15: Disorders of the Exocrine Pancreas

The pancreas is a gland with both exocrine and endocrine functions. The exocrine pancreas contains acini, which secrete pancreatic juice into the duodenum through the pancreatic ducts (Figure 15–1). Pancreatic juice contains a number of enzymes, some of which are initially secreted as zymogens in an inactive form. Once activated, these enzymes help to digest food and prepare it for absorption in the intestine. Disorders interfering with normal pancreatic enzyme activity (pancreatic insufficiency) cause maldigestion of fat and steatorrhea (fatty stools). Pathology of the exocrine pancreas results from inflammation (acute pancreatitis, chronic pancreatitis), neoplasm (ductal adenocarcinoma, neuroendocrine tumors, and other pancreatic neoplasms), or duct obstruction by stones or abnormally viscid mucus (cystic fibrosis).

The endocrine pancreas is composed of the islets of Langerhans. The islets are distributed throughout the pancreas and contain several different hormone-producing cells. The islet cells manufacture hormones such as insulin that are important in nutrient absorption, storage, and metabolism. Dysfunction of the endocrine pancreas may cause diabetes mellitus (see Chapter 18).

Both exocrine and endocrine pancreatic dysfunction occur together in some patients.

Anatomy

The pancreas is a solid organ that lies transversely in the retroperitoneum deep within the epigastrium. It is firmly fixed by fibrous attachments anterior to the suprarenal aorta and the first and second lumbar vertebrae. Thus, the pain of acute or chronic pancreatitis is situated deep in the epigastric region and frequently radiates to the back.

Normally, the pancreas is about 15 cm long, although the normal pancreas weighs less than 110 g. The organ is covered by a thin capsule of connective tissue that sends septa into it, separating it into lobules.

The pancreas can be divided into four parts: head, including the uncinate process; neck; body; and tail. The head is the thickest part of the gland (2–4 cm) and lies in the “C-loop” or curved space between the first, second, and third portions of the duodenum. The uncinate process is the portion of the head that extends dorsally and to the left behind the superior mesenteric vessels. The neck connects the head and body and sits immediately ventral to the superior mesenteric vessels. The body is situated transversely in the retroperitoneal space, bordered superiorly by the splenic artery and dorsally by the splenic vein. The tail of the pancreas is less fixed in the retroperitoneum and extends toward, and often immediately adjacent to, the hilum of the spleen.

Embryologically, the pancreas develops as two separate endodermal buds from the developing foregut. These separate dorsal and ventral elements of the primordial pancreas initially develop opposite each other but with rotation of the primitive gut end up fusing together toward the left of the duodenum. The development of the dorsal and ventral buds is regulated by a complex process of intrinsic signals from the endodermal cells themselves, as well as extrinsic signals from surrounding mesoderm. The dorsal bud differentiates into the more cephalad and anterior portion of the pancreatic head, as well as the neck, body, and tail of the pancreas. The dorsal bud contains the accessory pancreatic duct (duct of Santorini), which enters the duodenum at the minor papilla. The smaller ventral bud arises adjacent to the bile duct to become the more caudal portion of the pancreatic head and the uncinate process. The ventral bud contains the major pancreatic duct (duct of Wirsung), which enters the duodenum at the ampulla of Vater (Figure 15–1) together with the common bile duct. As cellular and mesenchymal elements differentiate, the dorsal and ventral buds develop a conjoined ductal system, and the whole organ eventually takes its place in the retroperitoneum of the upper abdomen. Primitive neuroendocrine cells arise amid the developing ductal structures and eventually form the interspersed islets of Langerhans.

The major pancreatic duct of Wirsung normally becomes the drainage for the majority of the pancreatic parenchyma. This duct is normally about 3–4 mm in diameter. In most individuals, the pancreatic duct enters the duodenum at the duodenal papilla alongside the common bile duct, surrounded by the sphincter of Oddi, which controls the drainage of both ducts. In about one-third of individuals, the duct of Wirsung and the common bile duct join to form a common channel before terminating at the ampulla of Vater (Figure 15–1).

Pancreas divisum, the most common congenital anomaly of the pancreas, occurs when the embryologic ventral and dorsal components of the pancreas fail to fuse, thus leaving two distinct ductal systems that do not communicate and that separately drain into the duodenum through two different papillae. The smaller system drains through the major papilla, but the dominant dorsal system drains through the minor papilla. This situation can cause a relative obstruction to flow of pancreatic juice and may be associated with the development of pancreatitis. Pancreas divisum occurs in up to 7% of autopsy series.

Histology

The exocrine pancreas consists of clusters of enzyme-secreting acini centered around and individually drained by ductules. The islets of Langerhans of the endocrine pancreas are clusters of a few hundred hormone-secreting cells, each located between the lobules.

Each pancreatic acinus is composed of several acinar cells surrounding a lumen (Figure 15–2). Centroacinar cells are centrally located in the acini, interpositioned between the acinar cells and ductal epithelium. The centroacinar cells are believed to have a primary role in the secretion of electrolytes and water into the pancreatic ductal system. The acinar cells synthesize and secrete enzymes. On histologic examination, acinar cells are typical exocrine glandular cells. They are pyramidal epithelial cells arranged in rows. Their apexes join to form the lumen of the acinus. Zymogen granules containing digestive enzymes or their precursors are found in the acinar cells. These granules are discharged by exocytosis from the apexes of the cells into the lumen. The number of zymogen granules in the cells varies; more are found during fasting and fewer after a meal.

Acini are centered on small branches of the pancreatic duct, which eventually converge to form the continuous lumen of the main pancreatic duct. The mature ducts are lined with a continuous layer of ductal epithelial cells, joined by tight junctions. The ductal epithelium contributes to the secretion of water and electrolytes into pancreatic secretions and forms the important epithelial barrier separating the pancreatic parenchyma from the enzyme-rich ductal secretions. Compromise of this epithelial barrier due to inflammation or trauma may be associated with significant peripancreatic inflammation and severe clinical sequelae.

Physiology

Composition of Pancreatic Juice

As much as 1500 mL of pancreatic juice is secreted each day by a normal pancreas. Disease states (eg, chronic pancreatitis) may be associated with a marked decrease in exocrine pancreatic secretion. Pancreatic juice contains water, ions, and a variety of proteins. The principal ions in pancreatic juice are HCO₃⁻, Cl⁻, Na⁺, and K⁺. Of these, HCO₃⁻ is particularly important. At maximum flow rates, the concentration of HCO₃⁻ in pancreatic juice may reach 150 mEq/L (vs. 24 mEq/L in plasma), and the pH of the juice may reach 8.3. The alkaline nature of pancreatic juice plays a major role in neutralizing the gastric acid entering the duodenum with ingested food (chyme) from the stomach. The pH of the duodenal contents rises to 6.0–7.0, and by the time the chyme reaches the jejunum, its pH is nearly neutral.

Proteomic analysis suggests that more than 200 proteins exist in pancreatic secretions. Many of these ubiquitous proteins have various roles in cellular growth and signaling, while others are engaged in cellular immunology. The remainder of the pancreatic proteins secreted are responsible for the digestive functions of the exocrine pancreas. A primary function of the pancreas is the digestion of proteins in partially digested ingested food. This digestion is mediated by trypsinogen and other secreted proteases. However, the exocrine pancreas also secretes enzymes responsible for metabolism and absorption of lipids (lipase, colipase) and carbohydrates (amylase, enolase).

Some of the pancreatic enzymes (lipase, amylase, deoxyribonuclease, and ribonuclease) are secreted by the acinar cells in their active forms. The remaining enzymes are secreted as inactive proenzymes or zymogens (trypsinogen, chymotrypsinogen, proelastase, procarboxypeptidase, and phospholipase A₂) that are activated in the lumen of the proximal intestine. Aberrant activation of zymogens within the acinar cell is hypothesized to lead to acute pancreatitis and pancreatic autodigestion.

When the pancreatic juice enters the duodenum, trypsinogen is converted to the active form trypsin by an enzyme found in the intestinal brush border called enteropeptidase. Trypsin then converts the remaining proenzymes into active enzymes (eg, chymotrypsinogen into chymotrypsin). Trypsin can also activate its own precursor, trypsinogen, producing the potential for an autocatalytic chain reaction.

When trypsinogen is activated within the pancreas itself, two known protective mechanisms are available. First, there is inhibition of activated trypsin by pancreatic secretory trypsin inhibitor (PSTI), also known as serine protease inhibitor, Kazal type 1 (or SPINK1), which can inhibit approximately 20% of trypsin activity. If trypsin activity overwhelms the SPINK1/PSTI inhibitory capacity, trypsin inactivation can occur through trypsin autolysis.

Regulation of Secretion of Pancreatic Juice

Recent advances in our understanding of pancreatic exocrine function reveal hormonal and neural factors as two distinct but interactive elements that regulate secretion. Two hormones in particular appear to have a primary role in the secretion of pancreatic enzymes, secretin and cholecystokinin (CCK). Both hormones are produced by specialized enteroendocrine cells of the duodenal mucosa and act by distinct but synergistic intracellular pathways on the pancreatic acinar cells.

Secretion of secretin is triggered by gastric acid and by the products of protein digestion in the duodenum. Secretin acts chiefly on the pancreatic ductal epithelial, centroacinar, and to a lesser extent acinar cells to produce HCO₃⁻, thus raising the pH of the pancreatic secretions. Secretion of H₂O is also increased in response to secretin, increasing the absolute volume of pancreatic juice. Mechanistic studies have demonstrated that secretin and the related hormone vasoactive intestinal peptide (VIP) act on ductal and acinar cells by activation of adenylate cyclase and subsequent cAMP-dependent protein kinase A. Proteomic analysis has revealed that secretin does not appear to alter the constituents of pancreatic juice but instead regulates the relative proportions of secreted enzymes.

Secretion of CCK is triggered by the products of protein and fat digestion (peptides, amino acids, and fatty acids) when they enter the duodenum. The release of CCK from specific intestinal cells appears to be regulated by a cholecystokinin-releasing peptide in the proximal small intestine that is trypsin sensitive and active in the lumen. CCK controls pancreatic exocrine secretion through two mechanisms: 1) activation of neurons located in the dorsal motor nucleus of the vagus motor neurons that control parasympathetic signals, and 2) direct action on pancreatic acinar cells. CCK release raises intracellular Ca²⁺ concentrations, which lead to release of pancreatic enzymes from the zymogen granules. Related enteric hormones acetylcholine and gastrin-releasing peptide (GRP) appear to act by similar calcium-dependent pathways. The integrated action of both secretin and CCK produces abundant secretion of enzyme-rich, alkaline pancreatic juice.

When both cAMP and calcium-dependent pathways are stimulated, the effect within the acinar cell is greater than the sum of their individual activities. Thus, CCK and secretin appear to act synergistically in response to a meal to stimulate the production of a large volume of alkaline pancreatic juice rich in digestive enzymes. Recent evidence also implicates a number of other gastrointestinal (GI) hormones and peptides (ghrelin, leptin, and melatonin) in the regulation of pancreatic endocrine and exocrine secretion.

Digestive Functions of Pancreatic Juice

The secretion of pancreatic juice aids digestion in several ways. The large amount of bicarbonate in the juice helps to neutralize the acidic chyme from the stomach so that the pancreatic enzymes can function optimally in a neutral pH range.

Each of the enzymes also has an important digestive function. In digesting carbohydrates, pancreatic amylase splits straight-chain glucose polysaccharides (so-called amyloses in starch) into smaller α-limit dextrins, maltose and maltotriose. Brush border enzymes in the small intestine complete the hydrolysis of these smaller sugars into glucose, which is transported across the intestinal epithelium by Na⁺-coupled transport. Pancreatic lipase contributes to fat metabolism by hydrolyzing triglycerides into fatty acids and a monoglyceride; this activity is most efficient in the presence of bile acids, which serve to emulsify the triglycerides. Phospholipase A₂ splits a fatty acid off from lecithin to form lysolecithin. Ribonuclease and deoxyribonuclease attack the nucleic acids. The remaining enzymes help to digest proteins. Trypsin, chymotrypsin, and elastase are endopeptidases (ie, they cleave peptide bonds in the middle of polypeptide chains). Carboxypeptidase is an exopeptidase (ie, it splits peptide bonds adjacent to the carboxyl terminals of peptide chains). Together, these proteases break down proteins into oligopeptides and free amino acids.

Acute Pancreatitis

Clinical Presentations

Acute pancreatitis is a clinical syndrome resulting from acute inflammation and destructive autodigestion of the pancreas and peripancreatic tissues. Acute pancreatitis is the third most common indication for hospital admission among GI diseases and is associated with significant morbidity and mortality. Data from the National Center for Health Statistics clearly document a near doubling of cases of hospital admissions due to acute pancreatitis between 1985 and 2005. Fortunately, the overall survival rate of patients with acute pancreatitis is increasing, despite the continued high mortality rate (20–25%) in patients admitted to the intensive care unit with severe acute pancreatitis.

Etiology

Acute pancreatitis has many causes, as summarized in Table 15–1. In clinical practice, biliary tract disease and alcohol ingestion account for the majority of cases, with metabolic causes, mechanical etiologies, drug reactions, and traumatic injuries accounting for almost all of the remaining cases. Regardless of etiology, the pathogenesis of pancreatic injury, associated systemic effects, and risk factors for severe acute pancreatitis appear to be similar.

Alcohol use is commonly associated with acute pancreatitis in developed countries. Acute pancreatitis typically occurs after a binge of heavy drinking; chronic heavy alcohol ingestion clearly may lead to chronic pancreatitis and may increase susceptibility to episodes of acute pancreatitis. A number of mechanisms are responsible for alcohol-induced damage to the pancreas. Alcohol or its metabolite, acetaldehyde, can have a direct toxic effect on pancreatic acinar cells, leading to intracellular trypsin activation by the lysosomal enzymes. Additionally, inflammation of the sphincter of Oddi can lead to retention of hydrolytic enzymes in the pancreatic duct and acini. Malnutrition may predispose alcoholics to pancreatic injury. For example, deficiencies of trace elements such as zinc or selenium occur in alcoholic patients and are associated with acinar cell injury. Metalloenzymes such as superoxide dismutase, catalase, and glutathione peroxidase are important scavengers of free radicals.

In patients who do not drink alcohol, the most common cause of acute pancreatitis is biliary tract disease. In such cases, the hypothesized mechanism is obstruction of the common bile duct and the main pancreatic duct when a gallstone or biliary sludge becomes lodged at the ampulla of Vater. Reflux of bile or pancreatic secretions into the pancreatic duct leads to parenchymal injury. Others have proposed that bacterial toxins or free bile acids travel via lymphatics from the gallbladder to the pancreas, giving rise to inflammation. In either case, acute pancreatitis associated with biliary tract disease is more common in women because gallstones are more common in women.

A significant proportion of “gallstone” pancreatitis is not associated with discrete, measurable gallstones passing through the bile duct and obstructing the ampulla. Instead, biliary sludge, or microlithiasis, is believed to play an etiologic role in many cases of pancreatitis, which were previously classified as idiopathic. Endoscopic retrograde cholangiopancreatography (ERCP) performed in such cases often identifies the microlithiasis and viscous particulate bile in the distal common bile duct, which can cause transient biliary obstruction and activate the same mechanistic pathways that lead to pancreatitis as happens with larger gallstones. An alternative mechanism that has been proposed is recurrent passage of microlithiasis causing papillary stenosis or sphincter of Oddi dysfunction.

Thus, the absence of obvious gallstones on imaging studies does not definitively rule out a biliary cause of acute pancreatitis. Biliary microlithiasis may be suspected when an ultrasound shows low-level echoes that gravitate toward the dependent portion of the gallbladder without the acoustic shadowing typical of gallstones. Microlithiasis is documented when cholesterol monohydrate crystals and calcium bilirubinate granules are found on light microscopy of an endoscopically acquired, centrifuged specimen of bile. In clinical practice, this diagnosis is often made in a patient with an appropriate presentation and with risk factors for biliary microlithiasis including pregnancy, rapid weight loss, critical illness, prolonged fasting, total parenteral nutrition, administration of certain drugs (ceftriaxone and octreotide), and bone marrow or solid organ transplantation.

Acute pancreatitis may result from a variety of infectious agents, including viruses (mumps virus, coxsackievirus, hepatitis A virus, HIV, or cytomegalovirus) and bacteria (Salmonella typhi or hemolytic streptococci). Patients with HIV infection can develop acute pancreatitis from the HIV infection itself, from related opportunistic infections, or from antiretroviral therapies. In HIV-infected patients, pancreatitis has been associated with intravenous drug abuse, pentamidine therapy, Pneumocystis jirovecii and Mycobacterium avium-intracellulare infections, and gallstones.

Blunt or penetrating trauma and other injuries may cause acute pancreatitis. Pancreatitis sometimes occurs after surgical procedures near the pancreas (duodenal stump syndrome, pancreatic tail syndrome after splenectomy). Shock and hypothermia may cause decreased perfusion, resulting in cellular degeneration and release of pancreatic enzymes. Radiation therapy of retroperitoneal malignant neoplasms can sometimes cause acute pancreatitis, likely by injury to the microvasculature and acinar architecture.

Marked hypercalcemia, such as that associated with hyperparathyroidism, sarcoidosis, hypervitaminosis D, or multiple myeloma, causes acute pancreatitis in about 10% of cases. Two mechanisms have been hypothesized. The high plasma calcium concentration may cause calcium to precipitate in the pancreatic duct, leading to ductal obstruction. Alternatively, hypercalcemia may stimulate activation of trypsinogen in the pancreatic duct.

Pancreatitis is also associated with hyperlipidemia, particularly those types characterized by increased plasma levels of chylomicrons (types I, IV, and V). In these cases, it is postulated that free fatty acids liberated by the action of pancreatic lipase cause gland inflammation and injury. Alcohol abuse or oral contraceptive use increases the risk of acute pancreatitis in patients with hyperlipidemia.

A variety of drugs have been associated with pancreatitis, including corticosteroids, thiazide diuretics, immunosuppressants, and cancer chemotherapeutic agents.

Rarely, acute pancreatitis may be familial, occurring with an autosomal dominant inheritance pattern. Hereditary pancreatitis typically presents as recurring acute pancreatitis in childhood, progressing to chronic pancreatitis by young adulthood in more than 50% of cases. Hereditary recurrent acute pancreatitis has been associated with mutations of the cationic trypsinogen gene (protease, serine, 1; PRSS1) mapped to chromosome 7q35. Two point mutations, R122H and N29I, account for most cases and can be detected by genetic testing. Studies have suggested that the R122H mutation is associated with more severe acute pancreatitis, leading to more frequent attacks and hospital admissions. Other families have mutations in SPINK1/PSTI. It appears that mutations in cationic trypsinogen enhance trypsinogen autoactivation by altering calcium-mediated regulatory pathways, and mutations in SPINK1/PSTI diminish inhibition of active trypsinogen. Other mutations eliminate the trypsin autolysis site. Patients demonstrated to have hereditary pancreatitis should be enrolled in a pancreatic cancer surveillance program, and total pancreatectomy should be considered in select cases, as approximately 40% of affected patients develop pancreatic cancer by age 70 years.

In recent years, our understanding of the diagnosis and classification of autoimmune pancreatitis has evolved. This chronic disease of fibrosis and lymphoplasmacytic inflammation may cause both acute episodes of pancreatitis as well as chronic injury. Two subtypes have been characterized. Type I autoimmune pancreatitis accounts for more that 80% of cases in the United States and is associated with elevated serum levels of IgG4 and with lymphocytic infiltration throughout the pancreatic parenchyma. Many patients with type I autoimmune pancreatitis have extrapancreatic manifestations and are often classified as having IgG4-related disease. Type II autoimmune pancreatitis is more common outside of the United States and does not appear to be IgG4-mediated. The pathognomonic histopathologic findings in this disease are granulocyte-epithelial lesions with neutrophilic infiltration. Type II autoimmune pancreatitis more often presents with acute pancreatitis compared with type I disease.

In about 15–25% of cases of acute pancreatitis, no etiologic factor can be identified. Idiopathic acute recurrent pancreatitis is seen in patients with more than one attack of acute pancreatitis when the underlying cause eludes detection despite a thorough search.

Pathology

The symptoms, signs, laboratory findings, and complications of acute pancreatitis can be explained on the basis of the pathologic damage to the ductules, acini, and islets of the pancreas. However, both the degree of damage and the clinical consequences are quite variable.

When the damage is limited in extent, the pathologic features consist of mild to marked swelling of the gland, especially the acini, and mild to marked infiltration with polymorphonuclear neutrophils. However, damage to tissue is usually only minimal to moderate, and there is no hemorrhage. In some cases, suppuration may be found along with edema, and this may result in tissue necrosis and abscess formation. In severe cases, massive necrosis and liquefaction of the pancreas occur, predisposing to pancreatic abscess formation. Vascular necrosis and disruption may occur, resulting in peripancreatic hemorrhage. While microvascular hemorrhage involving peripancreatic tissue is common in severe cases of acute pancreatitis, significant bleeding from large vessel erosion is a rare clinical entity and is more often seen in chronic pancreatitis.

Severe cases of pancreatitis may be associated with the formation of ascites, which is likely a combination of serous fluid excreted by the inflamed peritoneal surface, liquefied peripancreatic fat, blood from peripancreatic tissues, and necrotic pancreatic debris. In rare cases associated with ductal disruption of the pancreas, the ascites may contain frank pancreatic secretions rich in amylase and other pancreatic enzymes. Documentation of amylase-rich peritoneal fluid establishes the diagnosis of so-called pancreatic ascites. In cases of severe acute pancreatitis, the peritoneal surfaces have a characteristic appearance upon surgical exploration or autopsy; fat necrosis, or saponification, may occur in and around the pancreas, omentum, and mesentery, appearing as chalky white foci that may later calcify.

Histologic studies of pancreas tissue obtained from patients with first attacks of acute alcoholic pancreatitis who underwent surgery for complications have found that the acute pancreatitis (pancreatic necrosis, steatonecrosis, infiltration by inflammatory cells) sometimes develops in a gland already affected by chronic pancreatitis (perilobular and intralobular fibrosis, loss of exocrine parenchyma and atrophy of residual lobules, dilated interlobular and intralobular ducts lined with cuboidal or flattened epithelium, and protein plugs within dilated ducts). It has been conjectured that, if acute alcoholic pancreatitis develops in a pancreas already affected by chronic pancreatitis, it is due to obstruction of the ducts by protein plugs, an early lesion of chronic pancreatitis.

Pathogenesis

The pathogenesis of acute pancreatitis remains only partially understood. The central theory in this disease has long centered on the aberrant activation of trypsinogen and other enzymes within the pancreatic acini, causing autodigestion and a profound systemic inflammatory response. Recent evidence suggests that other events parallel to trypsinogen activation occur, such as the activation of NFκB (nuclear factor kappa-light-chain-enhancer of activated B cells, a protein complex that controls the transcription of DNA) that can induce acute pancreatitis in experimental models (Figure 15–3). However, elegant studies have confirmed that expression of active trypsin within pancreatic acini is itself sufficient to induce cell death and inflammation in acute pancreatitis. Thus, the in vivo role of alternative mechanisms of pancreatic autodigestion remains unclear.

Typsinogen activation is associated with a sustained cytosolic influx of calcium (Ca²⁺) mediated by calcium channels in the plasma membrane as well as by calcium receptors in the endoplasmic reticulum. Calcineurin is a likely downstream target of elevated intracellular Ca²⁺ levels, mediating some of the injury observed in acute pancreatitis via T-cell activation.

Trypsinogen is likely activated within membrane-bound intracellular compartments that exhibit dysregulated autophagy in the setting of acute pancreatitis. While cathepsin B within lysosomes has been shown to activate trypsinogen, this likely occurs only in certain pathologic conditions such as a low intracellular pH. The mechanism of pH disturbance within acinar cells is likely due to an alteration in cell signaling mechanisms and inhibition of acinar bicarbonate secretion. Moreover, although cathepsin L (an alternate isoform of cathepsin B) normally degrades trypsin in an important cellular protective mechanism, disturbance of the intracellular environment has been shown to contribute to an imbalance in cathepsin B activity relative to cathepsin L activity.

The pathogenesis of alcoholic pancreatitis may be unique and may involve disordered agonist-receptor interaction on the membrane of pancreatic acinar cells. According to this theory, alcohol increases activation of intrapancreatic digestive enzymes either by sensitizing acinar cells to pathologic stimuli or by stimulating the release of the secretagogue, cholecystokinin (CCK), from duodenal cells. The hyperstimulation of pancreatic acinar cells and their muscarinic receptors mimics the mechanism of acute pancreatitis caused by scorpion stings, anti-acetylcholinesterase-containing insecticide poisoning, or administration of supramaximal doses of secretagogues such as acetylcholine and CCK. CCK receptor activation can initiate different patterns of zymogen activation in pancreatic acinar cells, and the extent of activation is enhanced by a distinct set of short-chain alcohols. Whether ethanol or other alcohols mediate these effects by interfering with acinar cell signaling pathways or by affecting acinar cell membrane fluidity is currently under investigation.

The pathologic changes result from the action of activated trypsin and other pancreatic enzymes on the pancreas and surrounding tissues. Activated trypsin in turn activates the proenzymes of chymotrypsin, elastase, and phospholipase A₂, and those enzymes cause damage in several ways (Figure 15–4). For example, chymotrypsin activation leads to edema and vascular damage. Similarly, elastase, once activated from proelastase, digests the elastin in blood vessel walls and causes vascular injury and hemorrhage; damage to peripancreatic blood vessels can lead to hemorrhagic pancreatitis. Phospholipase A₂ splits a fatty acid off lecithin, forming lysolecithin, which is cytotoxic to erythrocytes and damages cell membranes. Formation of lysolecithin from the lecithin in bile may contribute to disruption of the pancreas and necrosis of surrounding fat. Phospholipase A₂ also liberates arachidonic acid, which is then converted to prostaglandins, leukotrienes, and other mediators of inflammation, contributing to coagulation necrosis. Pancreatic lipase, released as a direct result of pancreatic acinar cell damage, acts enzymatically on surrounding adipose tissue, causing the characteristic peripancreatic fat necrosis seen in severe acute pancreatitis (Figure 15–4).

Furthermore, trypsin and chymotrypsin activate kinins, complement, coagulation factors, and plasmin, leading to edema, inflammation, thrombosis, and hemorrhage within the gland. For example, trypsin activation of the kallikrein-kinin system leads to the release of bradykinin and kallidin, causing vasodilation, increased vascular permeability, edema, and inflammation (Figure 15–4), which all contribute to the systemic inflammatory response syndrome characteristic of acute pancreatitis. Circulating phospholipases interfere with the normal function of pulmonary surfactant, contributing to the development of an adult respiratory distress syndrome in some patients with acute pancreatitis. Elevated serum lipase levels are sometimes associated with fat necrosis outside of the abdomen.

Experimental models of acute pancreatitis suggest that NFκB activation occurs in parallel with trypsinogen activation. Pathologic Ca²⁺ influx appears to play a role in NFκB activation and may be a common activator of both parallel pathways of pancreatic injury. Moreover, as a downstream effector of CCK action, protein kinase C isoforms appear to play a role in NFκB activation, consistent with parallel activation of zymogen and NFκB pathways.

Finally, during acute pancreatitis, both the CC and CXC families of cytokines are implicated in the pathogenesis of the local and systemic inflammatory response. Cytokines and other inflammatory mediators such as tumor necrosis factor (TNF), interleukins (especially IL-1, IL-6, and IL-8), platelet-activating factor (PAF), and endotoxin are released rapidly and predictably from inflammatory cells. This release appears to be in response to the presence of active digestive enzymes, independent of the underlying cause. Production of cytokines during clinical pancreatitis begins shortly after pain onset and peaks 36–48 hours later. These agents are now thought to be principal mediators in the transformation of acute pancreatitis from a local inflammatory process to a systemic illness (Figure 15–5). The degree of TNF-induced inflammation correlates with the severity of pancreatitis. Cytokines rapidly enter the systemic circulation from the peritoneal cavity via the thoracic duct. In the systemic circulation, the cytokines affect many body systems and can produce the systemic inflammatory response syndrome (SIRS) and the multiorgan dysfunction syndrome typical of severe acute pancreatitis. Systemic complications of acute pancreatitis, such as respiratory failure, shock, and even multisystem organ failure, are accompanied by significant increases in monocyte secretion of TNF, IL-1, IL-6, and IL-8, and upregulation of the number of receptors for these cytokines on target cells. This finding suggests that TNF, IL-1, IL-6, and IL-8 play a central role in the pathophysiology of these manifestations.

Studies also suggest that substance P acting via neurokinin-1 (NK-1) receptors, PAF, and chemokines interacting with CCR1 receptors play important proinflammatory roles in determining the severity of acute pancreatitis. In particular, substance P and neurokinin-1 are involved in mediating acute lung injury. Substance P, a neuropeptide released from sensory afferent nerve endings, binds to the NK-1 receptor on the surface of effector cells and increases the permeability of vascular endothelium. The amount of substance P in the pancreas is increased during episodes of acute pancreatitis, and acinar cell expression of NK-1 receptors is markedly upregulated. Substance P appears to be a powerful proinflammatory mediator of both pancreatitis and associated lung injury. PAF also appears to play an important role in the development of pancreatitis and associated lung injury. Chemokines are chemoattractant cytokines that are involved in the activation and trafficking of various inflammatory cells. Chemokines acting via the chemokine receptor CCR1 appear to have a role in determining the severity of pancreatitis-associated lung injury but no effect on the severity of the pancreatitis itself. On the other hand, complement factor 5a (C5a) appears to act as an anti-inflammatory agent during the development of pancreatitis.

Various factors play active roles as proinflammatory or anti-inflammatory agents in acute pancreatitis. Drugs or other interventions to counteract those agents that are proinflammatory (eg, TNF, IL-1, IL-6, IL-8, and PAF) or to stimulate those that are anti-inflammatory (eg, IL-10) may eventually prove useful in treating patients with clinical pancreatitis to prevent severe injury to the pancreas and to prevent associated systemic manifestations, such as lung injury.

Clinical Manifestations

Acute pancreatitis may present in a highly variable manner, with the severity of inflammation and associated morbidity differing markedly among patients. Approximately 80% of patients experience a mild, self-limited illness to 2–3 days with no significant sequelae, but the remainder may develop severe acute pancreatitis, a life-threatening illness defined by the presence of associated organ system failure (typically pulmonary, cardiovascular, and/or renal systems). Acute pancreatitis may recur, depending primarily on its cause. With repeated attacks, the gland may eventually be permanently damaged, resulting in chronic pancreatitis or sometimes pancreatic insufficiency (see below). The distinction between acute pancreatitis and an acute exacerbation of chronic pancreatitis is determined by the clinical history and the characteristic finding on imaging of chronic pancreatitis. Acute and chronic pancreatitis have notably different management paradigms, so this distinction is important.

Recent consensus criteria require two of the following three criteria for the diagnosis of acute pancreatitis: abdominal pain, elevation of serum amylase or lipase (>3 times the upper limit of normal), and characteristic findings on computed tomography (CT) (or magnetic resonance imaging [MRI] or ultrasound). In practice, the first two elements are often present and suffice to make to the clinical diagnosis. Nonetheless, cross-sectional imaging (eg, contrast-enhanced CT may be useful in severe acute pancreatitis to assess the extent of associated pancreatic necrosis and other associated complications of the disease (Figure 15–6).

Signs and Symptoms at Presentation

Abdominal pain is nearly universal and a hallmark presentation of acute pancreatitis. In rare cases, patients may present with occult pancreatic inflammation evident by hyperamylasemia—eg, following pancreatic trauma, medication administration, or other known precipitants. However, such presentations are unlikely to be associated with clinically significant pancreatitis.

The pain of acute pancreatitis is characteristic, often described as an intense, deep, searing pain that radiates to the back. Frank peritoneal inflammation may lead to diagnostic confusion with other, more immediate surgical emergencies such as a perforated peptic ulcer, appendicitis, or diverticulitis.

The pain of acute pancreatitis is thought to derive in part from stretching of the pancreatic capsule by distended ductules and parenchymal edema, inflammatory exudate, digested proteins and lipids, and hemorrhage. In addition, these materials may seep out of the parenchyma into the retroperitoneum and lesser sac, where they irritate retroperitoneal and peritoneal sensory nerve endings and produce intense back and flank pain. The clinical findings of generalized peritonitis may follow.

Stretching of the pancreatic capsule may also produce nausea and vomiting. Increasing abdominal pain, peritoneal irritation, and electrolyte imbalance (especially hypokalemia) may cause a paralytic ileus with marked abdominal distention. If gastric motility is inhibited and the gastroesophageal sphincter is relaxed, there may be emesis. Both small and large bowel often dilate during an acute attack. Sometimes only a localized segment of bowel dilates. For example, there may be localized dilation of a segment of jejunum overlying the pancreas. In such cases, a plain x-ray film of the abdomen shows thickening of the valvulae conniventes and air-fluid levels (“sentinel loop”). In other cases, there may be segmental dilation of a portion of the overlying transverse colon. The x-ray film shows a sharply demarcated area of localized colonic dilation and edema (“colon cutoff sign”).

Almost two-thirds of patients with acute pancreatitis develop fever. The pathophysiologic mechanism responsible for fever involves the extensive tissue injury, inflammation, and necrosis and release of endogenous pyrogens, principally IL-1, from polymorphonuclear leukocytes into the circulation. In most cases of acute pancreatitis, fever does not indicate a bacterial infection. However, persistent fever beyond the fourth or fifth day of illness—or spiking temperatures to 40°C or more—may signify development of infectious complications such as infected peripancreatic fluid collections, infected pancreatic necrosis, or ascending cholangitis.

The cardinal laboratory finding in acute pancreatitis is elevation of the serum amylase, often up to 10- to 20-fold. The serum amylase elevation occurs almost immediately (within hours), but it usually returns to normal within 48–72 hours even if symptoms continue. The sensitivity of the serum amylase in acute pancreatitis is estimated to be 70–95%, meaning that 5–30% of patients with acute pancreatitis have normal or minimally elevated serum amylase values. The specificity of the test is considerably lower. Patients with marked (more than 3-fold) elevations of serum amylase usually have acute pancreatitis. Patients with lesser elevations of serum amylase often have one of a variety of other conditions.

The serum amylase concentration reflects the steady state between the rates of amylase entry into and removal from the blood. Hyperamylasemia can result from either an increased rate of entry or a decreased rate of metabolic clearance of amylase in the circulation. The pancreas and salivary glands have much higher concentrations of amylase than any other organs and probably contribute almost all of the serum amylase activity in healthy persons. Amylase of pancreatic origin can be now distinguished from that of salivary origin by a variety of techniques. Pancreatic hyperamylasemia results from injuries to the pancreas, ranging from minor (cannulation of the pancreatic duct) to severe (pancreatitis). In addition, injuries to the bowel wall (infarction or perforation) cause pancreatic hyperamylasemia as a result of enhanced absorption of amylase from the intestinal lumen. Salivary hyperamylasemia is observed in salivary gland diseases such as mumps parotitis but also (inexplicably) in a host of unrelated conditions such as chronic alcoholism, postoperative states (particularly after coronary artery bypass graft surgery), lactic acidosis, anorexia nervosa or bulimia nervosa, and certain malignancies. Hyperamylasemia can also result from decreased metabolic clearance of amylase caused by renal failure or macroamylasemia, a condition in which there are abnormally high-molecular-weight complexes of amylase bound to abnormal immunoglobulins in the serum.

Determination of serum lipase level is often helpful diagnostically. In acute pancreatitis, the serum lipase level is elevated, usually about 72 hours after onset of symptoms. The serum lipase measurement may be a better diagnostic test than serum amylase because it is just as easy to perform, may be more sensitive (85% vs. 79% sensitivity), is more specific for acute pancreatitis, and decreases to normal more slowly.

Early Complications of Acute Pancreatitis

Shock may occur in severe acute pancreatitis as a result of several interrelated factors. Hypovolemia results from massive exudation of plasma and hemorrhage into the retroperitoneal space and from accumulation of fluid in the gut as a result of ileus. Hypotension and shock may also result from release of kinins into the general circulation. For example, activation during acute inflammation of the proteolytic enzyme kallikrein results in peripheral vasodilation via liberation of the vasoactive peptides, bradykinin and kallidin. This vasodilation causes the pulse rate to rise and the blood pressure to fall. Cytokines like PAF, a very potent vasodilator and leukocyte activator, have been implicated in the development of shock and other manifestations of the SIRS. The contracted intravascular volume combined with the hypotension may lead to myocardial and cerebral ischemia, respiratory failure, metabolic acidosis, and decreased urinary output or renal failure as a result of acute tubular necrosis.

Tissue factor release and expression during proteolysis may cause activation of the plasma coagulation cascade and may lead to disseminated intravascular coagulation (DIC). In other cases, hypercoagulability of the blood is thought to be due to elevated concentrations of several coagulation factors, including factor VIII, fibrinogen, and perhaps factor V. Clinically affected patients may present with hemorrhagic discoloration (purpura) in the subcutaneous tissues around the umbilicus (Cullen sign) or in the flanks (Grey Turner sign). The splenic and portal veins are in close proximity to the pancreas and thus can become involved in the inflammatory process. Splenic vein thrombosis occurs in approximately 11% and portal vein thrombosis in approximately 2% of patients. Most thrombi are asymptomatic, but they may be associated with the development of venous hypertension and formation of varices over time.

Pulmonary complications are a dreaded manifestation of severe acute pancreatitis and occur in 15–50% of patients. The severity of pulmonary complications can vary from mild hypoxia to respiratory failure (acute respiratory distress syndrome [ARDS]). It is estimated that 50% of early deaths in patients with severe acute pancreatitis are associated with respiratory failure due to profound acute lung injury. The pathophysiology of this acute lung injury appears to involve an increase in permeability of the alveolar-capillary membrane. The endothelial cell destruction in the alveolar capillaries may be mediated by circulating activated pancreatic enzymes including elastase and phospholipase A₂. Pulmonary surfactant, another important alveolar barrier, appears to be destroyed by phospholipase A₂. Additional pulmonary injury appears to be mediated by inflammatory leukocytes that are sequestered in the alveoli and interstitial tissues, with subsequent release of proinflammatory cytokines and chemokines that lead to further tissue destruction. Elevated serum levels of IL-6 have been associated with the severity of lung injury in acute pancreatitis, an effect mediated by NFκB activation in pancreatic acinar cells. IL-6 and other inflammatory signaling pathways may prove to be appropriate therapeutic targets in severe acute pancreatitis, although to date no therapeutic agents have been found to be effective in clinical trials.

Acute pancreatitis may be accompanied by a small (usually left-sided) pleural effusion. The effusion may be reactive and hence secondary to a direct effect of the inflamed, swollen pancreas on the pleura abutting the diaphragm (typically transudative). Alternatively, in cases of severe acute pancreatitis, an effusion can be due to the tracking of exudative fluid from the pancreatic bed retroperitoneally into the pleural cavity through defects in the diaphragm. Characteristically, the pleural fluid in this latter circumstance is an exudate with high levels of protein, lactate dehydrogenase, and amylase. The effusion may contribute to segmental atelectasis of the lower lobes, leading to ventilation-perfusion mismatch and hypoxia.

Given the protean presentations of acute pancreatitis, there has been confusion regarding the classification of acute pancreatitis and any associated complications. Recent consensus guidelines have provided accurate criteria to aid in its diagnosis, treatment, and prognosis. The 2012 revision of the Atlanta classification represents the most recent standardized definitions for characterization of acute pancreatitis.

Acute pancreatitis is recognized to exist in two primary forms: interstitial edematous pancreatitis and necrotizing pancreatitis.

Interstitial edematous acute pancreatitis is characterized by enlargement of the pancreatic parenchyma with associated peripancreatic fluid but with uniform enhancement of the pancreatic parenchyma on contrast-enhanced CT. This form of the disease typically is clinically less severe, with symptoms routinely resolving within a week of presentation.

Necrotizing pancreatitis (necrosis of the pancreatic and peripancreatic tissues) occurs in approximately 5–10% of patients. While the degree of pancreatic necrosis is often detected by the lack of uniform parenchymal enhancement of contrast-enhanced CT, this process typically evolves over the first 1–2 weeks of illness, thereby rendering early imaging unreliable in predicting the severity of disease. The natural history of patients with necrotizing pancreatitis varies depending on whether pancreatic/peripancreatic necrosis remains solid or liquefies, becomes infected, persists, or resolves.

Infected pancreatic necrosis is a late complication of necrotizing pancreatitis. Rarely occurring in the first week of illness, procedures to diagnose this complication should be reserved for later in the patient’s clinical course. Infected pancreatic necrosis should be suspected when there is progressive clinical collapse with shock and end organ failure or failure to improve following initial stabilization. Infected pancreatic necrosis is suggested by the presence of pancreatic or peripancreatic necrosis with extraluminal gas on contrast-enhanced CT. However, it is important to document infected pancreatic necrosis by image-guided fine needle aspiration (percutaneous or endoscopic) and subsequent positive aspirate cultures because most pancreatic necrosis is in fact sterile. Infected pancreatic necrosis is a very serious complication of severe acute pancreatitis with a mortality rate of 25–50%. Consequently, it requires early pancreatic debridement.

Early complications of acute pancreatitis include both systemic and local problems. Systemic complications include the presence of organ failure, which defines severe acute pancreatitis. Organ failure may be transient (resolves within 48 hours) or persistent (affects prognosis). Early local complications of acute pancreatitis are defined by their presence within the first 4 weeks of disease onset. Acute peripancreatic fluid collections develop in the early phase of acute pancreatitis and may occur in the absence of pancreatic necrosis. On contrast-enhanced CT, these collections often have a poorly defined wall or boundary. These fluid collections are sterile and typically resolve without intervention. Acute necrotic collections occur in necrotizing pancreatitis and appear on contrast-enhanced CT as heterogeneous collections with variable amounts of fluid and solid debris. Sequential imaging studies may be necessary to define the evolution of these lesions. They may variably communicate with the pancreatic duct when necrosis is associated with ductal disruption, and they may become secondarily infected.

Late Complications of Acute Pancreatitis

Late complications of acute pancreatitis may similarly be divided into systemic and local effects. Systemic complications include persistent organ failure and need for prolonged intensive care, factors that portend a poor prognosis. Local complications are defined by their presence beyond 4 weeks from the onset of illness and are typically characterized by serial imaging that documents their evolution.

Pancreatic pseudocysts are nonepithelium-lined cavities that contain plasma, blood, pus, and pancreatic juice. They are the product of inflammatory fibrous or granulation tissue walling off a peripancreatic fluid collection. By definition, pseudocysts are distinguished from acute peripancreatic fluid collections by their persistence more than 4 weeks following an episode of acute pancreatitis. Pseudocysts generally occur after recovery from the acute attack and are the result of both parenchymal destruction and ductal obstruction or disruption. Some acini continue to secrete pancreatic juice, but because the juice cannot drain normally, it collects in an area of necrotic tissue, forming the ill-defined pseudocyst (Figure 15–7). As more juice is secreted, the cyst may grow progressively larger and may cause compression of nearby structures such as the portal vein (producing portal hypertension), common bile duct (producing jaundice or cholangitis), or gut (producing gastric outlet or bowel obstruction). Pancreatic pseudocysts are distinguished by their lack of solid debris, appearing as homogenous fluid-filled cavities on imaging studies.

Most pancreatic pseudocysts will resolve spontaneously and no specific intervention is required when asymptomatic. Indications for surgical, endoscopic, or percutaneous intervention include persistent symptoms or associated complications (obstruction of bowel or bile duct, hemorrhage, secondary infection). Treatment options for pseudocysts include external drainage either by surgical or percutaneous techniques or by internal drainage to the GI tract by surgical or endoscopic means. An infected pancreatic pseudocyst is typically designated a pancreatic abscess. As such, it is typically confined to a solitary cyst and occurs late in the course of disease. A pancreatic abscess can often be treated successfully. Percutaneous drainage is the mainstay of therapy, with surgical or endoscopic drainage reserved for refractory cases.

Walled-off necrosis is a mature, encapsulated collection of debris with a well-defined inflammatory rind that occurs beyond 4 weeks from onset of necrotizing pancreatitis. Sometimes, walled-off necrosis may be difficult to distinguish from a pancreatic pseudocyst, and contrast-enhanced CT may underestimate the amount of solid debris present in the walled-off necrosis. MRI and/or endoscopic ultrasound (EUS) may more reliably distinguish between these two entities and help define treatment strategies. In patients with persistent symptoms, with failure to improve clinically, or with secondary infection, intervention for walled-off necrosis may be necessary. Surgical necrosectomy, which may be done by open or minimally invasive techniques, or endoscopic transgastric necrosectomy may be considered.

Pancreatic ascites occurs when a direct connection develops between the pancreatic duct and the peritoneal cavity. Given its origin, it is not surprising that the ascitic fluid resembles pancreatic juice, characteristically an exudate with high protein and extremely high amylase levels. Left untreated, massive pancreatic ascites may lead to pleural effusions, subcutaneous fat necrosis, or abdominal compartment syndrome. Treatment typically involves drainage of the ascites and control of the pancreatic ductal disruption, either by endoscopic pancreatic duct stent placement or by surgical therapy.

Pancreatic fistulas, caused by disruption of the pancreatic duct, should be suspected in patients who develop pancreatic ascites or pleural effusions. Fistulas can be internal, connecting to pleural or pericardial spaces, colon, small intestine, or biliary tract, or external, draining through the skin.

Course & Prognosis

Most patients with acute pancreatitis recover completely with supportive medical management. The pancreas then regenerates and returns to normal except for some mild residual scarring. Diabetes mellitus almost never occurs after a single attack of pancreatitis, but both endocrine or exocrine insufficiency may occur following an episode of severe acute pancreatitis or repeated episodes of acute pancreatitis.

The initial course of alcoholic pancreatitis is characterized by recurrent acute exacerbations and the later course by progressive pancreatic insufficiency. However, among individuals with recurrent acute alcoholic pancreatitis, two groups can be distinguished in terms of prognosis. About 75% of these cases progress to advanced chronic pancreatitis, typically with pancreatic calcification and pancreatic insufficiency. The remainder do not progress and do not develop pancreatic duct dilation. The factors responsible for progression have not yet been elucidated.

The severity of acute pancreatitis can be estimated by various methods: clinical assessment, biochemical tests, peritoneal lavage, CT, and prognostic criteria (Table 15–2).

Studies have shown that important predictors of mortality are (1) failure of more than one organ system in the early phase of acute pancreatitis or (2) pancreatic necrosis associated with later development of multiple-organ failure. Organ failure may be defined by the modified Marshall scoring system, which includes assessment of respiratory failure (measured by PaO₂/FiO₂ ratio and need for supplemental oxygen), cardiovascular collapse (defined by systolic blood pressure, need for fluid resuscitation, and blood pH on arterial blood gases), and renal failure (defined by serum creatinine). Multiple-organ failure is defined as a syndrome of progressive but potentially reversible organ failure, involving two or more systems remote from the original insult. Persistent organ failure beyond 48 hours of presentation may be associated with a mortality as high as 36–50%.

Chronic Pancreatitis

Clinical Presentations

Chronic pancreatitis is a relapsing disorder causing severe abdominal pain, exocrine and endocrine pancreatic insufficiency, severe duct abnormalities, and pancreatic calcifications. The prevalence of the disorder is about 30 cases per 100,000 individuals, and the yearly incidence ranges from 3.5 to 10 cases per 100,000. In chronic pancreatitis, there is chronic inflammation of the parenchyma, leading to progressive destruction of the acini, stenosis and dilation of the ductules, and fibrosis of the gland. Eventually, there is impairment of the gland’s exocrine function (see Pancreatic Insufficiency later) and in severe cases loss of endocrine function as well (Chapter 18).

Etiology

It was at one time believed that chronic pancreatitis simply resulted from recurrent attacks of acute pancreatitis. However, there is some evidence that acute and chronic pancreatitis are distinct pathogenetic entities. Patients developing acute pancreatitis are a mean 13 years older than those with onset of chronic calcified pancreatitis. Furthermore, the two diseases have been linked to different causes. Finally, in acute pancreatitis, the pancreas is normal before the attack and the pathologic changes are completely reversible if the patient survives, whereas in chronic pancreatitis the gland is abnormal before the attack and the pathologic changes are not reversible.

The major cause of chronic pancreatitis is chronic alcoholism, which accounts for about 70–80% of cases. The remainder are due to diverse causes listed in Table 15–3. In 1788, Cawley first reported the association of alcoholism with chronic pancreatitis. He described a “free living young man” with diabetes and emaciation. At autopsy, his pancreas was “full of stones.” Patients with chronic pancreatitis resulting from alcohol abuse usually have a long history (6–12 years) of heavy alcohol consumption (150–175 g/d) before disease onset. In alcoholics, deficiencies of zinc and selenium may inhibit quenching of oxygen free radicals.

Recent epidemiologic evidence identifies cigarette smoking as a strong independent risk factor for the development of chronic pancreatitis. Moreover, tobacco exposure appears to have a dose-dependent relationship with its incidence. The number of daily cigarettes smoked as well as the duration of tobacco smoke exposure both appear to be important risk factors. Last, the combination of significant alcohol and cigarette use appears to be synergistic in augmenting the risk of chronic pancreatitis.

Long-term obstruction of the pancreatic duct can also cause chronic pancreatitis. The obstruction can be caused by a neoplasm, papillary stenosis, cystic lesions (cystic tumors or pseudocysts), scarring or stricture, or trauma. Pancreas divisum can cause chronic pancreatitis as a result of obstruction at the lesser papilla. Tropical chronic pancreatitis is a juvenile form of chronic calcific nonalcoholic pancreatitis, thought to be caused by protein or micronutrient deficiencies, which may cause impaired clearance of free radicals, or by ingestion of a toxic substance, such as cyanogens in cassava root. Chronic hypercalcemia may cause pancreatitis, as seen in 10–15% of patients with hyperparathyroidism. Intraductal precipitation of calcium and stimulation of pancreatic enzyme secretion are thought to be important in pathogenesis. In some cases of chronic pancreatitis with features of Sjögren syndrome, an autoimmune mechanism may be involved. Chronic hereditary pancreatitis, characterized by recurrent episodes of abdominal pain beginning in childhood, accounts for about 1% of cases. It is transmitted as an autosomal dominant genetic disorder with incomplete (~80%) penetrance. Hereditary chronic pancreatitis has also been associated with mutations in the cationic trypsinogen gene PRSS1 or in the SPINK1/PSTI gene (discussed previously). Some cases are due to cystic fibrosis (mucoviscidosis; see later). In some cases, no cause can be identified, and the disease is termed idiopathic chronic pancreatitis.

Pathology

Pathologically, chronic pancreatitis is characterized by scarring and shrinkage of the pancreas resulting from fibrosis and atrophy of acini and by stenosis and dilation of ductules. Grossly, the process usually involves the whole gland, but in about one-third of cases it is localized, most often involving the head and body of the gland. The ductules and ducts are often filled with inspissated secretions or calculi. Between 36% and 87% of patients with chronic pancreatitis have ductal stones. The gland may be rock hard as a result of diffuse sclerosis and calcification, and biopsy may be required to differentiate chronic pancreatitis from pancreatic carcinoma. Microscopically, there are loss of acini, dilation of ductules, marked fibrosis, and a lymphocytic infiltrate. The islets of Langerhans are usually well preserved.

In the early stage of chronic pancreatitis, pseudocysts are present in half (52%) of patients. A focally accentuated, perilobular fibrosis and a lesser degree of intralobular fibrosis are typically observed. Although intralobular fibrosis and perilobular fibrosis of the pancreas are hallmarks of alcoholic pancreatitis, they are also common among patients with alcohol dependence and abuse who have no history of pancreatitis. Marked fibrosis, ductal distortions, and the presence of intraductal calculi are the main features of advanced chronic pancreatitis. Pseudocysts occur less frequently (36%). CD4 and CD8 T lymphocytes are the predominant T-cell subsets in the inflammatory infiltrates in chronic pancreatitis.

In clinical practice, an important distinction must be made between patients with chronic pancreatitis who have “large duct” or “small duct” disease. The presence of a dilated main pancreatic duct, secondary to obstruction due to intraductal calculi and/or ductal strictures, is identified as large duct disease and is thought to produce symptoms of abdominal pain secondary to ductal hypertension. Such patients may be candidates for surgical decompression procedures, as described below. Patients with small duct disease tend to have small atrophic glands, often riddled with calcifications but without focal ductal abnormalities or dilatation. The pain syndrome in patients with small duct disease is attributed to local enzymatic activity and destruction of the perineural sheath, exposing axons to cytokines released by inflammatory cells and ultimately causing perineural fibrosis.

Pathogenesis

Table 15–4 presents a classification of pancreatitis based on pathogenesis, emphasizing the fundamental differences between acute and chronic pancreatitis. In Table 15–5, proposed pathogenetic mechanisms for chronic pancreatitis are listed, again emphasizing the differences between large duct and small duct pathologies and their associated causes.

As with acute pancreatitis, increasing understanding of genetic profiles of patients with chronic pancreatitis and more sophisticated knowledge of cell-signaling pathways have led to the appreciation of chronic pancreatitis as a complex genetic disorder. While a minority of patients have Mendelian disorders with single mutations that lead to pancreatitis (eg, hereditary pancreatitis, cystic fibrosis), the majority of patients likely have genetic susceptibilities that interact with environmental exposures to produce the clinical syndrome. At least five genes conveying susceptibility to pancreatitis have been identified, including variants in the cationic trypsinogen gene (PRSS1), the cystic fibrosis transmembrane conductance regular gene (CFTR), the pancreatic secretory trypsin inhibitor gene (SPINK1), the chymotrypsinogen C gene (CTRC), and the calcium-sensing receptor gene (CASR). Evidence suggests that these genes interact with each other as well as with environmental (eg, alcohol and tobacco) exposures in heterogeneous ways.

Mutations of the CFTR gene located on chromosome 7q32 are arguably the most well understood of the genetic susceptibilities to pancreatitis. In cystic fibrosis patients with chronic pancreatitis, mutation of the CFTR gene causes inadequate function of CFTR, the chloride channel located on the luminal surface of the pancreatic duct cell that is highly involved in bicarbonate secretion. Major mutations in both alleles lead to loss of CFTR function and the inability to hydrate mucus, resulting in inspissated secretions and obstruction of the ducts. Because pancreatic function may be maintained with CFTR function as little as 1% of normal, only severe CFTR mutations yielding little or no functional protein produce chronic pancreatitis and pancreatic insufficiency.

Chronic pancreatitis appears to occur in the context of one of several pathogenic pathways. In patients with large duct obstruction, the ductal lesion likely predates the development of pancreatic parenchymal abnormalities. The pathogenesis probably involves elevated pressures in the pancreatic duct, resulting in ischemia, necrosis, and inflammation of acinar cells. However, the ductal epithelium is preserved. Calcified protein plugs and stones are less often present, although some patients with lithogenic pancreatitis may develop secondary ductal obstruction and large duct disease over time. Many patients with idiopathic chronic pancreatitis also have ductal hypertension.

For chronic lithogenic pancreatitis, several different pathogenetic mechanisms have been postulated. One theory postulates acinar protein (trypsinogen) hypersecretion as an initial event (Figure 15–8A). Ultrastructural studies of exocrine pancreatic tissue from patients with chronic pancreatitis show signs of protein hypersecretion, including a larger diameter of cells, nuclei, and nucleoli; increased length of the endoplasmic reticulum; increased numbers of condensing vacuoles; and decreased numbers of zymogen granules. The hypersecretion of protein occurs without increased fluid or bicarbonate secretion by ductal cells. At the same time, there is an increase in the ratio of lysosomal hydrolases (cathepsin B) to digestive hydrolases (trypsinogen), resulting in activation of trypsinogen. Precipitation of intraductal protein into plugs is then thought to occur in the following fashion: Lithostathines (formerly called pancreatic stone proteins, or PSPs) are peptides secreted into pancreatic juice that normally inhibit the formation of protein plugs and the aggregation of calcium carbonate crystals to form stones. Acinar cell secretion of lithostathine is impaired by alcohol. Furthermore, when hydrolyzed by trypsin and cathepsin B, lithostathine H2/PSP-S1 is created. This insoluble peptide polymerizes into fibrils that form the matrix of protein plugs. At the same time, there is hypersecretion of calcium into the pancreatic juice. The calcium hypersecretion is first triggered by neural (cholinergic, vagally mediated) or hormonal stimuli. Later, as the basal lamina of the pancreatic duct is eroded by contact with the protein plugs, there is transudation of serum protein and calcium into the pancreatic juice. The combination of protein plug formation in pancreatic juice that is thick, viscid, and protein rich and supersaturated with calcium carbonate leads to formation of calculi (stones) (Figure 15–8B). Lithostathine deficiency is unexplained but may be hereditary or acquired. Chronic alcoholism and malnutrition are acquired causes of lithostathine deficiency. Decreased levels of other nucleation-inhibitory factors, such as local trypsin inhibitor and citrate, in pancreatic juice further enhance formation of pancreatic plugs and stones. Lactoferrin, an iron-containing macromolecular protein, is elevated in the pancreatic secretions of alcoholic patients with pancreatitis. Lactoferrin can produce aggregation of large acidophilic proteins, such as albumin, and thus may be partly responsible for the formation of protein plugs. Similarly, GP2, a glycosylphosphatidylinositol-anchored protein, might have a role in protein plug formation. GP2 is released from the apical surface of acinar cells into the pancreatic ducts in relatively high concentrations. GP2 aggregates at pH < 7.0, and pancreatic juice from patients with chronic pancreatitis usually has a pH < 7.0. Eventually, the stones provoke formation of fibrotic ductal strictures and ductal ectasia, acinar cell atrophy, and parenchymal atrophy distal to obstructed ducts in the advanced stages of chronic pancreatitis.

Another theory postulates a necrosis-fibrosis sequence, in which focal necrosis during recurrent attacks of acute pancreatitis induces scarring and fibrosis, leading to chronic lithogenic pancreatitis (Figure 15–9A). In this scenario, vascular damage in acute pancreatitis causes cellular anoxia, necrosis, chronic inflammation, and subsequent fibrosis. In particular, periacinar and periductal fat necrosis induce periductal fibrosis, which partially obstructs the interlobular ducts. Stasis within the ductules then leads to protein plug and stone formation in the pancreatic juice (Figure 15–9B). Subsequently, total obstruction of ducts by calculi induces acinar cell necrosis, inflammation, and fibrosis (Figure 15–9C). Transforming growth factor-β (TGF-β) appears to be a mediator of collagen synthesis after pancreatic injury.

Pathophysiology

Maldigestion in chronic pancreatitis results from several factors. Long-standing inflammation and fibrosis of the pancreas can destroy exocrine tissue, leading to inadequate delivery of digestive enzymes to the duodenum in the prandial and postprandial periods. This maldigestion is worsened by inadequate delivery of bicarbonate to the duodenum, with consequent gastric acid inactivation of enzymes and bile acids. Gastric dysmotility and mechanical obstruction from fibrosis in the pancreatic head may also contribute. Chronic pancreatitis may thus result in the profound steatorrhea of pancreatic insufficiency. There is a direct correlation between severity of histologic findings and exocrine pancreatic dysfunction as estimated by the CCK-secretin test (see later).

Studies of patients with chronic pancreatitis have found no abnormalities in basal plasma levels of CCK and pancreatic polypeptide (PP), but impaired interdigestive cycling and postprandial release of CCK and PP have been noted. Chronic pancreatitis does not seem to have any effect on intestinal motility.

In chronic pancreatitis, fecal bile acid excretion has been found to be three times that of healthy individuals. Bile acid malabsorption is related to impairment of pancreatic bicarbonate secretion; it is generally not observed until bicarbonate output is markedly reduced (<0.05 mEq/kg/h). Such bile acid malabsorption may cause the hypocholesterolemia seen in patients with chronic pancreatitis.

Impairment of exocrine function in chronic pancreatitis may also lead to increased CCK-mediated stimulation of the pancreas.

Hepatic insulin resistance has been demonstrated in patients with chronic pancreatitis, perhaps related to a decrease in high-affinity insulin receptors on the hepatocyte cell membrane. In rats, insulin binding improves after administration of pancreatic polypeptide.

Clinical Manifestations

The clinical manifestations of chronic pancreatitis are listed in Table 15–6. The major symptom of chronic pancreatitis is severe abdominal pain that can be either constant or intermittent. The abdominal pain often radiates to the midback or scapula and increases after eating. The pain of chronic pancreatitis is multifactorial, likely reflecting pancreatic ductal hypertension (eg, in patients with large duct disease) as well as chronic inflammatory neural injury (eg, in small duct disease). Patients may have recurrent attacks of severe abdominal pain, vomiting, and elevation of serum amylase (chronic relapsing pancreatitis). Continued alcohol intake may increase the frequency of painful episodes, at least when there is still relatively preserved pancreatic function; in severe pancreatic insufficiency, alcohol intake appears to have less influence on the development of abdominal pain. Pancreatic parenchymal pressure measurements have not been found to correlate with pain. From 10% to 20% of patients have “painless pancreatitis,” presenting with diabetes, jaundice, maldigestion, malabsorption, or steatorrhea. Anorexia and weight loss occur frequently, related to both poor nutrition and malabsorption from pancreatic insufficiency.

The diagnosis of chronic pancreatitis is based mainly on symptoms and signs. The serum amylase and lipase levels are elevated in only a minority of cases. In the remaining cases, the amylase and lipase levels are normal or low, probably because there is little residual functional pancreatic tissue and true acute inflammation is rare. Pancreatic parenchymal and main duct calcifications seen on CT or plain-film x-rays are pathognomonic of chronic pancreatitis. The calcifications are actually the intraductal pancreatic calculi composed of calcium carbonate and lithostathines. Pseudocyst formation may be evident on CT imaging as well.

EUS has become the test of choice for evaluation of patients with early or mild chronic pancreatitis. Studies correlating histologic findings with EUS scoring of chronic pancreatitis changes have confirmed the excellent sensitivity (85–91%) and specificity (70–86%) of EUS. The value of EUS is most apparent in patients without calcific disease, because those patients can have a definitive diagnosis made on CT and because they often have more severe or long-standing symptoms. A consensus conference established the Rosemont criteria as a scoring system composed of major and minor parenchymal and ductal features that has provided standardized criteria for diagnosing chronic pancreatitis.

About 5% of patients develop severe sclerosing pancreatitis involving the head of the pancreas, leading to obstruction of the common bile and pancreatic ducts. Obstruction of the common bile duct in the setting of chronic pancreatitis typically appears as a smooth, tapering stricture, rather than an abrupt cutoff, as is seen in bile duct obstruction due to pancreatic cancer. Obstruction may also be caused by a pseudocyst in the head of the pancreas. Common bile duct obstruction results in profound and persistent jaundice, resembling that produced by pancreatic carcinoma. The serum bilirubin and alkaline phosphatase are elevated.

ERCP is the best imaging procedure for assessing the severity and extent of ductal changes. ERCP findings include dilated ducts, frequently with adjacent areas of stricture, yielding a “chain of lakes” or “string of pearls” appearance, or ducts of normal caliber, with adjacent small ducts lacking side branches, yielding a “tree in winter” appearance. ERCP and MRCP, alternative imaging techniques that provide visualization of the pancreatic ductal system, may be used as confirmatory tests when EUS is not definitive or when specific focal abnormalities correlate with clinical symptoms (such as biliary obstruction or pancreatic ductal disruption).

Failure to secrete pancreatic juice results in malabsorption of fat (steatorrhea) and fat-soluble vitamins, leading to weight loss. Impairment of exocrine function is manifested by pancreatic insufficiency (see later). Studies screening patients with chronic pancreatitis have found that the majority develop exocrine dysfunction over time. One study documented that 63% developed exocrine dysfunction within 5 years and 94% after 10 years. Diabetes mellitus is a late complication of chronic pancreatitis and is not apparent until 80–90% of the gland is severely damaged.

The treatment of chronic pancreatitis is mainly symptomatic and directed toward relieving pain and treating exocrine and endocrine insufficiency (see below). Pain in these patients is often a serious clinical problem, leading to a significant compromise of quality of life and potential opioid tolerance and even addiction. If a precipitating factor such as an anatomic abnormality or metabolic condition is present, it may be treated with surgical or medical intervention. Methods of pain relief include abstinence from alcohol and use of conventional analgesics. If pain is not relieved, the use of opioids may be necessary. Invasive procedures, such as celiac plexus block, endoscopic procedures, and surgical drainage or resection may be indicated in select patients with debilitating symptoms.

The major complications of chronic pancreatitis are pseudocyst formation and mechanical obstruction of the common bile duct and duodenum. Less common complications include pancreatic fistulas with pancreatic ascites, pleural effusion, or sometimes pericardial effusion; splenic vein thrombosis and development of gastric varices; and formation of a pseudoaneurysm, with hemorrhage or pain resulting from expansion and pressure on adjacent structures. Fistulas result from disruption of the pancreatic duct. Splenic vein thrombosis occurs because the splenic vein, which courses along the posterior surface of the pancreas, may become involved in peripancreatic inflammation. Pseudoaneurysms may affect any of the arteries in proximity to the pancreas, most commonly the splenic, hepatic, gastroduodenal, and pancreaticoduodenal arteries.

In patients monitored for more than 10 years, the mortality rate is 22%; pancreatitis-induced complications account for 13% of the deaths. Older age at diagnosis, cigarette smoking, and alcohol intake are major predictors of mortality among individuals with chronic pancreatitis. Chronic pancreatitis of any cause has been associated with a 25-year cumulative risk of approximately 4% for the development of pancreatic cancer.

Pancreatic Insufficiency

Clinical Presentations

Pancreatic exocrine insufficiency is the syndrome of maldigestion resulting from disorders interfering with effective pancreatic enzyme activity. Because pancreatic lipase is essential for fat digestion, its absence leads to steatorrhea (the occurrence of greasy, bulky, light-colored stools). On the other hand, although pancreatic amylase and trypsin are important for carbohydrate and protein digestion, other enzymes in gastric and intestinal juice can usually compensate for their loss. Thus, patients with pancreatic insufficiency seldom present with maldigestion of carbohydrate and protein (nitrogen loss).

Etiology

Pancreatic insufficiency usually results from chronic pancreatitis in adults or cystic fibrosis (mucoviscidosis) in children (Table 15–7). In some cases, it is a consequence of pancreatic resection or carcinoma of the pancreas. Pancreatic insufficiency occurs after bone marrow transplantation and appears to be related to prior acute or chronic graft-versus-host disease. Each of these conditions markedly reduces the amount of pancreatic enzymes secreted, often to less than 5% of normal.

Pancreatic exocrine insufficiency is also a common occurrence in patients recovering from severe acute pancreatitis, and its severity correlates with the extent of pancreatic necrosis. Its severity also correlates with the severity of concomitant endocrine insufficiency, manifested by the new onset of diabetes mellitus.

Less commonly, pancreatic insufficiency results from disease states that cause hypersecretion of gastric acid. For example, excessive gastrin secretion from a gastrinoma (an islet cell neoplasm composed of G cells) leads to continuous hypersecretion of gastric acid and a very low pH of gastric juice. In affected patients, the excess gastric acid overwhelms the normal pancreatic bicarbonate production and results in an abnormally acidic pH in the duodenum. This acid pH, in turn, causes decreased activity of otherwise adequate amounts of pancreatic enzymes.

Pathology & Pathogenesis

Normally, the activities of the various pancreatic enzymes decrease during their passage from the duodenum to the terminal ileum. However, the degradation rates of individual enzymes vary; lipase activity is lost rapidly and protease and amylase activity is lost slowly. Lipase activity is usually destroyed by proteolysis, mainly by the action of residual chymotrypsin. This mechanism persists in patients with pancreatic insufficiency, helping to explain why fat malabsorption develops earlier than protein or starch malabsorption.

Patients with destruction of the exocrine pancreas develop impaired digestion and absorption of fat. Clinically, fat malabsorption is manifested as steatorrhea. Although the steatorrhea is caused mostly by the deficiency of pancreatic lipase, the absence of pancreatic bicarbonate secretion also contributes to its occurrence. Without bicarbonate, acidic chyme from the stomach inhibits the activity of pancreatic lipase and causes precipitation of bile salts. Deficiency of bile salts in turn causes failure of micelle formation and interference with fat absorption.

Pathophysiology

Causes of maldigestion from exocrine pancreatic insufficiency include chronic pancreatitis, cystic fibrosis, pancreatic cancer, partial or total gastrectomy, and pancreatic resection. Each of these causes is associated with specific related changes in GI physiology, including changes in intraluminal pH, bile acid metabolism, gastric emptying, and intestinal motility.

For example, during the course of chronic pancreatitis, there is a close relationship among gastric acidity, exocrine pancreatic insufficiency, and impaired digestion. Postprandial gastric acidification has been found to be significantly greater among patients with severe pancreatic insufficiency than among those with mild or no insufficiency. Inhibition of gastric acid secretion by H₂ blockers such as cimetidine or proton pump inhibitors such as omeprazole improves the response to pancreatic enzyme replacement and decreases fecal fat excretion. However, it does not lead to complete elimination of steatorrhea.

On the other hand, loss of the stomach can cause considerable change in function of the exocrine pancreas. After total gastrectomy, patients frequently develop severe primary exocrine pancreatic insufficiency with maldigestion and weight loss. Postoperatively, pancreatic juice volume, bicarbonate output, and enzyme (amylase, trypsin, and chymotrypsin) secretion are reduced significantly compared with preoperative levels. These reductions probably result from changes in GI hormone secretion, altering regulation of pancreatic function. For example, after gastrectomy, most patients exhibit decreased baseline and postprandial gastrin and pancreatic polypeptide secretion and increased postprandial CCK secretion.

Clinical Manifestations

The symptoms and signs exhibited by patients with pancreatic insufficiency (Table 15–8) vary to some extent with the underlying disease.

Steatorrhea

Patients with steatorrhea usually describe their stools as voluminous or bulky, foul-smelling, greasy, frothy, pale yellow, and floating. However, significant steatorrhea may occur without any of these characteristics. A 24-hour quantitative fecal fat test showing excretion of more than 6 g is necessary for the definitive diagnosis of steatorrhea. Steatorrhea responds, often dramatically, to treatment with oral pancreatic enzymes, ingested with each meal and with snacks. In severe cases of fat malabsorption, deficiencies of the fat-soluble vitamins (vitamins A, D, E, and K) may occur and require parenteral supplementation.

Diarrhea

In patients with fat malabsorption, diarrhea may result from the cathartic action of hydroxylated fatty acids. These fatty acids inhibit the absorption of sodium and water by the colon. Less commonly, watery diarrhea, abdominal cramping, and bloating are due to carbohydrate malabsorption. Indeed, because salivary amylase production remains undisturbed and because pancreatic amylase production must be markedly reduced before intraluminal starch digestion is slowed, symptomatic carbohydrate malabsorption is uncommon in pancreatic insufficiency.

Hypocalcemia

Hypocalcemia, hypophosphatemia, tetany, osteomalacia, osteopenia (low bone mineral density), and osteoporosis can occur both from deficiency of the fat-soluble vitamin D and from the binding of dietary calcium to unabsorbed fatty acids, forming insoluble calcium-fat complexes (soaps) in the gut.

Nephrolithiasis

The formation of insoluble calcium soaps in the gut also prevents the normal binding of dietary oxalate to calcium. Dietary oxalate remains in solution and is absorbed from the colon, causing hyperoxaluria and predisposing to nephrolithiasis.

Vitamin B12 Deficiency

About 40% of patients with pancreatic insufficiency demonstrate malabsorption of vitamin B₁₂ (cobalamin), although clinical manifestations of vitamin B₁₂ deficiency are rare (anemia, subacute combined degeneration of the spinal cord, and dementia). The malabsorption of vitamin B₁₂ appears to result from reduced degradation by pancreatic proteases of the normal complexes of vitamin B₁₂ and its binding protein (R protein), resulting in less free vitamin B₁₂ to bind to intrinsic factor in the small intestine.

Weight Loss

Long-standing malabsorption leads to protein catabolism and consequent weight loss, muscle wasting, fatigue, and edema. At times weight loss occurs in patients with chronic pancreatitis because eating exacerbates their abdominal pain or because narcotics used to control pain cause anorexia. In patients who develop diabetes mellitus, weight loss may be due to glycosuria.

Laboratory Tests & Evaluation

Because there is a direct correlation between duodenal (and therefore fecal) output of elastase 1 and duodenal output of lipase, amylase, trypsin, and bicarbonate, measurement of fecal elastase concentrations has been used as a screening test for exocrine pancreatic insufficiency. The diagnosis of pancreatic insufficiency is enhanced by several additional noninvasive tests of exocrine pancreatic function. These tests include the bentiromide test, pancreolauryl test, and cholesteryl-[¹⁴C]octanoate breath test. In these tests, substrates for pancreatic digestive enzymes are administered orally and their products of digestion are measured. In the bentiromide test, N-benzoyl-L-tyrosine-p-aminobenzoic acid is administered as a substrate for chymotrypsin. Enzymatic cleavage yields p-aminobenzoic acid, which is absorbed from the gut and measured in the urine. In the pancreolauryl test, fluorescein dilaurate is administered and pancreatic esterases release fluorescein, which is then absorbed and measured in the urine. The cholesteryl-[¹⁴C]octanoate breath test measures ¹⁴CO₂ output in the breath at 120 minutes after ingestion, allowing rapid detection of pancreatic exocrine insufficiency. Patients with chronic pancreatitis have markedly diminished excretion of p-aminobenzoic acid or fluorescein in the urine or output of ¹⁴CO₂ in the breath. In clinical practice, steatorrhea and associated weight loss are the most common and striking signs of exocrine pancreatic insufficiency. Therefore, providers must document and treat steatorrhea prior to proceeding with more specialized diagnostic testing.

Carcinoma of the Pancreas

Epidemiology & Etiology

Pancreatic carcinoma has become the fourth leading cause of cancer-related death in the Unites States, with an annual incidence and mortality approaching 40,000 cases per year. Delay in diagnosis, relative resistance to chemotherapy and radiation, and intrinsic biological aggressiveness manifested by early metastatic disease all contribute to the abysmal prognosis associated with pancreatic adenocarcinoma. Pancreatic cancer usually occurs after age 50 years and increases in incidence with age, with most patients diagnosed between 60 and 80 years of age. It is somewhat more frequent in men than in women. Autopsy series document that pancreatic cancer has been identified in up to 2% of individuals undergoing a postmortem examination. Despite advances in expanded awareness and understanding of the disease, diagnostic procedures, and surgical and medical therapies, overall 5-year survival for pancreatic adenocarcinoma remains approximately 5%.

Many risk factors for pancreatic adenocarcinoma have been identified. Cigarette smoking has the strongest overall association and is thought to account for one-quarter of cases diagnosed. The association between cigarette smoking and pancreatic cancer is thought to be related to N-nitroso compounds present in cigarette smoke. Exposure to these agents leads to pancreatic ductal hyperplasia, a possible precursor to adenocarcinoma.

Other factors associated with an increased risk of pancreatic adenocarcinoma include a high dietary intake of saturated fat, exposure to nonchlorinated solvents, and the pesticide dichlorodiphenyl trichloroethane (DDT), although the overall contribution of these factors is likely small. Diabetes mellitus has also recently been identified as a risk factor for the disease. Chronic pancreatitis increases the risk of developing pancreatic adenocarcinoma by 10- to 20-fold. The role of other dietary factors (coffee, high fat intake, and alcohol use) is much debated. Diets containing fresh fruits and vegetables are thought to be protective. There is an increased incidence of pancreatic cancer among patients with hereditary pancreatitis, particularly among those who develop pancreatic calcifications. Rarely, pancreatic carcinoma is inherited in an autosomal dominant fashion in association with diabetes mellitus and exocrine pancreatic insufficiency. A genetic predisposition has also been identified in a number of familial cancer syndromes, including the syndromes listed in Table 15–9. A number of genes linked with the familial syndromic and sporadic pancreatic cancer have been described. However, the penetrance of the disease in gene carriers is highly variable, and individual gene mutations have been variably linked to pancreatic oncogenesis. Importantly, the vast majority of pancreatic adenocarcinoma patients develop the disease without any identified genetic mutation or putative or established risk factor.

Pathology

Carcinomas occur more often in the head (70%) and body (20%) than in the tail (10%) of the pancreas. Grossly, pancreatic cancer presents as a profoundly desmoplastic, infiltrating tumor that obstructs the pancreatic duct and thus often causes fibrosis and atrophy of the distal gland. Carcinomas of the head of the pancreas tend to obstruct the common bile duct early in their course, with resulting jaundice, and can extend into the uncinate process to involve the superior mesenteric artery and vein, thus compromising surgical resectability. Tumors of the body and tail tend to present later in their course, as they cause few symptoms until they become quite large.

Microscopically, 90% of pancreatic cancers are adenocarcinomas; the remainder are adenosquamous, anaplastic, and acinar cell carcinomas. Pancreatic cancer tends to spread into surrounding tissues, invading neighboring organs along the perineural fascia, causing severe pain, and via the lymphatics and bloodstream, causing metastases in regional lymph nodes, liver, and other more distant sites (Figure 15–10).

Pancreatic adenocarcinomas consist of multiple cell types that each contribute to the clinical behavior of the disease. While mature cells in various stages of differentiation constitute the majority of the cellular elements, a small proportion of cancer stem cells account for the resistance to chemotherapy and radiation that is often characteristic of pancreatic cancer. Finally, pancreatic adenocarcinomas often have dense desmoplastic stromal elements that account for the tumor’s infiltrative and fibrotic nature.

Pathogenesis

As with other epithelial malignancies, pancreatic adenocarcinoma appears to develop through a series of progressive genetic mutations within the pancreatic ductal epithelium (Figure 15–11). These sequential genetic and epigenetic events correlate with the evolution from premalignant ductal lesions to invasive carcinoma. Pancreatic intraepithelial neoplasia (PanIN) is the most well characterized precursor to pancreatic adenocarcinoma. The evolution from minimal dysplasia (PanIN 1a and b) to severe dysplasia (PanIN 2 and 3) to adenocarcinoma appears to track with stepwise accumulation of genetic mutations that include the activation of the K-ras2 oncogene, inactivation of the tumor suppressor gene CDKN2a/INK4a, and finally, inactivation of the tumor suppressor genes TP53 and DPC4/SMaD4. Other precursor lesions of pancreatic adenocarcinoma likely exist in the form of mucin-producing pancreatic cystic neoplasms such as intraductal papillary mucinous neoplasms and mucinous cystic neoplasms.

Invasive pancreatic adenocarcinomas usually have one or more characteristic genetic mutations. Activating point mutations in the proto-oncogene K-ras at codon 12 have been identified in more than 90% of pancreatic cancers. Mutation in the TP53 tumor suppressor gene has been detected in 50–75% of adenocarcinomas of the pancreas. Concurrent loss of TP53 and K-ras function may contribute to the clinical aggressiveness of the cancer. In addition, in approximately 90% of cases, the P16 tumor-suppressor gene, located on chromosome 9p, is inactivated. DPC4 deletion is present in up to 50% of pancreatic adenocarcinomas and has been associated with increased metastatic potential.

Despite these prevalent mutations, comprehensive genomic analysis of human pancreatic cancer specimens has revealed tremendous genetic heterogeneity. Point mutations occur in numerous cellular pathways associated with neoplastic behavior, but few tumors share the same mutations or have defects in all pathways. Unfortunately, few targets susceptible to currently available drugs have been identified. Analyses of pancreatic cancer metastases have also revealed that the cellular clones that give rise to metastatic lesions may be distinct from the genetic fingerprint of the primary tumor. Although these characteristics complicate pancreatic cancer treatment, recent studies have attempted to identify tumor subtypes that differ in their response to differing chemotherapy regimens, potentially facilitating a future customized treatment regimen for individual tumor genotypes.

Mutations in DNA mismatch repair genes can also lead to pancreatic cancer. It appears that multiple mutations must occur for pancreatic cancer to develop. Familial pancreatic cancer syndromes arise from germline mutations. Examples include mutations in STK11 in Peutz-Jeghers syndrome and in DNA mismatch repair genes. The mismatch repair gene BRCA2 is inactivated in approximately 7–10% of pancreatic cancers. Familial syndromes and genetic alterations related to pancreatic cancer are summarized in Table 15–9. A 2012 consensus conference defined a group of high-risk individuals deemed appropriate for pancreatic cancer screening: first-degree relatives of patients with pancreatic cancer from a familial kindred (at least two affected first-degree relatives); patients with Peutz-Jeghers syndrome; and p16, BRCA2, and hereditary non-polyposis colorectal cancer (HNPCC) mutation carriers with one or more affected first-degree relative(s).

The tumor microenvironment (internal and surrounding stromal elements of pancreatic adenocarcinoma) is increasingly recognized both as central to the pathogenesis of the disease and as a potential target for therapy. Pancreatic stellate cells (myofibroblasts) that are responsible for stromal growth and turnover express growth factors and other peptides that may be associated with tumor behavior and prognosis.

In chronic pancreatitis, a common pathway for the development of pancreatic cancer may be through the chronic inflammatory process, including a pronounced stromal reaction. Mediators of chronic inflammation in the stroma likely support a transformation to malignancy, although the exact mechanisms remain unknown. Cytokines produced by the activated stroma appear to promote the aggressive behavior of pancreatic cancer cells.

Clinical Manifestations

The clinical presentation of pancreatic cancer may occasionally be indistinguishable from that of chronic pancreatitis, in part because inflammatory changes commonly occur in both chronic pancreatitis and pancreatic adenocarcinoma. The clinical manifestations (Table 15–10) of pancreatic cancer vary with location and histologic tumor type.

Patients with carcinoma of the head of the pancreas usually present with painless, progressive jaundice resulting from common bile duct obstruction (Figure 15–10). Sometimes the obstruction caused by carcinoma in the head of the pancreas is signaled by the presence of both jaundice and a dilated gallbladder palpable in the right upper quadrant (Courvoisier law). Patients with carcinoma of the body or tail of the pancreas usually present with epigastric abdominal pain, profound weight loss, abdominal mass, and anemia. These patients usually present at later stages and often have distant metastases, particularly in the liver. Splenic vein thrombosis may occur as a complication of cancers in the body or tail of the gland.

About 70% of patients with pancreatic cancer have impaired glucose tolerance or frank diabetes mellitus. While this may be due to proximal ductal obstruction and atrophy of the distal gland, some patients appear to have resolution of impaired glucose tolerance or diabetes with surgical resection, suggesting that pancreatic cancers elaborate a yet unidentified diabetogenic substance.

A variety of tumor markers, such as carcinoembryonic antigen (CEA), CA 19-9, α-fetoprotein, pancreatic oncofetal antigen, and galactosyl transferase II, can be found in the serum of patients with pancreatic cancer. However, none of these tumor markers have sufficient specificity or predictive value to be useful in screening for the disease. CA 19-9 may be useful to predict recurrence in patients following surgical resection or to follow disease burden in patients who are being treated with systemic chemotherapy.

In evaluating patients suspected of having pancreatic cancer, the initial diagnostic test of choice is a contrast-enhanced, thin-cut helical CT scan. For patients with an inconclusive CT scan, or in cases where a tissue diagnosis is needed, EUS with fine needle aspiration may aid in diagnosis. Endoscopic retrograde cholangiography (ERC) with endobiliary stent placement is typically used to palliate obstructive jaundice when present. In patients with pancreatic head lesions, brushing of the biliary or pancreatic duct during ERCP may confirm the diagnosis of pancreatic adenocarcinoma. In addition to aiding in diagnosis, helical CT is useful for delineating the regional vascular anatomy and to look for major vascular invasion by tumor, a sign of unresectability, or to determine the presence of metastatic disease.

Treatment with curative intent for pancreatic adenocarcinoma involves a multidisciplinary approach of surgical resection, systemic chemotherapy, and radiation therapy. Unfortunately, only 15–20% of patients are eligible for treatment with curative intent; all other patients with unresectable locally advanced pancreatic cancer and/or metastatic disease are candidates for palliative chemotherapy with only limited survival benefit. Advances in surgical strategies such as vascular resection and reconstruction, resection in elderly patients, minimally invasive pancreatectomy, and neoadjuvant chemoradiation regimens have all attempted to expand the population of patients eligible for surgical resection. However, the invasive growth behavior of pancreatic cancer into perineural and retroperitoneal tissues often makes achieving a negative microscopic margin challenging, and operations that leave even microscopic disease behind afford no real chance for long-term survival.

Of the patients eligible for surgical resection, overall 5-year survival rate is approximately 20%, while select patients with small tumors, negative lymph nodes, and a negative microscopic margin have a slightly better prognosis. Patients with unresectable locally advanced disease may survive 12–24 months with modern multimodality palliative regimens. Patients with metastatic disease at presentation have a median survival of 6 months or less. These ominous outcomes clearly indicate the need for improved treatment strategies. Given the significant expansion in the understanding of the genetic characteristics and cellular compartments of pancreatic adenocarcinoma, there is now more optimism that targeted agents and personalized treatment strategies will ultimately lead to improved survival for patients stricken with this aggressive disease.

Yeong Kwok, MD

Case 74

Acute Pancreatitis

An admitting physician is called to the emergency department to evaluate a 58-year-old woman who presents with a 2-day history of fever, anorexia, nausea, and abdominal pain. Suspecting pancreatitis, the physician inquired about a history of similar symptoms. She was seen 2 months ago in the emergency department for an episode of unrelenting, achy right upper quadrant abdominal pain, at which time ultrasound imaging demonstrated multiple gallstones without evidence of cystic duct obstruction or gallbladder wall edema. At this time, serum amylase and lipase levels are both grossly elevated. On day 3 of her hospital course, the physician is called urgently to evaluate the patient for hypotension, increased shortness of breath, and ensuing respiratory failure. She requires endotracheal intubation and mechanical ventilation. A chest radiograph and severe hypoxia support the diagnosis of acute respiratory distress syndrome.

Questions

A. By what mechanism can biliary stones cause pancreatitis?

Biliary tract disease is a common cause of acute pancreatitis. It is hypothesized that the inciting event is obstruction of the common bile and main pancreatic ducts by a gallstone lodged in the ampulla of Vater. Parenchymal injury may be caused by the local reflux of bile or duodenal contents. It has also been proposed that inflammation is caused by bacterial toxins or free bile acids transported from the gallbladder to the pancreas through lymphatics.

B. At the time of admission, what additional historic features and laboratory studies should be obtained to further clarify the etiology of her pancreatitis?

Although choledocholithiasis appears to be the most likely cause of this patient’s acute pancreatitis, other causes should be considered, for example, alcohol use, infection (viral, bacterial, and parasitic), concomitant drugs, recent surgeries, comorbid rheumatologic disease, and a family history of pancreatitis. Laboratory studies such as a serum calcium and lipid panel, including triglycerides, would be helpful in ruling out important metabolic causes of pancreatitis. Of note, however, the cause of the pancreatitis remains unclear despite workup in approximately 15–25% of cases. To help guide prognosis, Ranson criteria require an assessment of the white blood cell count, serum glucose, LDH, and AST.

C. Describe how acute pancreatitis may be complicated by acute respiratory distress syndrome.

Acute respiratory distress syndrome (ARDS) may be caused, in part, by activated pancreatic enzymes such as circulating phospholipases, which are released systemically and interfere with the normal function of pulmonary surfactant. In addition, the systemic release of both the CC and CXC families of cytokines and endotoxin, beginning shortly after pain onset and peaking 36–48 hours later, corresponds temporally with the profound clinical decline observed. In particular, substance P, neurokinin-1, and platelet activating factor (PAF) are involved in the proinflammatory responses seen in pancreatitis-associated acute lung injury. Elevated serum levels of IL-6 have been associated with the severity of lung injury in acute pancreatitis, an effect mediated by NFκB activation in pancreatic acinar cells. IL-6 and other inflammatory signaling pathways may prove to be appropriate therapeutic targets in severe acute pancreatitis, although to date no therapeutic agents have been found to be effective in clinical trials.

Case 75

Chronic Pancreatitis

A 52-year-old man with a 20-year history of alcohol abuse presents to his primary care provider complaining of recurrent episodes of epigastric and left upper quadrant abdominal pain. Over the past month, the pain has become almost continuous, and he has requested morphine for better pain control. He has a history of alcohol-related acute pancreatitis. Examination reveals a 10-pound weight loss over the past 6 months. He has some mild muscle guarding over the epigastrium with tenderness to palpation. Bowel sounds are somewhat decreased. Serum amylase and lipase are mildly elevated. A plain film of the abdomen demonstrates pancreatic calcifications.

Questions

A. How often do heavy drinkers develop chronic pancreatitis?

Alcoholism is the most common cause of chronic pancreatitis, accounting for 70–80% of cases. The risk is directly related to the duration and amount of alcohol consumed, but in fact, only 5–10% of heavy drinkers actually develop the disease. Recent epidemiologic evidence identifies cigarette smoking as a strong independent risk factor for the development of chronic pancreatitis. Moreover, tobacco exposure appears to have a dose-dependent relationship with its incidence. The number of daily cigarettes smoked as well as the duration of tobacco smoke exposure appear to be important risk factors. Last, the combination of significant alcohol and cigarette augments the risk of chronic pancreatitis.

B. What are the proposed mechanisms of alcohol-induced chronic pancreatitis?

It is thought that ethanol causes secretion of insoluble pancreatic proteins that calcify and occlude the pancreatic duct. This results in progressive fibrosis and subsequent destruction of glandular tissue. In addition, deficiencies of dietary antioxidants such as zinc and selenium may lead to the buildup of toxic free radicals. Unlike other forms of chronic pancreatitis, alcohol-related chronic disease may evolve from multiple episodes of severe acute pancreatitis.

C. Why may a proton pump inhibitor be helpful for this patient?

Proton pump inhibitors may be helpful adjuvant therapy along with pancreatic enzyme replacement by decreasing postprandial gastric acid secretion, commonly seen in patients with severe pancreatic insufficiency.

Case 76

Pancreatic Insufficiency

A 15-year-old boy with a history of cystic fibrosis comes to see you because of worsening diarrhea and weight loss. His lung disease has been relatively well controlled, but recently he has lost 5 kg unintentionally. His stools have also become loose and are very bulky, greasy, and foul smelling, especially after fatty meals. On examination, he is thin but otherwise normal appearing with a weight of 45 kg and height of 160 cm. Lung examination is notable for scattered rhonchi and crackles, but the rest of the examination, including the abdominal exam, is normal. Stool collection verifies the presence of steatorrhea. He is started on pancreatic enzymes with resolution of his gastrointestinal symptoms.

Questions

A. Why is fat malabsorption such a prominent finding in pancreatic insufficiency?

Because pancreatic lipase is essential for fat digestion, its absence leads to steatorrhea (the occurrence of greasy, bulky, light-colored stools). On the other hand, although pancreatic amylase and trypsin are important for carbohydrate and protein digestion, other enzymes in gastric and intestinal juice can usually compensate for their loss. Thus, patients with pancreatic insufficiency seldom present with maldigestion of carbohydrate and protein (nitrogen loss).

B. What are other consequences of pancreatic insufficiency?

In severe cases of fat malabsorption, deficiencies of the fat-soluble vitamins (vitamins A, D, E, and K) may occur and require parenteral supplementation. Diarrhea results from the cathartic action of hydroxylated fatty acids. These fatty acids inhibit the absorption of sodium and water by the colon. Hypocalcemia, hypophosphatemia, tetany, osteomalacia, osteopenia (low bone mineral density), and osteoporosis can occur both from deficiency of the fat-soluble vitamin D and from the binding of dietary calcium to unabsorbed fatty acids, forming insoluble calcium-fat complexes (soaps) in the gut. These soaps also prevent the normal binding of dietary oxalate to calcium. Dietary oxalate remains in solution and is absorbed from the colon, causing hyperoxaluria and predisposing to nephrolithiasis. About 40% of patients with pancreatic insufficiency demonstrate malabsorption of vitamin B₁₂ (cobalamin), although clinical manifestations of vitamin B₁₂ deficiency (anemia, subacute combined degeneration of the spinal cord, and dementia) are rare. The malabsorption of vitamin B₁₂ appears to result from reduced degradation by pancreatic proteases of the normal complexes of vitamin B₁₂ and its binding protein (R protein), resulting in less free vitamin B₁₂ to bind to intrinsic factor in the small intestine. Finally, long-standing malabsorption leads to protein catabolism and consequent weight loss, muscle wasting, fatigue, and edema. At times, weight loss occurs in patients with chronic pancreatitis because eating exacerbates their abdominal pain or because narcotics used to control pain cause anorexia.

Case 77

Carcinoma of the Pancreas

During a family reunion, a 62-year-old widower describes to his son a 1-month history of lethargy. He attributed it to the stress of a recent move from a large three-bedroom house into an apartment. His granddaughter comments that his eyes appear “yellow” and that he has lost a significant amount of weight since their last visit with him. Corroborating the finding of painless jaundice, his internist orders a contrast-enhanced spiral CT, revealing a 3-cm mass in the head of the pancreas.

Questions

A. On physical examination, the patient has a palpable and mildly tender gallbladder. What is the significance of this finding?

The Courvoisier law distinguishes the causes of the gallbladder findings on physical examination. A palpable gallbladder makes gallstones of the common bile duct less likely than carcinoma of the pancreas because gallstones typically result in inflammation and subsequent scarring, resulting in a shrunken, and not a distended, gallbladder.

B. What hematologic abnormalities may be associated with pancreatic cancer?

Adenocarcinomas of the pancreas may present with anemia, migratory thromboembolic disease, or disseminated intravascular coagulation. The coagulopathies may be related to thromboplastins released within the mucinous secretions of the adenocarcinoma.

C. What are some important clinical prognostic factors?

Clinical prognostic factors include tumor size, site, clinical stage, lymph node metastasis, type of surgery, anemia requiring blood transfusion, performance status, and adjuvant radiation therapy. The poor overall prognosis (5-year survival of <5%, and only 15–20% of patients undergoing curative tumor resections living >5 years) can be attributed primarily to the advanced stage of the disease by the time it presents clinically, the rapid rate of local tumor expansion, and the early systemic dissemination.