Background and Aims Alcohol‐related liver disease is the most frequent cause of cirrhosis and a major indication for liver transplantation. Several alcohol use biomarkers have been developed in recent years and are already in use in several centers. However, in patients with liver disease their diagnostic performance might be influenced by altered biomarker formation by hepatic damage, altered excretion by kidney dysfunction and diuretics use, and altered deposition in hair and nails. We systematically reviewed studies on the diagnostic accuracy of biomarkers of alcohol use in patients with liver disease and performed a detailed study quality assessment. Methods A structured search in PubMed/Medline/Embase databases was performed for relevant studies, published until April 28, 2019. The risk of bias and applicability concerns was assessed according to the adapted quality assessment of diagnostic accuracy studies‐2 (QUADAS‐2) checklist. Results Twelve out of 6,449 studies met inclusion criteria. Urinary ethyl glucuronide and urinary ethyl sulfate showed high sensitivity (70 to 89 and 73 to 82%, respectively) and specificity (93 to 99 and 86 to 89%, respectively) for assessing any amount of alcohol use in the past days. Serum carbohydrate‐deficient transferrin showed low sensitivity but higher specificity (40 to 79 and 57 to 99%, respectively) to detect excessive alcohol use in the past weeks. Whole blood phosphatidylethanol showed high sensitivity and specificity (73 to 100 and 90 to 96%, respectively) to detect any amount of alcohol use in the previous weeks. Scalp hair ethyl glucuronide showed high sensitivity (85 to 100%) and specificity (97 to 100%) for detecting chronic excessive alcohol use in the past 3 to 6 months. Main limitations of the current evidence are the lack of an absolute gold standard to assess alcohol use, heterogeneous study populations, and the paucity of studies. Conclusions Urinary and scalp hair ethyl glucuronide are currently the most validated alcohol use biomarkers in patients with liver disease with good diagnostic accuracies. Phosphatidylethanol is a highly promising alcohol use biomarker, but so far less validated in liver patients. Alcohol use biomarkers can complement each other regarding diagnostic time window. More validation studies on alcohol use biomarkers in patients with liver disease are needed.