Vitamin K antagonists (VKA) continue to be the standard of long-term anticoagulation. Direct oral anticoagulants(DOAC) are increasingly used. In many trials DOAC were at least as effective as VKA. In this study we evaluate the bleeding profiles, frequencies and etiologies of patients receiving DOAC versus VKA in a real-life setting. All patients presenting with suspected gastrointestinal bleeding (GIB) in the emergency department of the University Hospital Erlangen in one year were enrolled in this study. They were looked up for the intake of either DOAC (dabigatran, rivaroxaban and apixaban) or VKA. The results showed that 406 patients with suspected GIB were admitted to the emergency unit of the University Hospital Erlangen. In 228 of those patients GIB could be verified (56.2%). Fifty four of those patients (23.7%) were administered either VKA or DOAC. In 35 of those 54 patients (64.8%) GIB was classified as 'major bleeding'. In 27 patients with administration of VKA upper GIB was recorded and lower GIB was detected four times. In 16 patients with administration of DOAC upper GIB was found and lower GIB was found in 7 patients. The presented data do not show higher GIB rates for DOAC (mainly dabigatran and rivaroxaban), but do also not indicate a significantly higher safety of DOAC concerning GIB than VKA. This finding represents a clear contrast to the reduced bleeding rates of DOAC for intracerebral bleeding and other non-GIB events. According to our study, the absolute number of DOAC-associated GIB events is lower than in the VKA group.