Why do clinicians withhold anticoagulation in patients with atrial fibrillation and CHA2DS2-VASc score ≥ 2?

Abstract

Current guidelines recommend oral anticoagulant (OAC) to prevent stroke and systemic embolism for atrial fibrillation (AF) patients at high thromboembolic (TE) risk (CHA2DS2-VASc score ≥ 2). Approximately 40% of eligible patients do not receive recommended OAC. The reasons for guideline non-adherence are unclear. To identify patient characteristics associated with non-use of OAC for AF despite CHA2DS2-VASc score ≥ 2. Patient characteristics associated with OAC in the GARFIELD-AF registry were assessed using logistic regression analysis [OR (95%)]. The rate of annual all-cause mortality, ischemic stroke or systemic embolism (SSE), and major bleeding (%/y) were compared between patients with and without OAC with a log-rank test. P -values < 0.05 were considered significant. Subsequently, a sample of European physicians was questioned through a web-based survey to identify factors that may influence their decision-making. Among 52,014 patients enrolled in GARFIELD-AF, 42471 (83%) had a CHA2DS2-VASc score ≥ 2 and 12,884 (30%) of them did not receive OAC. Use of antiplatelet therapy [OR = 15.0(14.1–15.8)] and a history of bleeding [OR = 2.5(2.2–3.0)] were the strongest predictors of withholding OAC, followed by no hypercholesterolemia [OR = 1.4(1.3–1.5)] and no vascular disease [OR = 1.5(1.4–1.6)]. Compared to patients with OAC, those without OAC had an increased annual all-cause mortality (5.3 vs 3.9%, P < 0.001) and SSE (1.6 vs 1.1%, P < 0.001), and a decreased in major bleeding (0.5 vs. 0.8%, P < <0.001). Table 1 shows the most frequently reported characteristics leading to a preference to withhold OAC among 266 survey respondents. Antiplatelet therapy and history of bleeding were the strongest predictors of OAC non-use in AF patients with high TE risk in GARFIELD-AF registry. Major/critical site bleeding and cirrhosis may also influence non-use of OAC by some physicians. Guideline-based treatment with OAC was associated with better outcomes.