Increased risk of stroke in young (aged 55 to 64) screen-detected and clinically diagnosed atrial fibrillation patients who do not meet criteria for oral anticoagulation therapy at time of diagnosis:

Purpose Opportunistic screening for AF is recommended in population aged above 65. However, risk of ischemic stroke is higher in Asians. We aimed to determine longitudinal change of CHA2DS2VA, anticoagulation therapy (OAC) and risk of ischemic stroke in a Chinese 55-64yr population undergoing AF screening. Methods Consecutive patients aged 55-64 years attending medical outpatient clinics in Hong Kong underwent AF screening using a handheld single-lead ECG (AliveCor) from Dec 2014 to Dec 2017 (NCT02409654). Repeated screening was performed in patients who had >1 clinic visit during the study period. OAC indication (i.e., CHA2DS2VA >=2) and OAC use were handled as time-dependent variables. Cumulative incidence of OAC indication, OAC prescription were estimated and compared between screen-detected AF and clinically-diagnosed AF during follow-up, using Fine and Gray regression accounting for death as competing risk. Response to OAC was evaluated by comparing risk of ischemic stroke, stratifying all AF categories by OAC use and using no AF group as reference. Results Of 3926 subjects enrolled, 341 (8.7%) had previously known AF, 75 (1.9%) clinically-diagnosed AF during FU, and 35 (0.9%) screen-detected AF. Over a median FU duration of 8.49 (IQR: 7.9-9.6) years, proportion of AF patients indicated to OAC therapy increased from 16.0% to 50.2% (Table 1). Compared to no AF, no statistically significant difference was observed between groups (subdistribution hazard ratio (SHR) for AF not on OAC =2.48, 95% CI 0.6-11.0, p=0.23, SHR for AF on OAC=1.72, 95% CI 0.5-5.8, p=0.38) (Figure 1A). While 43.3% of strokes happened before an OAC indication was reached, the subsequent stroke rate was 7.7% for patients with screen-detected AF, compared to just 1.0% for those who screened negative (p<0.001). Reaching indication of OAC further increased the risk of stroke (aHR=1.73,95% CI 1.0-3.0, p=0.05). Prior stroke history was the sole predictors of subsequent stroke (HR=3.53, p<0.001). OAC prescription generally increased over time regardless of categories of AF (clinically-diagnosed AF or screen-detected AF), but plateaued earlier in screen-detected AF (2.5 years after AF diagnosis for screened-detected AF vs 5.0 years for clinically-diagnosed AF) (Figure 1B). Proportions of OAC use for indications other than AF (i.e., deep vein thrombosis, pulmonary embolization) among screen-detected and clinically-diagnosed AF were 5.7% and 8.0%, respectively. Conclusion Approximately two-fifth of strokes occurred before reaching OAC indication in this younger population. OAC prescription plateaued earlier among patients with screen-detected AF, implying a gap in downstream management after screening, warranting further study.