RESUMO
BACKGROUND: The Atlanta criteria for acute pancreatitis (AP) has been revised recently. This study was to evaluate its practical value in classification of AP, the severity assessment and management. METHODS: The clinical features, severity classification, outcome and risk factors for mortality of 3212 AP patients who had been admitted in Ruijin Hospital from 2004 to 2011 were analyzed based on the revised Atlanta criteria (RAC) and the original Atlanta criteria (OAC). RESULTS: Compared to the OAC group, the incidence of severe acute pancreatitis (SAP) was decreased by approximately one half (13.9% vs 28.2%) in the RAC group. The RAC presented a lower sensitivity but higher specificity, and its predictive value for severity and poor outcome was higher than those of the OAC. The proportion of SAP diagnosis and ICU admission in the early phase in the RAC group was significantly lower than that in the OAC group (P<0.05). Based on the RAC, the risk factors for death among SAP patients were older age, high CT severity index (CTSI), renal failure, cardiovascular failure, acute necrotic collection and walled-off necrosis. Compared to the OAC, the acute physiology and chronic health evaluation II (APACHE II) score, Ranson score, idiopathic etiology, respiratory failure and laparotomy debridement were not risk factors of death in contrast to walled-off necrosis. Interestingly, hypertriglyceridemia-related SAP had good outcomes in both groups. CONCLUSIONS: The RAC showed a higher predictive value for severity and poorer outcome than the OAC. However, the RAC resulted in fewer ICU admissions in the early phase due to its lower sensitivity for diagnosis of SAP. Among SAP cases, older age, high CTSI, renal and cardiovascular failure, complications of acute necrotic collection and walled-off necrosis were independent risk factors for mortality.