Background: mortality in severe acute pancreatitis varies from 10% to 20%. The early identification of patients with higher risk of complications is crucial to treat them properly. APACHE II is the most used scoring system to determine the severity of acute pancreatitis. However, some problems have been related, as the overestimation of severity, and other scores have been proposed, such as SOFA and Marshall. The aim of this study is to determine variables related to the development of organ failure, mortality and necrotizing pancreatitis in patients with severe acute pancreatitis.
Methods: evaluation of all patients with acute pancreatitis admitted in this hospital, including in the analysis only patients with APACHE II > 8 at admission. SOFA score and Marshall score were also obtained. The variables analyzed were age, sex, aetiology, hematocrit, leukocytes, C-reactive protein, computerized tomography and length of stay. These variables were related with the development of organ failure, mortality and necrotizing pancreatitis.
Results: One hundred and seventy-five patients were admitted with acute pancreatitis, and 39 (22.3%) were classified as severe acute pancreatitis due to APACHE II > 8. The mean APACHE II value was 11.6 ± 3.1, SOFA score 3.2 ± 2 and Marshall 1.5 ± 1.9. Respiratory failure was present in six (15,4%) patients with severe acute pancreatitis. Eleven patients developed necrotizing pancreatitis. Mortality of patients with APACHE II > 8 was 7.7%. The variables related with organ failure were APACHE II, SOFA > 3 and Marshall > 3, and variables related with mortality were SOFA > 3 and leukocytosis > 19,000. C-reactive protein > 19.5 mg/dl and length of stay were related to necrotizing pancreatitis.
Conclusion: the scoring systems, particularly the SOFA score, are related to the development of organ failure and mortality. C-reactive protein demonstrates relationship with necrotizing pancreatitis. There is no relationship between scoring systems and necrotizing pancreatitis in severe acute pancreatitis.