Background Patients taking direct oral anticoagulant medications (DOACs) commonly undergo computed tomography (CT) head scanning following mild head injury, regardless of symptoms or signs. International guidelines have noted a lack of evidence to support management decisions in such patients.

Methods A systematic review, pre-registered (CRD42017071411) and following Cochrane Collaboration recommendations, was performed. Studies of adults with mild head injury (GCS 13–15) taking DOACs, which reported the risk of adverse outcome following the head injury, were eligible for inclusion. A comprehensive range of bibliographic databases and grey literature were examined using a sensitive search strategy. Selection of eligible studies, data extraction, and risk of bias was evaluated independently by separate reviewers. A random effects meta-analysis was used to provide a pooled estimate of the risk of adverse outcome. The overall quality of evidence was assessed using the Grades of Recommendation, Assessment, Development and Evaluation Working Group approach.

Results 4,185 articles were screened for inclusion, of which 7 cohort studies, including 346 patients, met inclusion criteria. All studies were at high or unclear risk of bias secondary to selection and information bias. Estimates of adverse outcome (any death, intracranial hematoma (ICH), or neurosurgery) ranged from 0% to 8%. A random effects meta-analysis showed a weighted composite outcome risk of 4% (95% CI 2–6%, I²=3%). The overall quality of the body of evidence was low secondary to imprecision, indirectness and risk of bias.

Conclusions There is limited data available to characterize the risk of adverse outcome in patients taking DOACs following mild head injury. A sufficiently powered prospective cohort study is required to validly define this risk, identify clinical features predictive of adverse outcome, and inform future head injury guidelines.