this trial?\u003c\/li\u003e\n\u003cli\u003eWhat is my risk of bleeding, and how does my HAS-BLED score factor into the decision?\u003c\/li\u003e\n\u003cli\u003eWhich specific anticoagulant is best for me, and at what dose?\u003c\/li\u003e\n\u003cli\u003eWhat signs of bleeding should I watch for, and when should I seek medical attention?\u003c\/li\u003e\n\u003cli\u003eDo I need antiplatelet therapy as well, or should I avoid it unless there is a specific reason?\u003c\/li\u003e\n\u003c\/ol\u003e\n\n\u003cp\u003eDo not stop or start any anticoagulant on your own. The decision depends on your individual stroke risk, bleeding risk, kidney function, other medications, and personal circumstances.\u003c\/p\u003e\n\n\u003cp\u003eReport any unusual bleeding, such as nosebleeds that will not stop, blood in the urine or stool, or unexplained bruising, to your doctor promptly. Also report any signs of stroke, such as sudden weakness on one side, difficulty speaking, or sudden vision loss, by seeking emergency care immediately.\u003c\/p\u003e\n\n\u003c!-- ddn:faq:start --\u003e\n\u003ch2 id=\"ddn-faq\"\u003eFrequently Asked Questions\u003c\/h2\u003e\n\u003ch3\u003eWhat is the SINGLE-AF trial and what did it find?\u003c\/h3\u003e\n\u003cp\u003eIt was a randomized trial at 18 centers in South Korea. It enrolled 1803 patients with atrial fibrillation at intermediate stroke risk. Patients took a direct oral anticoagulant or no anticoagulant. After 24 months, the combined endpoint of stroke, systemic embolism, major bleeding, or cardiovascular death occurred in 0.5% of the DOAC group versus 1.5% of the no-anticoagulant group.\u003c\/p\u003e\n\u003ch3\u003eWho was eligible for this trial?\u003c\/h3\u003e\n\u003cp\u003ePatients were eligible if they were at least 19 years old, had atrial fibrillation, and were at intermediate stroke risk. Intermediate risk meant a CHA2DS2-VASc score of 1 in men or 2 in women. The trial excluded people with significant kidney or liver disease, certain heart valve problems, or significant structural heart disease.\u003c\/p\u003e\n\u003ch3\u003eDid the trial show a benefit for stroke alone?\u003c\/h3\u003e\n\u003cp\u003eStroke occurred in 0.3% of the DOAC group and 1.1% of the no-anticoagulant group, with a hazard ratio of 0.30. However, the 95% confidence interval ranged from 0.08 to 1.08 and crossed 1.0, so the stroke result alone did not reach conventional statistical significance. The combined primary endpoint did reach significance.\u003c\/p\u003e\n\u003ch3\u003eWhat were the bleeding risks in this trial?\u003c\/h3\u003e\n\u003cp\u003eMajor bleeding occurred in 0.3% of the DOAC group and 0.5% of the no-anticoagulant group, with a hazard ratio of 0.75. Clinically relevant nonmajor bleeding was numerically more common with DOACs, at 2.5% versus 1.6%, but the confidence interval crossed 1.0. Serious adverse events were similar between groups, at 8.9% and 9.3%.\u003c\/p\u003e\n\u003ch3\u003eWhat are the limitations of this trial?\u003c\/h3\u003e\n\u003cp\u003eThe trial was open-label, so patients and doctors knew the assigned treatment. It enrolled patients in South Korea, so results may not apply exactly to other populations. Only 4 events occurred in the DOAC group and 13 in the no-anticoagulant group, making estimates less precise. Secondary and subgroup results were not adjusted for multiple testing.\u003c\/p\u003e\n\u003ch3\u003eWhat should I discuss with my doctor about anticoagulants?\u003c\/h3\u003e\n\u003cp\u003eAsk your doctor what your CHA2DS2-VASc score is and whether you are in the intermediate-risk group studied. Discuss your bleeding risk and HAS-BLED score, which anticoagulant and dose may suit you, what bleeding signs to watch for, and whether you need antiplatelet therapy. Do not start or stop any anticoagulant on your own.\u003c\/p\u003e\n\u003ch3\u003eI have atrial fibrillation with an intermediate stroke risk score — when should I get a second opinion about whether to start a blood thinner?\u003c\/h3\u003e\n\u003cp\u003eA second opinion is reasonable when your CHA2DS2-VASc score is 1 (men) or 2 (women), the intermediate-risk group studied here, because guidelines give only a class IIa recommendation for considering oral anticoagulants. In this trial, DOAC therapy lowered the combined endpoint of stroke, systemic embolism, major bleeding, or cardiovascular death at 24 months, from 1.5% to 0.5%, though the stroke result alone did not reach statistical significance and bleeding risk must be weighed. An independent review can help you and your doctor apply this evidence to your own situation. Diagnostic Detectives Network provides independent expert second opinions.\u003c\/p\u003e\n\u003c!-- ddn:faq:end --\u003e\n\n\u003ch2 id=\"source\"\u003eSource Information\u003c\/h2\u003e\n\n\u003cp\u003e\u003cstrong\u003eOriginal article title:\u003c\/strong\u003e Anticoagulation for Atrial Fibrillation with Intermediate Stroke Risk\u003c\/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eAuthors:\u003c\/strong\u003e Daehoon Kim, M.D.; Young Soo Lee, M.D.; Jaemin Shim, M.D.; Hee Tae Yu, M.D.; Junbeom Park, M.D.; Jin-Kyu Park, M.D.; Il-Young Oh, M.D.; Ki-Woon Kang, M.D.; Eue-Keun Choi, M.D.; Kyoung-Min Park, M.D.; Hyoung-Seob Park, M.D.; Dae-Hyeok Kim, M.D.; Hyung Wook Park, M.D.; Jun Kim, M.D.; Jum-Suk Ko, M.D.; Dongmin Kim, M.D.; Jong-Youn Kim, M.D.; Jin-Bae Kim, M.D.; Jung-Hoon Sung, M.D.; Tae-Hoon Kim, M.D.; Jae-Sun Uhm, M.D.; Hui-Nam Pak, M.D.; and Boyoung Joung, M.D., for the SINGLE-AF Investigators\u003c\/p\u003e\n\n\u003cp\u003e\u003cstrong\u003ePublication:\u003c\/strong\u003e The New England Journal of Medicine. Published online August 28, 2026. DOI: 10.1056\/NEJMoa2607978. Copyright © 2026 Massachusetts Medical Society.\u003c\/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eTrial registration:\u003c\/strong\u003e SINGLE-AF, ClinicalTrials.gov number NCT04437654.\u003c\/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eFunding:\u003c\/strong\u003e The Ministry of Health and Welfare, South Korea, and others.\u003c\/p\u003e\n\n\u003cp\u003e\u003cem\u003eThis patient-friendly article is based on peer-reviewed research.\u003c\/em\u003e\u003c\/p\u003e","brand":"DiagnosticDetectives.Com","offers":[{"title":"Default Title","offer_id":47807398740124,"sku":null,"price":0.0,"currency_code":"USD","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0599\/5449\/5644\/files\/ddn-medical-article-blood-thinners-for-atrial-fibrillation-in-patients-at-intermediate-stroke-risk-what-a-major-randomized-trial-found-hero.png?v=1792909278","url":"https:\/\/diagnosticdetectives.com\/de\/products\/blood-thinners-for-atrial-fibrillation-in-patients-at-intermediate-stroke-risk-what-a-major-randomized-trial-found","provider":"DiagnosticDetectives.Com","version":"1.0","type":"link"}