Chronic pancreatitis is an inflammatory and fibrosing disease of the exocrine pancreas characterized by irreversible morphological changes and permanent loss of function. The incidence of chronic pancreatitis has increased approximately fourfold over the past several decades. This apparent increase is due in part to a broadening of its definition and the inclusion of patients with earlier-stage disease. The natural history of chronic pancreatitis is unpredictable. Affected individuals typically suffer a pattern of persistent or recurrent attacks of pain and progressive pancreatic exocrine dysfunction. Additional symptoms may develop from extension peripancreatic inflammation and fibrosis to adjacent organs and vascular structures. In later stages, pancreatic endocrine insufficiency may develop. Decision making in the management of chronic pancreatitis must be individualized the specific anatomic and pathological circumstances, taking into account the extent of local expertise in various diagnostic and therapeutic modalities as well as the fact that there is a relative paucity of high-quality data on the clinical effectiveness of surgical and medical interventions. Optimal management is facilitated by a multidisciplinary approach that includes surgical, endoscopic, and radiological expertise in addition to nutrition, endocrinology, pain management, and psychosocial support.
DEFINITION AND RISK FACTORS
Pancreatitis is thought to have its origin as an autodigestive disease initiated by inappropriate activation of pancreatic zymogens. The terms acute pancreatitis and chronic pancreatitis are often used to draw the temporal distinction between an isolated episode and a more persistent illness associated with a gradual, progressive loss of pancreatic function. In fact, pancreatitis represents a far more heterogeneous clinical entity than can be captured by these two simple descriptors. A number of international conferences have been held in order to develop uniform terminology to characterize the spectrum of morphology seen in acute and chronic pancreatitis.
According to the Marseilles-Rome classification of 1988, the term acute pancreatitis is used to refer to single or repeated episodes of abdominal pain associated with a range of potentially reversible pancreatic lesions including pancreatic edema, necrosis, and hemorrhage, as well as peripancreatic fluid collections, necrosis, and pseudocysts. Chronic pancreatitis is used to refer to recurrent or persistent abdominal pain that is associated with irreversible and ongoing inflammatory destruction of exocrine parenchyma and eventually, islets. In practice, however, the distinction between acute and chronic pancreatitis is rarely made based on tissue sampling, and there is no consensus on the definition of irreversible morphological change.1 It is also acknowledged that certain forms of chronic pancreatitis can occur in the absence of pain.
The Marseilles-Rome classification further divides chronic pancreatitis into several morphological subtypes that may coexist in the same patient. Chronic obstructive pancreatitis is characterized by exocrine atrophy and is associated with duct stenosis caused by tumors, pseudocyst, or scarring from prior acute pancreatitis. Chronic calcifying pancreatitis is characterized by intraductal calcifications and protein plugs, and is often associated with atrophy, stenotic ducts, and areas of acute inflammation or pseudocyst. Chronic inflammatory pancreatitis consists of dense infiltration of mononuclear inflammatory cells. Retention cysts and pseudocysts, seen in both calcifying and obstructive forms, may also become infected. Fibrosis may develop in the absence of symptoms.
Chronic pancreatitis lacks a simple unifying theory of disease pathogenesis. The precise mechanism by which any specific agent or circumstance induces pancreatitis remains obscure. Acute pancreatitis clearly has the potential to evolve into chronic disease. However, repeated episodes of acute pancreatitis do not invariably lead to chronic pancreatitis, and chronic pancreatitis may present without prior acute attacks. Excessive alcohol ingestion has been associated with chronic pancreatitis since the term was introduced by Comfort in 1946.2 Alcohol use continues to be the most commonly identified environmental risk factor clinically. The precise relationship between alcohol and chronic pancreatitis remains, however, poorly understood. Alcohol ingestion does not lead to pancreatitis in experimental animal models. Alcoholism is by no means uniform among human pancreatitics. Chronic pancreatitis in humans occurs in the absence of significant alcohol usage, and fewer than 5% of alcoholics actually develop pancreatic disease.3,4 Acute and chronic forms of pancreatitis have been associated with exposure to toxic agents other than alcohol. As with alcohol, most individuals exposed to the other toxic substances associated with pancreatitis do not develop the chronic form of the disease.
Given the lack of an identified, uniform pathogenic trigger, the concept of risk modifiers rather than etiologies or causes of chronic pancreatitis may be more appropriate in classifying the disease, particularly when making decisions regarding patient management. Chronic pancreatitis is not simply a “drunkard’s disease,” but rather is more appropriately attributed to a variety of genetic, environmental, anatomic, immunologic, and other poorly understood susceptibility factors that interact to initiate and perpetuate the pathology. Whitcomb4 has proposed the TIGAR-O system (Table 71-1) as a potential framework that allows various risk factors associated with the disease to be logically organized into categories: Toxic or metabolic, Idiopathic, Genetic, Autoimmune, Recurrent acute, and Obstructive. The TIGAR-O system reflects the fact that in chronic pancreatitis there is a diversity of etiologic risk factors that contributes to a spectrum of pathological and functional derangements, clinical features, and natural history.
|Chronic renal failure (uremia)|
|Recurrent and severe acute pancreatitis|
|Sphincter of Oddi dysfunction|
|Malignant obstruction of the pancreatic duct|
|Primary pancreatic duct stones|
The majority of patients with chronic pancreatitis (55%-80%) will report significant alcohol intake over the years prior to diagnosis. A relationship between dose and duration of alcohol use has been repeatedly documented, and there appears to be a threshold level for the risk of pancreatitis at approximately 50 g (four drinks) per day.5 Several mechanisms have been proposed to account for pancreatic injury, including alterations in pancreaticobiliary secretory flow, ductal plugging, and direct toxic action on acinar cells. Chronic pancreatitis in the setting of alcohol use is associated with pancreatic calcification and ductal stone formation, but none of the proposed mechanisms is convincingly supported experimentally, and the hypotheses are not mutually exclusive. Several other toxic agents have been identified as risk factors for pancreatitis. Included among these is tobacco, which has been shown to confer increased risk of chronic pancreatitis independent of alcohol use.6 Several medications have been implicated in acute pancreatitis but probably do not play a role in the chronic form of the disease. Similarly, hypercalcemia (eg, associated with hyperparathyroidism) and various forms of hyperlipidemia (eg, hypertriglyceridemia) are linked to acute but not chronic pancreatitis. So-called tropical chronic pancreatitis, described in children living in developing parts of the world, is thought to be either due to a dietary toxin or to an unidentified micronutrient deficiency.
About 20% of patients with chronic pancreatitis have no clinically obvious risk factor. It is suspected that a great many of these idiopathic cases will ultimately prove to harbor yet-unidentified genetic or molecular derangements that explain the process. In recent years, many patients previously considered to be idiopathic recurrent acute and chronic pancreatitis have been found to carry mutations, polymorphisms, or splice variants of the gene associated with cystic fibrosis (CF). Recent evidence also suggests that polymorphisms in genes associated with oxidative stress and xenobiotic metabolism may be more prevalent in patients with what is now characterized as idiopathic disease.7 Thus, as new genetic associations that predispose to the development of chronic pancreatitis become recognized, the percentage of patients with truly idiopathic disease will decrease.
Hereditary pancreatitis was first characterized in 1952 as early onset of chronic pancreatitis clustering in family members without other risk factors.8 At least half of hereditary pancreatitis kindreds have been found to carry germline mutations in the cationic trypsinogen (PRSS1) gene.3,4,9 The arginine-to-histidine (R122H) substitution is the most common defect. Hereditary pancreatitis has an autosomal dominant pattern of inheritance, with a high degree of penetrance. Cationic trypsinogen is produced in the pancreatic acinar cells and, upon cleavage by duodenal enteropeptidase, forms trypsin. Trypsin is a protease that acts to hydrolyze dietary proteins and plays the key role in both initial activation of other pancreatic zymogens (including trypsinogen itself) and in their subsequent proteolytic inactivation. Trypsin encoded by pancreatitis-associated PRSS1 mutations is unusually stable and resists autolytic inactivation, predisposing to premature and extended activation of trypsin within the pancreatic parenchyma.10 Mutations in other genes such as anionic trypsinogen (PRSS2) or the calcium-sensing receptor (CASR) have also been reported in some cases of hereditary pancreatitis, although in many other kindreds, the responsible gene has not yet been identified.11 Other gene associations with hereditary or otherwise idiopathic chronic pancreatitis will undoubtedly emerge over the next several years. Recently, for example, inactivating mutations in the gene encoding for the trypsin-degrading enzyme chymotrypsin C have been identified in a German cohort.12
Another genetic disorder associated with pancreatic pathology is CF, a disease linked to mutations in the CF transmembrane conductance regulator (CFTR) gene.9,13–15 CFTR is a chloride ion channel involved in water, chloride, and bicarbonate secretion by epithelial cells such as those lining the gastrointestinal tract and respiratory system. In the pancreas, CFTR is localized to centroacinar and proximal lobular duct cells.16 Over 90% of CF patients are pancreatic insufficient, and while severe pancreatic fibrosis is common, acute pancreatitis is rare.17 However, a subset of patients with otherwise idiopathic recurrent acute and chronic pancreatitis has been noted to have borderline abnormalities in functional tests for CF such as sweat chloride content. These patients harbor at least an eightfold increase in CF-associated CFTR mutations on a single allele. Various other CFTR mutations, polymorphisms, and splice variants not associated with classical pulmonary manifestations of CF are also frequently identified in patients with recurrent acute and chronic pancreatitis. The CFTR gene shows autosomal recessive inheritance with incomplete penetrance, and thus a family history of CF or pancreatic disease is usually absent in CFTR-associated pancreatitis. The mechanism of CFTR-associated pancreatitis is thought to involve the viscous, low-volume, low-bicarbonate containing pancreatic fluid secretion leading to duct sludge, ductal obstruction, and enzyme hyperconcentration, enhancing the potential for intraglandular enzyme activation.
Mutations and polymorphisms in other genes may also modify susceptibility to chronic pancreatitis. Pancreatic serine protease inhibitor Kazal type 1 (SPINK1) is a natural protease inhibitor that localizes with trypsinogen within zymogen granules. SPINK1 binds to and inhibits activated trypsin, thus serving as a “buffer” of sorts against inappropriate early trypsinogen activation. Mutations of the SPINK1 gene (notably N34S) appear to increase the risk of recurrent acute and chronic pancreatitis, particularly in patients who harbor two mutated alleles.3,4,9 A single mutated SPINK1 allele appears to increase the risk of alcohol-associated pancreatitis and tropical pancreatitis.
Autoimmune pancreatitis (AIP), also known as lymphoplasmocytic sclerosing pancreatitis, is characterized by diffuse glandular enlargement and infiltration with CD4- or CD8-positive lymphocytes and IgG4-positive plasma cells.3,4,18,19 The exact immunologic etiology is unknown, although circulating antibodies with homology both to a peptide sequence associated with a protein from Helicobacter pylori (infection with which is associated with various autoimmune disorders including AIP) and to a protein highly expressed in pancreatic acinar cells have recently been found in over 90% of patients.20 Inflammatory infiltrates are particularly concentrated in duct rather than acinar zones, however, and thus a duct-origin autoantigen has been postulated. Notably, diffused ductal narrowing rather than dilation is usually observed. Initially described predominately in young men, AIP has been increasingly recognized as a cause of biliary obstruction and pseudotumor in older individuals.21 Most patients report little in the way of pain, and prior attacks of acute pancreatitis are unusual. It has been associated with serologic elevation of IgG4 levels in about two-thirds of patients and with other autoimmune conditions in approximately 20%, including Crohn’s disease, ulcerative colitis, Sjögren syndrome, primary biliary cirrhosis, or primary sclerosing cholangitis.18,21
Recurrent episodes, or even a single severe episode of acute pancreatitis, may lead to chronic pancreatitis, but the basis for progression is poorly understood. Patients with prior episodes of necrosis appear to be at particular risk for developing chronic disease. In many cases, progression may be due to postpancreatic ductal scarring, persistent activation of pancreatic stellate cells, and neuroplasticity leading to hyperalgesia.
Post-traumatic duct strictures, or obstruction associated with tumors including cystic neoplasms, neuroendocrine lesions, and pancreatic adenocarcinoma have been associated with pancreatic pathology consistent with chronic pancreatitis, although these patients are often asymptomatic. Chronic pancreatitis has also been associated with anomalous anatomic variations in the pancreatic ductal system, most notably pancreas divisum, and it has been postulated that relative obstruction to pancreatic flow through the dorsal duct and minor papilla predisposes to recurrent acute and chronic pancreatitis. The evidence supporting the association to chronic pancreatitis in particular is largely circumstantial and may reflect referral bias,22 but pancreas divisum may be a contributing factor in the presence of certain genetic risk factors (Fig. 71-1). Some cases of chronic pancreatitis are attributed to sphincter of Oddi dysfunction, although rigorous evidence to support this association is also lacking.
Recurrent acute and chronic pancreatitis in a 41-year-old woman with pancreas divisum and a pancreatitis-associated mutation in the cystic fibrosis (CF) gene. Magnetic resonance cholangiopancreatography (MRCP) demonstrates noncommunicating dorsal (arrow) and ventral (arrowhead) pancreatic ducts.
PATHOPHYSIOLOGY AND MECHANISM OF PAIN
Progress in elucidation of the pathogenesis of chronic pancreatitis has been hampered by the lack of a suitable experimental model that adequately recapitulates the features of the disease seen in humans.23 However, existing evidence suggests a number of useful conceptual frameworks that may help guide efforts to treat patients with chronic pancreatitis. Traditional theories of the pathogenesis of acute pancreatitis include the toxic-metabolic or oxidative stress hypotheses, in which normal acinar cell processing and release of zymogens are disrupted by a toxic or oxidative stressor, and the ductal obstruction hypothesis that proposes a mechanical role for ductal plugs and stones causing disruption of the integrity of the acinar cell (common in alcoholic and tropical disease). In certain situations, notably autoimmune disease, pancreatitis may begin not in the acinar cell but in the duct cell, triggered by the development of an as-yet-unidentified autoantigen on the duct epithelium. Recently, attention has focused on understanding the mechanism of pancreatic fibrosis, the central histological feature that characterizes the evolution from acute disease to chronic pancreatitis. One attractive hypothesis is that a sentinel acute pancreatitis event (SAPE) primes the pancreas for fibrogenesis.23 According to the SAPE concept, local inflammatory cytokines released during acute pancreatitis activate circulating macrophages that infiltrate the gland as well as resident pancreatic stellate cells, myofibroblast-like cells that are normally quiescent. During the subsequent healing phase, anti-inflammatory mediators (particularly anti-inflammatory cytokines such as tumor growth factor beta [TGF-β]) drive stellate cells and tissue macrophages to synthesize and deposit fibrogenic matrix proteins. The pancreatic parenchyma may return to normal after a mild self-limited episode. However, the damage may not completely resolve after a severe attack, particularly if there has been significant tissue necrosis. Thus, following the SAPE, the local pancreatic environment may be permanently altered by the persistent presence of anti-inflammatory and profibrogenic cell populations that are perpetually activated by ongoing toxic-metabolic, oxidative, or mechanical stress. The pancreas then becomes subject to repeated cycles of inflammation and progressive fibrosis.
A comprehensive mechanistic explanation for pain, often the most debilitating symptom of chronic pancreatitis, also remains elusive.23 One hypothesis is that pain results from capsular stretch associated with ductal or organ hypertension. This hypothesis is supported by the favorable results of surgical or endoscopic ductal drainage in patients with chronic pancreatitis associated with dilated pancreatic ducts, and the success of surgical resection in other selected patient populations. An alternative, possibly complementary, hypothesis is that the pain represents a neuropathy caused by repeated inflammatory insults and damage to retroperitoneal sensory nerves.23–25 Recent evidence demonstrating neuroplasticity in nociceptive dorsal root ganglia in chronic inflammatory states, with evidence of upregulation of nociceptors such as TRPV1 by proteolytic enzymes such as trypsin,26 supports this theory.
As in acute pancreatitis, pain in chronic pancreatitis typically localizes to the left upper quadrant or epigastric region, often radiating around or into the back. The pattern of pain is variable. Some patients experience recurrent attacks of moderate to severe pain interspersed with periods of relative or complete quiescence. In others, the pain may be persistent and lead to significant incapacitation and chronic disability. During acute exacerbations, the pain may be increased by food intake and is frequently associated with nausea and vomiting.
Weight loss and malnutrition are common, due to both decreased intake as well as exocrine insufficiency, with consequent malabsorption of protein and fat. Exocrine insufficiency is usually obvious in patients with classical steatorrhea (loose, bulky bowel movements that may be greasy, sticky, oily, or foul-smelling), but these symptoms are obscured by narcotic-associated constipation.
Endocrine insufficiency typically occurs late in the course of disease, often after exocrine insufficiency has appeared, and usually not before about 90% of the pancreatic parenchyma has been replaced by fibrosis. For reasons that are unclear, there is a relative sparing of islet cells until late in the course of the disease. Diabetes is more common in patients with alcohol-associated chronic calcifying pancreatitis, with 80% of these individuals demonstrating endocrine insufficiency within 10 years of the development of severe exocrine insufficiency.27 Histologically, pancreatic islets are seen to persist within areas of extensive fibrotic replacement of exocrine tissue (Fig. 71-2). Because diabetes of chronic pancreatitis is associated with indiscriminate destruction of all cell types within the islets of Langerhans, counterregulatory glucose control may be considerably more labile than in either type I or type II diabetes. Less is known regarding the natural history of nonalcohol-associated chronic pancreatitis, but the risk of diabetes appears to be lower.27 Both endocrine and exocrine insufficiency occur later and less frequently in patients with chronic pancreatitis associated with gene mutations than those without gene mutations.28
On occasion, the initial manifestation of chronic pancreatitis will be related to extrapancreatic complications such as intestinal or biliary obstruction due to compression by a pseudocyst or progressive peripancreatic fibrosis and gastrointestinal hemorrhage due to blood lost into the pancreatic duct (hemosuccus pancreaticus) or due to rupture of pseudoaneurysm into a pseudocyst or to splenomesenteric vein thrombosis.
The diagnosis is usually suspected based on an appropriate clinical history and is confirmed by imaging studies. Laboratory investigation is of limited value. Acute exacerbation of abdominal pain may be paralleled by a transient increase in serum amylase or lipase, but these may be normal with progressive destruction of acinar cell mass. Elevation of liver function tests, particularly serum bilirubin and alkaline phosphatase, may indicate the presence of bile duct obstruction.
The diagnosis of chronic pancreatitis is usually confirmed by imaging studies, most commonly computed tomography (CT). CT findings depend on the morphologic type of chronic pancreatitis, the duration of disease, and the presence of complications. In the early phases of chronic pancreatitis, ductal or parenchymal changes may be rather subtle, but as the disease advances, progressive and irreversible changes in organ architecture are readily apparent. Chronic pancreatitis associated with toxic-metabolic or genetic risk factors, and idiopathic chronic pancreatitis may demonstrate calcifications either focally or scattered throughout the organ. There may be evidence of acute inflammatory changes or focal areas of enlargement associated with areas of dense calcifications, particularly in the pancreatic head (Fig. 71-3); this so-called “inflammatory head mass” appears to be more common in European than American cohorts.29 There may be evidence of segmental or diffuse pancreatic ductal dilation related to stricture formation, and pseudocyst formation and evidence of extrapancreatic complications such as duodenal or biliary obstruction, or splenomesenteric vein thrombosis (Fig. 71-4). In AIP, calcifications are almost uniformly absent and the pancreas is usually diffusely enlarged, although a focal mass-forming variant is occasionally encountered.18 In obstructive forms of chronic pancreatitis, the pancreatic duct is dilated upstream of the area of stenosis, and the acinar parenchyma appears atrophic.
Pancreatic ductography complements CT imaging. Endoscopic retrograde cholangiopancreatography (ERCP) has long served as a gold standard of sorts in mapping duct pathology and offers endotherapeutic options including sphincterotomy and stent placement (Fig. 71-5). Magnetic resonance cholangiopancreatography (MRCP) is less invasive and provides image quality that rivals ERCP; the addition of secretin stimulation further enhances duct visualization and allows some assessment of pancreatic exocrine function (Fig. 71-6).30 Anatomic ductal anomalies such as pancreas divisum are readily defined by ERCP or MRCP, as are dominant focal duct strictures that might be amenable to endoscopic stenting or surgical drainage procedures. MR imaging integrates information regarding parenchymal and ductal involvement and may be particularly helpful when the disease is regionally heterogeneous and architecturally complex.
It is not difficult to establish the diagnosis of chronic pancreatitis in its advanced stages, when classical clinical symptoms are present or when imaging studies demonstrate obvious abnormalities such as strictures, ductal dilation, or pancreatic calcifications. Recognition of disease in its earlier stages presents more of a challenge. A 1983 conference held in Cambridge, England categorized chronic pancreatitis as equivocal, mild, moderate, or marked, and established criteria (Table 71-2) according to combinations of features seen in the main and side branch pancreatic ducts on CT and ductograms.1,31,32 Although this consensus approach has proven useful over the years, there continues to be a subset of patients with symptoms suspicious for chronic pancreatitis but in whom imaging studies are negative. Some of these patients may suffer from functional abdominal pain disorders rather than pancreatic disease.33 Others may have early forms of chronic pancreatitis. Consensus workshops by the Japan Pancreas Society (1995 and 2001) continue to address the ongoing challenge of so-called “minimal change” disease in the context of evolving imaging and diagnostic modalities.
|Grade of Pancreatitis||Main Pancreatic Duct Appearance||Side Branch Pancreatic Duct Appearance|
|Equivocal||Normal||<3 abnormal branches|
|Mild||Normal||>3 abnormal branches|
|Moderate||Abnormal||>3 abnormal branches|
Abnormal plus any of the following:
Cavity >10 mm
Intraductal filling defects
Contiguous organ involvement on CT
Severe duct dilation or irregularity
|>3 abnormal branches|
Endoscopic ultrasound (EUS) appears to be valuable in evaluation of the suspicious pancreatic mass and in characterizing cystic lesions of the pancreas.34 EUS generally adds little to the evaluation of chronic pancreatitis in its advanced stages but has potential applicability in early-stage minimal-change disease where other imaging modalities fail to establish the diagnosis.3,35–37 EUS appears to be more sensitive than ERCP or MRCP in detecting early parenchymal fibrosis and subtle ductal changes occurring in early forms of chronic pancreatitis. Various systems using up to 11 different parenchymal and ductal endosonographic criteria (Table 71-3) to diagnose chronic pancreatitis have been proposed.38 There is, however, no gold standard grading system or agreement on the threshold number of abnormalities that must be present for the diagnosis of chronic pancreatitis. Because of this, the value of EUS in making an early diagnosis of chronic pancreatitis remains uncertain. EUS may have more practical utility in cases of suspected AIP. Surgical interventions may be avoided in some of these patients who present with a mass-forming variant by EUS-directed core needle biopsy demonstrating the pathognomonic lymphoplasmacytic infiltrate39 and thus ruling out malignancy.
Functional testing to demonstrate pancreatic exocrine insufficiency is occasionally helpful, although from a practical standpoint, the condition is usually clinically obvious. Symptoms of steatorrhea, postprandial gaseous distension, or progressive weight loss despite adequate caloric intake are all suggestive of exocrine insufficiency. Quantification of fecal fat content or measurement of fecal human elastase (FE-1) levels can confirm the diagnosis and can be used to monitor efficacy of enzyme supplementation and surgical intervention.40 Unfortunately, these studies are most reliable in those patients in whom the diagnosis is clinically obvious. They are of questionable accuracy in the setting of patients with more subtle symptoms where objective documentation of exocrine insufficiency might be most needed.