DEVELOPMENT AND VALIDATION OF A PREDICTION RULE FOR BENEFIT AND HARM OF ORAL ANTICOAGULATION IN NON-VALVULAR ATRIAL FIBRILLATION

Abstract

ABSTRACTBackgroundOral anticoagulation therapy (OAC) remains the gold standard for stroke prevention in patients with atrial fibrillation (AF). In real life, there are patients who do not receive OAC due to high bleeding risk. In those patients, left atrial appendage closure (LAAC) has emerged as a potential alternative for stroke prevention. With this study, we aimed to develop a clinical decision tool to identify patients expected to derive harm vs benefit from OAC therapy.MethodsAmong 14,915 AF patients with CHA2DS2-VASc ≥ 1 (2 for women) from CardioCHUVI-AF registry (78% with OAC), a prediction rule was derived using a linear regression model to predict the stroke-bleeding balance. This rule was externally validated in the Taiwan AF registry, with 26,595 patients (70.5% with non OAC therapy).ResultsA simplified risk score was created using 7 clinical variables. The low-score group (≤ −8 points) was associated with higher rates of bleeding than stroke (7.25 vs 1.11 and 3.27 vs 2.58 per 100 patients/year in derivation and validation cohorts, respectively; p<0.001). In those patients, OAC was harmful. In contrast, high-score group (≥ +6 points) was associated with higher stroke risk than bleeding risk (2.32 vs 1.71 and 4.19 vs 1.64 per 100 patients/year in derivation and validation cohorts, respectively; p<0.001), with great benefit of OAC therapy.ConclusionsA prediction rule balancing stroke and bleeding risks correctly identify patients with harm vs benefit from OAC therapy. This rule requires further prospective evaluation to assess potential effects on patient care to select candidates for LAAC.CLINICAL PERSPECTIVEWhat is new?We developed and externally validated a simple user-friendly clinical tool -OAC score- to balance both embolic and bleeding risks in atrial fibrillation patients with CHA2DS2-VASC ≥1 (excluding female sex).According to OAC score, we can identify atrial fibrillation patients expected to derive benefit vs harm from anticoagulation therapy.Patients with a low OAC score have much higher rates of bleeding than stroke, which is exaggerated with anticoagulation. The opposite occurs in patients with a high score.What are the clinical implications?A standardised stratification of both stroke and bleeding risk in atrial fibrillation patients at high embolic risk can be performed using an user-friendly, comprehensive tool (the OAC score).In patients with low risk OAC score, despite high risk CHA2DS2-VASC, anticoagulation is not beneficial and left atrial appendage closure could be an alternative.Further studies are needed to assess the impact of oral anticoagulation and left atrial appendage closure in patients with atrial fibrillation according to the OAC score risk groups.