When” and “how” should antithrombotic therapy be administered in elderly patients with atrial fibrillation, especially in the elderly? Antithrombotic strategies should be determined according to the urgency of the clinical event, taking into account the prognostic impact of antithrombotic therapy and the search for safer and more effective anticoagulants. In 2011, Lip et al. compared the value of various stroke scoring systems in the anticoagulation of elderly patients with atrial fibrillation in a group of 662 patients with atrial fibrillation, mean age 74 years, 36.1% of whom were women, at a mean follow-up of 3.6 years, to observe stroke and thrombotic events. The results showed that all stroke scoring systems had moderate predictive value, but the CHADS2 CHA2DS2-VASc had the highest predictive value for thrombosis. Therefore, CHA2DS2-VASc is better than CHADS2 to differentiate between low-risk and intermediate-risk patients and truly identify low-risk patients, and Liu Hongliang, Department of Geriatrics, The First Affiliated Hospital of Henan College of Traditional Chinese Medicine. This part of patients does not need antithrombotic therapy. Hemorrhage risk assessment HAS-BLED hemorrhage score clinical application affirms that major bleeding increases the risk of death In 2010, R. De Caterina et al. analyzed data from the ACTIVE-W survey in which enrolled patients with atrial fibrillation were randomized to the vitamin K inhibitor anticoagulation group (3371 cases) and the clopidogrel + aspirin combination antithrombotic therapy group (3335 cases). The results showed that both ischemic stroke and hemorrhagic stroke increased mortality, but the risk of hemorrhagic stroke was higher than that of ischemic stroke; transient ischemic attack (TIA) did not increase mortality. Only major bleeding events (68% of all major bleeding events) increased mortality. The difference between non-serious events with major bleeding and serious events with major bleeding was that non-serious major bleeding events required transfusion of ≥ 2 U but < 4 U. Patients treated with clopidogrel combined with aspirin antithrombotic therapy had more minor bleeding and less major or serious bleeding compared with vitamin K inhibitor anticoagulation. Reduced hemoglobin after hemorrhage leads to decreased tissue oxygenation, hemodynamic instability, sympathetic nervous system activation, vasoconstriction, and increased cardiac load all contribute to a poor prognosis. The immediate prognosis for the occurrence of bleeding events is primarily related to the site of bleeding, and the subsequent frequent interruption of antithrombotic therapy increases the risk of thrombosis. Be alert to the overlap of bleeding and thrombosis Patients at high risk for bleeding events are usually older and have more comorbidities. Studies have shown that in older adults, the risk of bleeding increases twofold. The relative risk of intracranial hemorrhage (especially at high INR values) is 2.5 in elderly patients ≥ 85 years of age compared to those aged 70-74 years. The AFFIRM study showed that for every 1 year increase in the incidence of major bleeding, there was a 5% increase in risk. In a group of 105 elderly patients (mean age 85 years) from our center, 21% had previous cerebral hemorrhage, hemorrhagic gastritis, duodenal ulcer, renal anemia, and thrombocytopenia; 26.7% had previous cerebral infarction, inadequate basilar artery supply, atrial thrombosis, lower extremity deep vein thrombosis, and peripheral vascular embolism; and 8.6% had both bleeding and thrombotic risk disorders. Other risk factors include liver or kidney disease, alcohol dependence, tumors, decreased platelets, risk of rebleeding, uncontrolled hypertension, anemia, genetics, fall risk, previous stroke or TIA. This scoring system is complex and difficult to master for clinical application. Following this, Shireman et al. developed a bleeding score system including 8 factors in 26,345 elderly patients ≥65 years of age with AF, and found that this score was 0.9% for patients at low risk of bleeding, 2.0% for those at intermediate risk of bleeding, and 5.4% for those at high risk of bleeding. The equation was as follows: risk score = 0.49 (age ≥ 70 years) + 0.32 (female) + 0.58 (distant bleeding events) + 0.62× (recent bleeding events) + 0.71 (alcohol/drug abuse) + 0.27 (diabetes) + 0.86 (anemia) + 0.32× (antiplatelet therapy), with a score ≤ 1.07 as low risk, 1.07-2.19 as intermediate risk, and ≥ 2.19 as high risk. , ≥ 2.19 as high risk. The HAS-BLED recommended by the 2010 ESC Guidelines for the Management of Atrial Fibrillation is a well-used bleeding scoring system in recent years. Risk factors include hypertension, abnormal liver and kidney function, stroke, history or tendency to bleed, unstable INR, age >65 years, medication or alcohol abuse. In a study of 220 elderly patients with atrial fibrillation (mean age 83.1 years), the mean CHADS2 score was 3.0, the CHA2DS2-VASc score was 4.6, and the HAS-BLED score was 3.0. The risk of stroke on the CHADS2 and CHA2DS-VASc basis and the risk of bleeding on the HAS-BLED basis increased with increasing age. The increase in bleeding risk was more pronounced with increasing age, with the annual incidence of major bleeding risk increasing from 1.9% in the 65-79 year old group to 8.7% in the 90+ year old group. The risk of stroke was significantly higher than the risk of bleeding in both the 65-79 and 80-89 age groups, but the increased risk of bleeding (8.7%) in the high advanced age group (age >90 years) was close to the risk of stroke (9.8%) on a CHA2DS2-VASc basis. This was accompanied by a significant increase in the occurrence of bleeding events in the older age groups (3.1%, 14.6%, and 21.2% in the 65-79, 80-89, and 90+ age groups, respectively) and an increase in mortality from bleeding events. The increase in bleeding events was mainly due to major bleeding events. About one-third of the elderly AF patients who had a bleed had a thrombotic event after the bleed. In addition, 5.5% of elderly patients with atrial fibrillation had a “bidirectional event,” which is a thrombotic and bleeding event during the course of the disease. The average age of elderly patients with “bidirectional events” was 86.3 years, and most patients with “bidirectional events” will have a recurrent thrombotic event with a mortality rate of 22%. The BAFTA study of 973 elderly patients with a mean age of 81.5 years who were treated with warfarin (INR 2-3) or aspirin (75 mg/d) for antithrombotic prophylaxis was well documented. The annual incidence of stroke and embolic events was lower for warfarin than for aspirin (1.8% vs. 3.8%) and for extracranial hemorrhage (1.4% vs. 1.6%) in this group. The 2010 ESC Guidelines for the Management of Atrial Fibrillation recommend oral anticoagulation (OAC) therapy (Table 1). Patients aged ≥75 years with AF who have ≥1 risk factor (blood pressure, heart failure, diabetes mellitus, stroke, or TIA) should be treated with anticoagulation. However, patients with advanced AF should be adequately evaluated for risk of stroke and major bleeding. Dabigatran may be used as an alternative to warfarin anticoagulation in elderly patients with AF at high risk of bleeding, as it has the same antithrombotic effect as warfarin, but with a reduced risk of major bleeding such as intracranial hemorrhage. The guidelines recommend dabigatran (150 mg , bid) if the patient is at low risk for bleeding (HAS-BLED score 0-2) and dabigatran (110 mg , bid) if at high risk for bleeding (HAS-BLED score ≥ 3). The “Chinese Cardiovascular Disease Expert Consensus” recommends an INR safety window of 1.5 to 2.5 for Chinese older adults and an INR of 1.0 to 2.0, with a median of 1.8 and preferably no more than 1.8, for those at high risk for stroke and bleeding, most patients will benefit from OAC, but in patients at relatively low risk for stroke and high risk for bleeding, most patients will benefit from OAC. Patients with a HASBLED score ≥3 should be routinely monitored for bleeding risk after initial antithrombotic therapy, regardless of whether OAC or antiplatelet agents are used. In patients at high risk for thrombosis and bleeding, it is difficult to accurately identify the “gray area” between the occurrence of bleeding events and thrombotic events and to maximize the benefit of antithrombotic therapy. Antithrombotic treatment after a “bidirectional event” is also a challenge for clinicians, who need to carefully weigh the risk of bleeding and thrombosis, as well as the prognosis of different clinical events, taking into account the overall benefit of antithrombotic treatment on patient prognosis. Anticoagulation should be discontinued for at least 3 months in patients who have had a hemorrhagic stroke. However, in the case of a recurrent thrombotic event requiring emergency anticoagulation less than 3 months after a hemorrhagic stroke, OAC therapy can be resumed after “bridge anticoagulation” therapy (2 weeks of heparin therapy) while closely monitoring the INR to maintain it at the low end of the attainment limit. However, in elderly patients with atrial fibrillation, it is often difficult to achieve INR values after a “bidirectional event” and clinical signs of bleeding (e.g., oral mucosa, subcutaneous bleeding, urinary and fecal occult blood) need to be closely monitored to maximize the benefit of anticoagulation. In October 2010, the New England Journal of Medicine published the THINRS study to compare weekly self-measurement of INR at home with monthly INR anticoagulation monitoring in patients after atrial fibrillation or mechanical valve placement. The results showed that 80% of patients had the ability to self-monitor INR at home and adjust INR to target values. There were no significant differences between the two groups in the time to primary endpoints of stroke, major bleeding, and fatal events; nor were there significant differences in the secondary endpoints of non-stroke embolic events and myocardial infarction. In the self-monitoring group, the frequency of monitoring increased and the time to INR increased, but there was no significant reduction in the occurrence of major events and an increase in minor bleeding. Why did the increased time to INR not produce the expected better protection against thrombosis? It may be related to the narrow therapeutic window of warfarin. New oral anticoagulants developed in recent years include direct thrombin inhibitors and Xa factor inhibitors. The RE-LY study included 18 113 patients with non-valvular atrial fibrillation in the low-dose dabigatran (110 mg, bid), high-dose dabigatran (150 mg, bid), and warfarin groups. The results showed that the low-dose dabigatran group had lower bleeding than the warfarin group, but similar prevention of thrombotic events. 2011, Circulation published an observation of anticoagulation in 1270 patients with electrically resuscitated atrial fibrillation, which showed that for 30 d of electrically resuscitated atrial fibrillation, the incidence of stroke and major bleeding was lower or similar in both dabigatran-treated groups compared with the warfarin group. In contrast, in an analysis of 7258 patients ≥75 years with non-valvular atrial fibrillation, the incidence of hemorrhagic stroke was 0.20%, 0.14%, and 0.47%/year in the low-dose, high-dose dabigatran, and warfarin groups, respectively, and the preventive effect of dabigatran on stroke was independent of age and renal function. Currently, drugs in clinical studies, such as Rivaroxaban, Apixaban, Endoxaban, Betrixaban, etc., are expected to have more favorable results.