Frequency of bleeding and thromboembolism in patients with atrial fibrillation on warfarin at a tertiary care hospital in Sri Lanka

Atrial fibrillation (AF), the most common sustained cardiac arrhythmia, significantly increases stroke risk with approximately one in five strokes attri-buted to AF1,2,3. Oral anticoagulation, primarily with warfarin, is crucial for AF management, especially in valvular AF where it remains the drug of choice. However, warfarin’s narrow therapeutic range poses a significant bleeding risk, with intracranial haemorrhage being the most serious complication. Time in Therapeutic Range (TTR) is a key metric for assessing anticoagulation efficacy and safety, with TTR values above 70% correlating with better out-comes. This is a retrospective descriptive cross-sectional study aimed to determine the frequency of bleeding and thromboembolism in AF patients on warfarin over six months at the National Hospital of Sri Lanka (NHSL), describe their demo-graphic profile, determine predicted bleeding risk using the HAS-BLED score, and evaluate anti-coagulation effectiveness via TTR. The cohort had a mean age of 62.41 years, with a female predo-minance (73%). Seventeen percent of patients experienced at least one bleeding episode during follow-up, with 4% being major bleeds. All major haemorrhages occurred with supratherapeutic INR values (INR >5), although no significant correlation was found between bleeding type and INR values (p=0.187). No significant correlation between HAS-BLED score and bleeding (p=0.156), but 3 out of 4 major bleeds were of high-risk HAS-BLED category. Twenty-six percent had stroke/ transient ischemic attack (TIA) during follow-up; only 1 had a stroke in the past 3 months who had a subtherapeutic INR at the time. There was no correlation between stroke/ TIA/ thromboembolism and CHA2DS2VASc score (p=0.36). Only 45.435% of the patients had INR values within the therapeutic range and 62%of patients had a TTR below 60%. There was no significant association between TTR and either bleeding or thrombotic episodes. The study concluded that warfarin can be safely used in this setting for AF patients, but the observed lack of predictive power for HAS-BLED and CHA2DS2VASc scores, alongside suboptimal TTR, highlights the need for further larger-scale, multicenter studies.