Pancreatitis is due to enzymatic digestion of the gland is supported by the finding of proteolytic enzymes in ascitic fluid and increased amounts of phospholipase A and lysolecithins in pancreatic tissue from patients with acute process.
Trypsin has not been found in excessive amounts in pancreatic tissue from affected humans, possibly because of inactivation by trypsin inhibitors. Nevertheless, although the available evidence is inconclusive, the autodigestion theory is almost universally accepted. Other proposed factors are vascular insufficiency, lymphatic congestion, activation of the kallikrein-kinin system.
For many years, trypsin and other proteases were held to be the principal injurious agents, but recent evidence has emphasized phospholipase A, lipase, and elastase as perhaps of greater importance. Trypsin ordinarily does not attack living cells, and even when trypsin is forced into the interstitial spaces, the resulting pancreatitis does not include coagulation necrosis, which is so prominent in human pancreatitis.
Phospholipase A, in the presence of small amounts of bile salts, attacks free phospholipids (eg, lecithin) and those bound in cellular membranes to produce extremely potent lyso-compounds. Lysolecithin, which would result from the action of phospholipase A on biliary lecithin, or phospholipase A itself, plus bile salts, is capable of producing severe necrotizing pancreatitis. Trypsin is important in this scheme, because small amounts are needed to activate phospholipase A from its inactive precursor.
Elastase, which is both elastolytic and proteolytic, is secreted in an inactive form. Because it can digest the walls of blood vessels, elastase has been thought to be important in the pathogenesis of hemorrhagic pancreatitis.
If autodigestion is the final common pathway in pancreatitis, earlier steps must account for the presence of active enzymes and their reaction products in the ducts and their escape into the interstitium. The following are the most popular theories that attempt to link the known etiologic factors with autodigestion:
Ligation of the pancreatic duct generally produces mild edema of the pancreas that resolves within a week. Thereafter, atrophy of the secretory apparatus occurs. On the other hand, partial or intermittent ductal obstruction, which more closely mimics what seems to happen in humans, can produce frank pancreatitis if the gland is simultaneously stimulated to secrete. The major shortcoming of these experiments has been the difficulty encountered in attempting to cause severe pancreatitis in this way. However, since the human pancreas manufactures ten times as much phospholipase A as does the dog or rat pancreas, the consequences of obstruction in humans conceivably could be more serious.
Common Channel Theory
Opie, having observed pancreatitis in a patients with a gallstones impacted in the ampulla of Vater, speculated that reflux of bile into the pancreatic duct might have initiated the process. Flow between the biliary and pancreatic duct requires a common channel connecting these two systems with the duodenum. Although these ducts converge in 90% of humans, only 10% have a common channel long enough to permit biliary-pancreatic reflux if the ampulla contained a gallstone. Experimentally, pancreatitis produced by pancreatic duct obstruction alone is similar in severity to pancreatitis following obstruction of a common channel, so biliary reflux is discounted as an etiologic factor in this disease.
The above theories do not explain activation of pancreatic enzymes, a process that normally takes place through the action of enterokinase in the duodenum. In experimental animals, if the segment of duodenum into which the pancreatic duct empties is surgically converted to a closed loop, reflux of duodenal juice initiates severe pancreatitis (Pfeffer loop). Pancreatitis associated with acute afferent loop obstruction after Billroth II gastrectomy is probably the result of similar factors. Other than in this specific example, there is no direct evidence for duodenal reflux in the pathogenesis of pancreatitis in humans.
Back Diffusion Across the Pancreatic Duct
Just as the gastric mucosa must serve as a barrier to maintain high concentrations of acid, so must the epithelium of the pancreatic duct prevent diffusion of luminal enzymes into the pancreatic parenchyma. Experiments in cats have shown that the barrier function of the pancreatic duct is vulnerable to several injurious agents, including alcohol and bile acids. Furthermore, the effects of alcohol can occur even after oral ingestion, because alcohol is secreted in the pancreatic juice. Injury to the barrier renders the duct permeable to molecules as large as MW 20,000, and enzymes from the lumen may be able to enter the gland and produce pancreatitis.
The studies by Steer and his coworkers have shown that a very early event in several forms of experimental pancreatitis, including that due to pancreatic duct obstruction, consists of zymogen activation within acinar cells by lysosomal hydrolases (eg, cathepsin B). This may represent the long-sought unifying explanation. Other factors must be postulated, however, to account for the variations in severity of the disease. In biliary pancreatitis, transient obstruction of the ampulla of Vater by a gallstone is most likely the first event. Alcoholic pancreatitis probably has several causes, including partial ductal obstruction, secretory stimulation, acute effects on the ductal barrier, and toxic actions of alcohol on parenchymal cells.
Severe acute pancreatitis may be complicated by multiple organ failure, principally respiratory insufficiency, myocardial depression, renal insufficiency, and gastric stress ulceration. The pathogenesis of these complications is similar in many respects to that of multiple organ failure in sepsis, and in fact, sepsis due to pancreatic abscess formation is a contributing factor in some of the severe cases of pancreatitis. During acute pancreatitis, pancreatic proteases, bacterial endotoxins, and other active agents are liberated into the systemic circulation. The concentrations of serum factors able to complex with the proteases (eg, 2-macroglobulin) decrease in proportion to the severity, and complexed 2-macroglobulin, which normally is cleared rapidly by macrophages, accumulates. These circulating complexes, which retain proteolytic activity, are thought to contribute to systemic toxicity. The endotoxin probably originates from bacteria that translocate through an abnormally permeable intestinal mucosa. Within the circulation, the proteases and the endotoxin activate the complement system (especially C5) and kinins. Complement activation leads to granulocyte aggregation and accumulation of aggregates in the pulmonary capillaries. The granulocytes release neutrophil elastase, superoxide anion, hydrogen peroxide, and hydroxide radicals, which in concert with bradykinin exert local toxic effects on the pulmonary epithelium that result in increased permeability. Arachidonate metabolites (eg, PGE2, PGI2, leukotriene B4) may be involved in some way. Analogous events are thought to occur in other organs.