Evidence network of included studies.

The width of lines for each connection in the evidence network are proportional to the number of randomized controlled trials (RCTs) comparing each pair of treatments. Multiarm trials are represented in dotted lines. The size of each treatment node is proportional to the number of randomized participants (sample size). Direct oral anticoagulants for the main analysis included apixaban, dabigatran and rivaroxaban, whereas unmarketed drugs (shaded in gray) (idraparinux, sulodexide and ximelagatran) were used for sensitivity analyses only (see online supplement). ASA: aspirin; RCT: randomized controlled trial; VKA: vitamin K antagonist.