Anticoagulation usage and thrombolytic therapy in subjects with atrial fibrillation associated ischaemic stroke

Background: Atrial Fibrillation (AF) causes up to 20% of ischaemic strokes and 30% in some populations such as those over 80 years. Previous research in our population showed that widespread adoption of Direct Oral Anticoagulant (DOAC) prophylaxis had not been associated with a reduction in AF associated stroke prevalence but there was a considerable rate of breakthrough stroke in patients receiving anticoagulation and anticoagulation rate may be affected by increasing use of DOACs.

Aims: We undertook a more detailed study using the Irish National Audit of Stroke (INAS) to determine the characteristics of anticoagulation practice in AF associated stroke, particularly breakthrough stroke, adherence to prescribing guidelines and effect on thrombolysis rate.

Methods: Data from INASwere analysed for the period 2017-2022 inclusive as part of a cross-sectional, cohort study. Data on pre stroke, and discharge disability (modified Rankin Score (mRS)) were collected in addition to age, sex, length of stay and thrombolysis rate. An enhanced dataset with additional questions about adherence with prescription, reasons for non-concordance and International Normalized Ratio (INR) control for receiving Warfarin was collected for 2022 was also considered separately. Comparisons for continuous / quantitative data were made using Student's t tests and for proportional data using Pearson's Chi Square statistics and logistic regression analyses

Results: Complete AF data were available on 22,485 of 26,829 strokes admitted over this period. Of these strokes with AF data, 19,260 (85.6%) were ischemic, mean age was 71.8 years and 57.1% male. AF was found in 5321 of these ischemic strokes, and this AF was identified pre-stroke in 2835 (53.3%). In total, 80.4% of patients with known AF had been prescribed anticoagulation. The population with previously unknown AF was significantly younger on average than those on anticoagulation (76.8 years vs. 79.1 years (p < 0.0001, t-test)), or those with AF and not anticoagulated (78.8 years (p < 0.0001)); they were also much more likely to have received thrombolysis (17.3% vs. 4.0% (χ2, p < 0.0001)). Of 1794 patients receiving DOACs, only 52 received thrombolysis (2.9%) compared with 37 (8.5%) of the Vitamin K antagonist (VKA) group (χ2 = 29.0, p < 0.0001). On regression analysis, anticoagulation was not associated with differences in excellent outcome (modified Rankin Score (mRS) 0 or 1) achieved (odds ratio (OR) = 1.064, p = 0.41) or with mortality (OR = 1.014, p = 0.89). There were 4999 strokes in 2022, 4272 (85.4%) were ischemic, and 1270 (29.7%) of ischemic strokes were AF associated. Of the 557 AF-associated ischemic strokes anticoagulated at presentation, 84.6% were prescribed DOACs. Forty-eight (9.5%) had their anticoagulation paused, and 40 admitted to poor compliance (7.9%).

Conclusion: Nearly half of people with AF identified had not previously had the arrhythmia detected pre-stroke. Those admitted with known AF were mainly appropriately treated with DOACs and constitute breakthrough strokes. Subjects receiving DOACs were much less likely to receive thrombolytic therapy even than those taking Warfarin.

Data access: INAS data is publicly accessible. Data is available on application to National Office of Clinical Audit, Ireland.