A best evidence topic in cardiac surgery was written according to the structured protocol. The question addressed was about the best time to restart anticoagulation in patients with intracranial bleed with a prosthetic valve in situ. This difficult clinical decision has to balance the risk of thromboembolism during the period that the anticoagulation was reversed and later withheld vs the risk of haematoma expansion or rebleed if the anticoagulation was started early. Altogether, more than 80 papers were found using the reported search, of which 10 represented the best evidence to answer the clinical question. There were two prospective studies and eight retrospective studies. There were no randomized controlled trials on this topic. The authors, journal, date and country of publication, patient group studied, study type, relevant outcomes and results of these papers are tabulated. Seven studies reported the strategy of reversal of anticoagulation with vitamin K, fresh frozen plasma or prothrombin concentrate. The emphasis was on prompt initial reversal of anticoagulation; however, the best agent for reversal was not defined. Four studies dealt exclusively with intracranial bleed in patients with prosthetic valve in situ. The remaining six studies on intracranial bleed had only a subset of patients with a prosthetic valve in situ. The anticoagulation was restarted with heparin and later switched to oral anticoagulant. Thromboembolic events during the period of reversal and cessation of anticoagulants were low (5%) as was the incidence of rebleed or haematoma expansion (0.5%). We conclude that anticoagulation can safely be withheld for a short period, up to 7-14 days in a patient with intracranial bleed with a very low probability of thromboembolic phenomenon. In patients with prosthetic valves, in situ anticoagulation in the form of heparin can safely be restarted as early as 3 days and switched to oral anticoagulation in the form of warfarin at 7 days without major concerns of bleeding.