Based on the "Chinese Guidelines for the Prevention of Cardiogenic Stroke (2019)," the CHA2DS2-VASc scoring system is recommended for atrial fibrillation patients, suggesting oral anticoagulant therapy for males with a score ≥ 2 and females with a score ≥ 3 (19). This study included 410 patients, with 42.7% receiving oral anticoagulants during acute hospitalization, slightly higher than the global registry study—GARFIELD (20)—which reported a 28% anticoagulation rate among Chinese patients with a CHA2DS2-VASc score ≥ 2. However, our study focused on acute cerebral infarction patients with a higher average age, indicating a higher risk population than the aforementioned studies, and suggesting an issue of insufficient anticoagulation. Oral anticoagulant therapy is the cornerstone for primary and secondary prevention of ischemic stroke in patients with atrial fibrillation. Novel oral anticoagulants (NOACs) have been shown in previous research to be just as good as warfarin at preventing stroke. They have the same or a lower risk of major bleeding events and a much lower risk of intracranial hemorrhage (21–25). The RAF study demonstrated a better prognosis for patients with nonvalvular atrial fibrillation (NVAF) and AIS treated solely with oral anticoagulants compared to those treated with low molecular weight heparins (LMWH) alone or LMWH followed by oral anticoagulants. However, the optimal timing for initiating anticoagulation after acute ischemic stroke remains uncertain (26), as key large-scale studies comparing NOACs with warfarin excluded patients who had recently (within 7–30 days) experienced a stroke (21, 23–24). Due to a lack of evidence, current international guidelines do not provide specific recommendations on the optimal timing for initiating anticoagulation therapy after an acute cerebral stroke. The latest research (TIMING, ELAN) suggests that early anticoagulation is more effective than delayed anticoagulation (27, 28).
The risk of ischemic stroke recurrence or post-stroke hemorrhagic transformation is highest within the first few days after an ischemic stroke (12). Such events may offset the advantages of acute secondary prevention. Previous studies indicate that patients receiving IVT, EVT, or IVT + EVT treatments might have higher symptomatic hemorrhagic transformation rates than control groups, but better 90-day outcomes [mRS score 0–2] (29, 30). In this study, the early anticoagulation group (< 14 days) had a 90-day good prognosis rate of 86.9%, compared to 54.9% in the late anticoagulation group, with 90-day ischemic stroke recurrence and bleeding event rates of 4.0% and 6.9%, respectively. Referring to the TIMING trial for anticoagulation timing groups, the early anticoagulation initiation group (≤ 4 days) included 60 people, with a 90-day good prognosis rate of 93.3%, while the delayed or non-initiation group (> 4 days) with 350 people had a 90-day good prognosis rate of 64.3%. There were no new ischemic strokes in the early anticoagulation group within 90 days, but there were 5 new bleeding events, compared to 25 new ischemic strokes and 22 new bleeding events in the delayed group. The early (< 4 days) anticoagulation group had a higher proportion of a good prognosis at 90 days and fewer new ischemic strokes and bleeding events than the delayed group, aligning with the latest findings from the TIMING and ELAN studies. However, it should be noted that only 60 patients in our study initiated anticoagulation treatment early (≤ 4 days). The average age of the people in our study was 72.66 years, and their NIHSS score was 13. This is higher than the range of 2–10 in previous observational studies. Also, 17.3% of the people in our study had severe strokes (NIHSS > 25), which is much higher than in previous studies (5, 18, 31, 32). Patients with early severe symptomatic bleeding or a high risk of bleeding did not meet the criteria for initiating anticoagulation therapy.
The median time to initiate oral anticoagulant therapy during hospitalization in our study population was 7 days, higher than the currently recommended early initiation time (28, 33). Only 42.7% of patients underwent anticoagulation therapy within 14 days of AIS occurrence, with 41.0% in the IVT group. However, the Irene (18) trial reported an 82% anticoagulation rate within 14 days, with 85.1% receiving IVT treatment and a median NIHSS score of 10 (6–16) at admission, compared to a median NIHSS score of 13 (8–20) in our study, with 10 in the IVT group and 15 in the EVT and IVT + EVT groups. For patients with atrial fibrillation undergoing EVT treatment, initiating anticoagulation between 5–14 days resulted in the lowest recurrence rate of ischemic cerebrovascular events (34). Thus, a higher NIHSS score, more severe stroke, and endovascular intervention may significantly impact the timing of anticoagulation initiation and outcome events (32, 35, 36), including increasing the risk of bleeding. Previous studies have shown that the ICH (intracranial hemorrhage) incidence rate in IVT patients is 2.7–5.1% and even lower in EVT patients (37–39). In our study, there were 116 cases of ICH, with a 7.6% incidence rate of large parenchymal hematoma (PH2). Moreover, our findings suggest that pre-anticoagulation antiplatelet drug use is a protective factor for early initiation of anticoagulation treatment (OR = 0.122, 95% CI: 0.065–0.228, P < 0.001), which is inconsistent with previous experimental results (40). This suggests that physicians' use of antiplatelet drugs before anticoagulation typically indicates no bleeding events in these patients. However, in the GARFIELD-AF study, the use of NOACs in conjunction with antiplatelet drugs, regardless of the sequence, increased the risk of stroke or bleeding in patients with atrial fibrillation (40).
Currently, there are few studies on the status of anticoagulation therapy in patients with NVAF and AIS undergoing vascular recanalization therapy. In the RAF and TIMING studies, only 22.4% and 35.2% of patients, respectively, underwent vascular recanalization therapy. Compared to patients who did not receive such therapy, these patients likely had more severe strokes, with higher mRS and NIHSS scores at admission. Pathologically, early in a stroke, particularly during ischemic reperfusion, the risk of hemorrhagic transformation increases due to impaired autoregulation and blood-brain barrier disruption; thus, early anticoagulation might elevate the risk of ICH (41). Additionally, for patients receiving EVT, initiating anticoagulation within 4 days did not significantly improve functional outcomes (34), potentially due to the small sample size of 60 patients initiating anticoagulation within this period in our study. The high proportion of good outcomes at 90 days in this group compared to the delayed anticoagulation group may indicate selection bias. Therefore, it's essential to further explore a balanced timing or period for anticoagulation therapy in such patients.
The strength of this study is that it is a multicenter, real-world study that provides new evidence regarding the anticoagulation strategy after reperfusion therapy for patients with AF-related AIS.
However, the study also has certain limitations. First, patient outcomes were assessed by telephone follow-ups, and due to the limited-literacy, some patients may not have been able to accurately report discontinued antiplatelet drugs or switch to anticoagulants. This could introduce some information bias into the study. Second, this is a retrospective study, with a larger proportion of patients receiving EVT, who are at a higher risk of bleeding. There is a selection bias in anticoagulation therapy, which limits the generalizability of the study results to other populations.