Abstract Aims Incidence of stroke in patients with an advanced stage hypertrophic cardiomyopathy (HCM) is associated with adverse outcome, impaired quality of life and loss of productivity. Still today, however, the real burden of stroke in both patients with and without atrial fibrillation (AF) is unresolved. To assess the prevalence and incidence of AF and stroke in patients with an advanced stage HCM implanted with cardiac implantable electronic devices (CIEDs) at our institution, a long-standing high flow referral centre for cardiomyopathies. Methods and results Clinical and instrumental data of HCM patients implanted with CIEDs [either pacemakers (PM) or implantable cardioverter defibrillator (ICD)] from 1998 to 2019 were retrospectively reviewed. Inclusion criteria were site-designated diagnosis of HCM, age at diagnosis >18 years, Follow-up >1 year. HCM phenocopies (e.g. Fabry disease) were carefully excluded. Patients were divided into three categories according to presence of AF (‘AF prior to CIED implantation’ vs. ‘AF after CIED implantation’ vs. ‘sinus rhythm’). Outcome was measured against incidence of thromboembolic events [stroke or transient ischaemic attack (TIA)] at Follow-up. Patients were also stratified by left atrial diameter (LAD) in two groups (<47 vs. > 48mm). Of 1861 patients followed at our Unit, a total of 185 (9.9%) patients implanted with a CIED were included (57% men, mean age at implantation 54 ± 17 years). At baseline, AF was present in 72 (36%) patients. Mean CHA2DS2VASc was 1.7 + 1.3 with no differences among patients with or without AF. Patients with AF at baseline had a more pronounced LAD dilation (51 ± 7 vs. 44 ± 8, P < 0.001) and a lower ejection fraction (55 ± 11 vs. 64 ± 12, P < 0.001). After 5.0 ± 3.8 years from CIED implantation, de novo AF was detected in 24 (21%) individuals, resulting in an annual incidence rate of 4.1%/year. Overall, 89 (48%) of patients remained is sinus rhythm. Stroke/TIAs were reported in 19 (10.3%) patients: seven (37%, 1.1%/year) occurred in patients with prior history of AF, three (16%, 2.2%/year) in patients with de novo AF, and nine (63%, 2.3%/year) in patients with no history of arrhythmias documented at CIED interrogation. Among patients in sinus rhythm, those with a LAD > 48 mm had the greatest risk of stroke (4.8%/year vs. 0.5%/year, P < 0.01, for LAD > 48 vs. LAD < 47, respectively). At multivariable analysis, after adjustment for CHA2DS2VASc, AF, and obstructive physiology, only LAD was associated with a higher risk for stroke (HR: 1.09, 95% CI: 1.03–1.11, P < 0.001). Conclusions In a large cohort of consecutive high risk HCM patients referred to CIED implantation, the incidence of stroke was high, with 1-in-10 patients experiencing at least one event. Among patients in sinus rhythm, those with a marked left atrial dilatation were at highest risk of ischaemic stroke, suggesting the existence of an unmet need to stratify risk of stroke even in patients with no detected arrhythmias.