INTRODUCTIONAcute pancreatitis (AP) contributes to thousands of annual hospital admissions, of which severe acute pancreatitis (SAP) accounts for 10%-20% [1][2][3] . Despite considerable improvement in the treatment, the mortality of SAP still ranges between 10%-15% [2][3][4][5] . The course of SAP tends to be prolonged and the patients usually are hypermetabolic and high protein catabolic due to systemic inflammatory response syndrome (SIRS) induced by acute local inflammatory process and subsequent vital-organ dysfunction [6] . Thus, if nutritive support is not appropriately administrated to match rapidly increased demand in the treatment of SAP, the patients consequently come down with metabolic disorder and nutrition deficiency, which is considered to increase mortality due to impaired immune function, increased risk of infections and intractable vital-organ failure.In recent years, research showed a conditional deficiency of glutamine would be an independent predictive factor for a poor outcome and its correction improved survival by restoring cellular protective mechanisms, improving immune function and resistance of the gut barrier to hypoperfusion, metabolic stress and subsequent bacterial translocation, and decreasing the risk of infection in critical illness [7][8][9][10] . Since free glutamine is instable in solution, intravenous administration is limited [7] . Alanyl-glutamine
METHODS:Eighty patients with SAP were randomized and received 100 mL/d of 20% AGD intravenously for 10 d starting either on the day of (early treatment group) or 5 d after (late treatment group) admission. Groups had similar demographics, underlying diseases, Ranson score, Acute Physiology and Chronic Health Evaluation Ⅱ (APACHE Ⅱ) score, and Balthazar 's computed tomography (CT) score at the beginning of the study and underwent similar other medical and nutritional management.
RESULTS:The duration of acute respiratory distress syndrome (2.7 ± 3.3 d vs 12.7 ± 21.0 d, P < 0.01), renal failure (1.3 ± 0.5 d vs 5.3 ± 7.3 d, P < 0.01), acute hepatitis (3.2 ± 2.3 d vs 7.0 ± 7.1 d, P < 0.01), shock (1.7 ± 0.4 d vs 4.8 ± 3.1 d, P < 0.05), encephalopathy (2.3 ± 1.9 d vs 9.5 ± 11.0 d, P < 0.01) and enteroparalysis (2.2 ± 1.4 d vs 3.5 ± 2.2 d, P < 0.01) and hospital stay (28.8 ± 9.4 d vs 45.2 ± 27.1 d, P < 0.01) were shorter in the early treatment group than in the late treatment group. The 15-d APACHE Ⅱ score was lower in the early treatment group than in the late treatment group (5.0 ± 2.4 vs 8.6 ± 3.6, P < 0.01). The infection rate (7.9% vs 26.3%, P < 0.05), operation rate (13.2% vs 34.2%, P < 0.05) and mortality (5.3% vs 21.1%, P < 0.05) in the early treatment group were lower than in the late treatment group.Xue P et al . Impact of AGD on SAP 475 www.wjgnet.com dipeptide (AGD), however, can be taken via vein and hydrolyzed into alanine and glutamine in circulation as a substitute [7] . Presently AGD supplement in parenteral nutrition is a worth-trying approach and an evidence-based recommendation in the management of SAP [11] , but ...