Ascites fluid in severe acute pancreatitis: from pathophysiology to therapy

Several pathophysiological mechanisms are involved in the development of the inflammatory necrotizing process that takes place in the retroperitoneal area during the early phase of acute pancreatitis. They include premature intraglandular activation of pancreatic proenzymes (zymogens) and in particular trypsin, early microcirculatory impairment with subsequent ischaemia/reperfusion and overstimulation of immune effector cells. Although intra-acinar or interstitial activation of trypsinogen is most probably the trigger of acute pancreatitis, in recent years much emphasis has been put on the role of leukocytes. Based on numerous experimental and human data several pro-inflammatory mediators including cytokines, arachidonic acid derivatives, activated oxygen species and proteases are released locally by overactivated neutrophils and monocytes/macrophages among other cells. They are now believed to play a central role in the development of pancreatic necrosis and, once they gain access to the systemic circulation, in the emergence of early multisystem organ failure. However the sequential and relative contribution of each of these 3 pathophysiological mechanisms remain controversial and the precise identification of the mediators incriminated in local and remote tissue injury is still awaited. Severe acute pancreatitis still carries a mortality of 20% to 30%. With advances in intensive care management 80% of the deaths occur somewhat late in the attack due to infected pancreatic necrosis. Nevertheless early remote organ failures still remain a lifethreatening condition for most of these patients. A peritoneal exudate rich in activated lipolytic and proteolytic enzymes, vasoactive substances and several other pro-inflammatory mediators collect in over 60% of the patients with severe acute pancreatitis. On the basis of favourable animal experiments early percutaneous or surgical peritoneal lavage with or without the addition of antiproteases has been carried out in human acute pancreatitis. The rationale behind this procedure was the washout of potential toxic mediators from the peritoneal cavity before they gain access to the systemic circulation. Contrary to animal and uncontrolled human data no prospective randomized study could ever demonstrated a significant effect of peritoneal lavage neither in the prevention and control of remote organ failures or in early mortality and ultimate survival after severe acute pancreatitis in humans. Differences between experimentally-induced pancreatitis, difference in the timing of the initiation of lavage and a type II error in controlled human studies may account for the discrepancy in the outcome between these studies. Anyway, this disparity should raise the question as whether the peritoneal cavity acts simply as a reservoir or as a route of transfer of toxic mediators to the systemic circulation. Although data are scarce, conflicting and limited to animal experiments and to a few molecules, peripancreatic veins and lymphatics seem to be the major routes of transfer whereas transperitoneal absorption is trivial. Nevertheless early peritoneal aspiration of ascitic fluid in acute pancreatitis and measurement of trypsinogen activation peptides may be used as a means of severity assessment and identification of pancreatic necrosis. This implies that even if not taking part actively in the emergence of remote organ failures ascitic fluid may reflect the peripancreatic necrotizing process. So careful comparative analysis of peritoneal exudate, plasma and lymph with regards to putative mediators of local and remote injury may provide essential pathophysiological clues. At the time of trials of antimediator therapy early in the attack this kind of insight is essential.