Acute pancreatitis is a common digestive disease of which the severity may vary from mild, edematous to severe, necrotizing disease. An improved outcome in the severe form of the disease is based on early identification of disease severity and subsequent focused management of these high-risk patients. However, the ability of clinicians to predict, upon presentation, which patient will have mild or severe acute pancreatitis is not accurate. Prospective systems using clinical criteria have been used to determine severity in patients with acute pancreatitis, such as the Ranson's prognostic signs, Glasgow score, and the acute physiology and chronic health evaluation II score (APACHE II). Their application in clinical practise has been limited by the time delay of at least 48 h to judge all parameters in the former two and by being cumbersome and time-consuming in the latter. Contrast-enhanced computed tomography is presently the most accurate non-invasive single method to evaluate the severity of acute pancreatitis. It cannot, however, be performed to all patients with acute pancreatitis. Therefore, considerable interest has grown in the development of reliable biochemical markers that reflect the severity of acute pancreatitis. In this article we critically appraise current and new severity markers of acute pancreatitis in their ability to distinguish between mild and severe disease and their clinical utility.