90 Thromboembolic events and vascular dementia in contemporary patients with atrial fibrillation and low apparent stroke risk

Background Thromboembolism in patients with atrial fibrillation (AF) can be prevented, however oral anticoagulation is typically reserved for older patients or those with specific historical comorbidities. Risk scores are in widespread use, but with limited predictive accuracy for stroke and no consideration of current major challenges such as dementia. This study provides contemporary data on the risk of thromboembolism, including both cardiac and cerebral damage.

Methods Population-based matched cohort study of primary care patients across the UK from 2005 to 2020 using structured, electronic healthcare record data. Inclusion criteria were patients with a pre-defined coded entry for AF (exposed group; including past or resolved AF), aged between 40-75, no previous history of stroke, CHA2DS2-VASc risk score zero or one, and not receiving an oral anticoagulant. Patients were matched by age, sex and region with up to 4 patients without an AF exposure diagnosis. Outcomes were all-cause mortality, stroke, arterial thromboembolism, ischaemic heart disease and dementia (categorised into all-cause, Alzheimer’s and vascular dementia), analysed by Cox proportional hazard ratios (HR) adjusted for age, sex, socioeconomic deprivation, ethnicity and clinical factors.

Results Real-world healthcare data from 828 general practices were included (a total of 16,587,749 patients), of which 290,525 patients from 5,199,994 (5.6%) had evidence of AF exposure and were aged 40-75 years. 36,340 patients with prior AF exposure met inclusion criteria (no prior stroke, CHA2DS2-VASc zero or one, no anticoagulation), and were matched to 117,298 controls without documented AF; Figure 1. Demographics were comparable across groups with a median age of 58 years (interquartile range [IQR] 12 years), 17.4% women, 2.3% with diabetes mellitus and 12.4% with hypertension; consistent with a population at low thromboembolic risk (mean CHA2DS2-VASc score 0.5, SD 0.5). The median follow-up was 4.4 years for those with AF exposure (IQR 1.9-8.0) and 5.0 years for controls (IQR 2.2-8.7), with a combined total of 831,005 person-years of follow-up. The rates of cardiovascular outcomes were consistently two-fold higher in those with AF exposure compared to control; for stroke 3.8% vs 1.5% (adjusted HR 2.06, 95% CI 1.91-2.21; p<0.001), arterial thromboembolism 0.3% vs 0.1% (HR 2.39, 95% CI 1.83-3.11; p<0.001) and ischaemic heart disease 5.6% vs 2.7% (HR 1.88, 95% CI 1.77-1.99; p<0.001). All-cause mortality was also significantly higher in patients with AF exposure (HR 1.44 vs control, 95% CI 1.38-1.50; p<0.001). AF exposure was associated with all-cause dementia (HR 1.17, 95% CI 1.04-1.32; p=0.010), driven predominantly by vascular dementia (HR 1.68, 95% CI 1.33-2.12; p<0.001) as opposed to Alzheimer’s disease (HR 0.85, 95% CI 0.70-1.03; p=0.09); Figure 2.

Conclusions Large-scale routine healthcare data demonstrates that thromboembolic risk, including the propensity to vascular dementia, is substantially increased in contemporary patients with any exposure to AF. This includes those who are younger or without conventional stroke risk factors.