Anticoagulants or antiplatelets for secondary prevention of cryptogenic stroke: an updated systematic review and meta-analysis

Background Patients with cryptogenic stroke or embolic stroke of undetermined source (ESUS) face a high risk of recurrent ischaemic stroke, but the optimal antithrombotic strategy remains unclear. This systematic review and meta-analysis compared the effectiveness and safety of oral anticoagulants (OACs) versus antiplatelets in these populations, with a focus on subgroup effects by key clinical characteristics. Methods Six databases were searched through March 2024 to identify randomised controlled trials (RCTs) comparing OACs and antiplatelets in patients with cryptogenic stroke or ESUS. The primary outcome was recurrent ischaemic stroke. Subgroup analyses evaluated treatment effects based on supracardiac atherosclerosis risk, presence of patent foramen ovale (PFO) and signs or risk factors for atrial cardiopathy. Meta-regression with interaction p values was employed to assess differences in treatment effects between subgroups. Results Nine RCTs comprising 15 451 participants were included. In the overall population, there was no significant difference in recurrent ischaemic stroke risk between OACs and antiplatelets (relative risk (RR) 0.90, 95% CI 0.79 to 1.02; I 2 =0%). Subgroup analyses showed that OACs reduced ischaemic stroke risk in patients with low-risk supracardiac atherosclerosis (RR 0.53, 95% CI 0.35 to 0.80; I 2 =0%) compared with those with high-risk supracardiac atherosclerosis (RR 0.91, 95% CI 0.78 to 1.06; I 2 =0%) and evidence of supracardiac atherosclerosis (RR 1.13, 95% CI 0.84 to 1.53; I 2 =0%) (p interaction=0.0002). Similarly, OACs were more effective in patients with signs or risk factors for atrial cardiopathy (RR 0.84, 95% CI 0.70 to 0.99; I 2 =0%) than in those without atrial cardiopathy (RR 1.05, 95% CI 0.85 to 1.30; I 2 =0%) (p interaction=0.02). There was no significant interaction by PFO status (p interaction=0.28). While the risk of major bleeding risk was comparable between groups (RR 1.34, 95% CI 0.73 to 2.44; I 2 =65%), a significantly higher risk of major bleeding other than intracerebral haemorrhage was observed in patients taking OACs compared with antiplatelets (RR 1.69, 95% CI 1.18 to 2.43; I 2 =0%). Conclusions OACs are more effective than antiplatelets for preventing ischaemic stroke in patients who had a cryptogenic stroke or ESUS with low-risk supracardiac atherosclerosis or atrial cardiopathy. The findings highlight the need for personalised treatment strategies and further trials in these subgroups. PROSPERO registration number CRD42024518903.