Abstract Aims To evaluate the safety and efficacy of left atrial appendage occlusion (LAAO) with the Amplatzer™ Amulet™ occluder. Methods and results Patients with atrial fibrillation eligible for LAAO were recruited to a prospective global study. Implant procedures were undertaken with echocardiographic guidance. Transoesophageal echocardiography (TOE) was undertaken 1–3 months post-LAAO. Implant and follow-up TOEs were evaluated by a CoreLab. The primary endpoint was a composite of ischaemic stroke and cardiovascular death at 2 years. Serious adverse events were adjudicated by an independent clinical events committee. A total of 1088 patients were enrolled, aged 75.2 ± 8.5 years; 64.5% were male. CHA2DS2-VASc and HAS-BLED scores were 4.2 ± 1.6 and 3.3 ± 1.1, respectively. A total of 71.7% had prior major bleeding, and 82.8% had contraindications to oral anticoagulants. Implant success was 99.1%. Major adverse events (≤7 days post-procedure) occurred in 4.0%, including death (0.3%), stroke (0.4%), major vascular (1.3%), and device embolization (0.2%). A total of 80.2% of patients were discharged on antiplatelet therapy alone. Peridevice flow was <3 mm in 98.4% at follow-up TOE. Device-related thrombus (DRT) was seen in 1.6% of cases. Cardiovascular death or ischaemic stroke occurred in 8.7% of patients at 2 years. The ischaemic stroke rate was 2.2%/year—a 67% reduction compared to the CHA2DS2-VASc predicted rate. Major bleeding (Bleeding Academic Research Consortium type ≥ 3) occurred at rates of 10.1%/year (year 1) and 4.0%/year (year 2). Conclusion Following LAAO with the Amplatzer Amulet device, the ischaemic stroke rate was reduced by 67% compared to the predicted risk. Closure was complete in 98.4% of cases and DRT seen in only 1.6%.