Published: February 2014
Chronic pancreatitis (CP) is a progressive inflammatory disease of the pancreas, characterized by irreversible morphologic changes and gradual fibrotic replacement of the gland. Loss of exocrine and endocrine function results from parenchymal fibrosis. The primary symptoms of CP are abdominal pain and maldigestion. Because of diagnostic and therapeutic challenges, an interdisciplinary management strategy is required.
The incidence of CP ranges from 1.6 to 23 cases per 100,000 population per year worldwide. Chronic pancreatitis in the United States results in more than 122,000 outpatient visits and more than 56,000 hospitalizations per year.
Grossly, the pancreas may be enlarged or atrophic, with or without cysts or calcifications. The ducts may be dilated, irregular, or strictured. Essential pathologic features include irregular and patchy loss of acinar tissue, chronic inflammation, ductal changes, and fibrosis. These gross changes are end-manifestations of complex pathogenic mechanisms that are associated with gene mutations, metabolic and environmental factors.
Several past theories have been developed to explain the pathogenesis of chronic pancreatitis. The premise of the oxidative stress hypothesis is that reactive by-products of hepatic mixed function oxidase activity damage the pancreas through chronic reflux of bile into the pancreatic duct. The toxic-metabolic theory is that alcohol is directly toxic to the acinar cell through a change in intracellular metabolism. The stone and duct obstruction theory suggests that alcohol increases the lithogenicity of pancreatic juice and causes stone formation. Chronic contact of the stones with duct epithelial cells produces ulceration, scarring, and obstruction of the acinar glands. The necrosis-fibrosis theory emphasizes that acute and chronic pancreatitis represents a spectrum of disease. Inflammation from acute pancreatitis leads to scarring and extrinsic compression of the pancreatic ductules. Recently, research in the immunology of pancreatitis has demonstrated the primary role of pancreatic stellate cells. The sentinal acute pancreatitis event (SAPE) hypothesis encompasses past theories and the new knowledge about pancreatic stellate cells. This model includes 3 sequential phases on a continuum – pre acute pancreatitis, the sentinel attack of acute pancreatitis and progression phase. It presupposes that CP occurs in patients who are genetically or environmentally at risk of pancreatitis (pre-acute pancreatitis stage). The sentinel event of pancreatitis (second stage) in a predisposed patient leads to triggering of chronic inflammation in pancreas and importantly, activation and recruitment of stellate cells. Perpetuation of these immune responses from variety of insults (e.g. recurrent AP from mutations or excessive alcohol) leads to progression (third stage) and irreversible changes.1,2
Heavy and prolonged alcohol use is a common cause of CP. In contrast to other causes, alcohol-related CP is associated with more-severe pain, extensive calcification and ductal changes, and more rapid progression to endocrine and exocrine insufficiency. Most patients experience recurrent episodes of acute pancreatitis for several years before CP develops. Interestingly, only 3% of alcoholics develop CP, implying the presence of cofactors that amplify the effect of alcohol. A high-fat diet and smoking might also contribute to pancreatic disease in alcoholics. Smoking adversely affects pancreatic bicarbonate and water secretion, induces oxidative stress, and increases the rate of pancreatic calcification. Indeed, there have been several studies that demonstrated smoking as an independent risk factor for both development and progression of chronic pancreatitis.
Tropical pancreatitis is endemic to certain developing regions, such as India, Africa, and South America. Episodic abdominal pain begins in childhood and is followed by rapid progression to endocrine and exocrine insufficiency. Nutritional factors such as micronutrient deficiencies (zinc, copper, and selenium) may be involved in the pathogenesis of tropical pancreatitis. Causes of obstructive CP include pancreatic adenocarcinoma, neuroendocrine tumors, and intrapapillary mucinous tumors. Autoimmune pancreatitis is a recently described disorder which resides on the continuum of IgG4 related sclerosing disease. Pancreatic involvement often presents with obstructive jaundice or acute recurrent pancreatitis due to pancreatic swelling and ductal strictures compressing the bile duct or pancreatic duct. Laboratory features include an elevated immunoglobulin IgG4 level. Imaging features include focal or diffuse pancreatic enlargement and a narrowed pancreatic duct. The clinical and radiographic features of the disease improve rapidly with corticosteroid therapy.
Important discoveries have been made in the genetic basis of pancreatic disease. Hereditary pancreatitis is a rare autosomal dominant disease that causes recurrent painful episodes of acute pancreatitis in childhood, leading to CP and pancreatic cancer in adulthood. Hereditary pancreatitis occurs through a mutation of the cationic trypsinogen gene (PRSS1), leading to loss of autoregulation of activated trypsin. Additionally, studies have demonstrated a high prevalence of CFTR gene mutations in patients presenting with idiopathic acute and chronic pancreatitis. Although 85% of cystic fibrosis patients have the severe form of cystic fibrosis, with respiratory disease and pancreatic insufficiency, the remaining 15% possess lower sweat chloride levels and might express other phenotypes, including pancreatitis. Most patients who present with pancreatitis as the sole phenotypic feature of cystic fibrosis have one or two mild CFTR mutations. Other susceptibility genes that have been established include the pancreatic secretory trypsin inhibitor gene (SPINK1), the chymotrypsinogen C gene (CTRC), and the calcium sensing receptor gene (CASR).
Severe hypercalcemia is known to trigger episodes of acute pancreatitis through trypsin-mediated mechanisms and can progress to CP. Chronic renal failure is associated with an increased prevalence of CP, perhaps related to a direct toxicity of uremia on the pancreas. Hypertriglyceridemia and gallstones rarely cause CP.
Ten percent to 30% of patients with CP possess no identified risk factors. Idiopathic CP has a bimodal age presentation. Early-onset idiopathic CP manifests with severe abdominal pain in childhood, with relatively few structural and functional changes. Late-onset idiopathic CP manifests in late adulthood, often with minimal pain and pronounced exocrine insufficiency.
A commonly used etiologic classification of CP is Toxic Idiopathic Genetic Autoimmune Recurrent Obstructive (TIGAR-O) system that reflects different underlying mechanisms of pancreatic injury. (See Table 1.)
|Etiologic Risk Factors|
|Chronic renal failure|
|Cause unknown; likely genetic|
|Isolated autoimmune chronic pancreatitis|
|Associated with the following:|
|Recurrent and severe acute pancreatitis|
|Postnecrotic (severe acute pancreatitis)|
|Pancreas divisum (controversial)|
|Sphincter of Oddi dysfunction (controversial)|
|Duct obstruction (tumors, post-traumatic)|
Abdominal pain occurs in 50% to 80% of cases and is responsible for most hospitalizations related to this illness. Pancreatic pain is dull or boring in quality and worsens after eating. The pain is located in the epigastric area and often radiates to the back. There may be associated nausea and vomiting with exacerbations of pain. Two patterns of abdominal pain have been described in CP. Type A pain is characterized by short relapsing episodes lasting days to weeks, separated by pain-free intervals. Type B pain implies prolonged, severe, unrelenting pain. Recent study suggests that type B pain is associated with worse quality of life, greater healthcare need and disability. Pain exacerbations are not always associated with elevations of serum amylase and lipase levels. Previously, results of some studies suggested diminishment of pancreatic pain with progression of CP. However, subsequent large prospective studies refuted this notion and findings suggest progression of disease may in fact lead to a state where pain management is less effective. There are several proposed pathogenic mechanisms of pancreatic pain in CP, including intraductal hypertension, neural inflammation, upregulation of pain mediators, and proliferation of unmyelinated nerve fiber, neurohormonal changes, and abnormal feedback mechanisms. Concomitant gastroparesis, and complications of CP (see section on complications (Table 2)) can contribute to abdominal pain and appropriately investigated if suspected. Some patients develop nonvisceral (central) pain, often after years of narcotic dependency.
Gradual pancreatic fibrosis produces a steady deterioration in enzyme output, leading to steatorrhea and weight loss. Clinically apparent steatorrhea does not occur until 90% of pancreatic function has been lost. The sudden development of steatorrhea suggests main pancreatic duct obstruction by inflammatory strictures, stones, or cancer. Endocrine insufficiency does not occur until late in the disease course. Pancreatic diabetes requires insulin and is typically brittle because of concomitant glucagon deficiency. Weight loss in CP is multifactorial, related to maldigestion, fear of eating, anorexia, nausea, and vomiting. Severe or rapid weight loss is a red flag for pancreatic cancer.
Complications of chronic pancreatitis are listed in Table 2.
|Signs and Symptoms||Treatment|
Mild elevations in amylase and lipase levels
|Drainage for large or symptomatic pseudocysts
Endoscopic drainage (transmural or transpapillary)
Surgical drainage (cyst gastrostomy or cyst jejunostomy)
|Jaundice||Drainage of obstructing pseudocyst
|Gastric Outlet Obstruction|
Nausea and vomiting
|Drainage of pseudocyst
|Consider surgical resection
|Increased abdominal girth
|Endoscopic stent placement
Total parenteral nutrition
|Shortness of breath
High-amylase pleural fluid
Endoscopic stent placement
Total parenteral nutrition
|Splenic vein thrombosis|
|Bleeding from gastric varices||Splenectomy|
Chronic pseudocysts are benign cysts formed of pancreatic fluid and surrounded by a fibrous wall. The pathogenesis of pseudocysts in CP is believed to be ductal obstruction, leading to upstream dilation and cyst formation. ERCP might show communication of the cyst with the main pancreatic duct. The typical clinical presentation of a pseudocyst is worsening abdominal pain in the setting of known CP, with or without mild elevation in the serum amylase and lipase levels. Biliary obstruction and gastric outlet obstruction can result from compression of the bile duct and duodenum from severe fibrosis, enlarging pseudocysts, or pancreatic cancer. Pancreatic adenocarcinoma develops in 4% of patients with long-standing CP. Imaging tests often produce uncertainty in differentiating cancer from inflammatory masses.
Pancreatic fistulas, ascites, and pleural effusions arise from a communication of pancreatic pseudocysts with adjacent cavities or from disruption of the pancreatic ducts. The diagnoses of pancreatic ascites and pleural effusions are based on the finding of elevated fluid amylase. Endoscopic stent placement across the pancreatic duct disruption can ameliorate these complications. Splenic vein thrombosis is common and usually asymptomatic; however, recurrent bleeding from secondary gastric varices develops in some patients.
Diagnostic tests for chronic pancreatitis are shown in Table 3.
Almost 100% specificity
|CT||++||0||++||Detects advanced disease|
|MRI/MRCP||+++||0||+++||Assesses ducts and parenchyma
Secretin enhancement may improve sensitivity
|EUS||+++||++||+++||Assesses ducts and parenchyma
|ERCP||++++||+++||+++||Detects early ductal changes|
|Hormone-stimulated PFT||++++||++||++||Traditional methods not widely available
Endoscopic methods in development
CT, computed tomography; ERCP, endoscopic retrograde cholangiopancreatography; EUS, endoscopic ultrasound; MRI/MRCP, magnetic resonance imaging/MR cholangiopancreatography; PFT, pancreatic function test
Advanced CP is most easily diagnosed with a pancreatic-protocol CT scan. The cardinal CT features of CP are pancreatic atrophy, calcifications, and main pancreatic duct dilation. Using these criteria, CT has a sensitivity of 74% to 90% and a specificity of 84% to 100%. Additionally, CT can detect CP complications, including pseudocysts, splenic artery pseudoaneurysm, and biliary obstruction. The finding of pancreatic head enlargement suggests pancreatic cancer or an inflammatory mass.
The diagnosis of early (minimal change) CP is more challenging. ERCP involves the endoscopic injection of contrast into the pancreatic duct via the papilla of Vater for fluoroscopic imaging. ERCP ductal changes are graded from equivocal (class I) to severe (class IV) (Cambridge classification). Although ERCP is considered accurate for early CP, it carries a 5% to 10% risk of acute pancreatitis. The role of ERCP in the diagnosis of chronic pancreatitis has recently decreased because safer and less-invasive techniques are available.
EUS is a minimally invasive test that allows a close sonographic inspection of the head, body, and tail of the pancreas from gastric and duodenal stations. The finding of five or more EUS criteria has been an approximate sensitivity of 75% and specificity of 80% for CP in studies incorporating ERCP as reference standard (Table 4).
|Side branch dilation|
MRCP is comparable to CT for detecting major structural abnormalities (atrophy, pseudocysts). Gadolinium-enhanced MRCP can offer improved differentiation of neoplastic and inflammatory masses compared with CT. T1-weighted images showing parenchymal abnormalities can correlate with early parenchymal fibrosis. Heavily T2-weighted images produce bright enhancement of fluid-filled structures, allowing a noninvasive alternative to ERCP for imaging the pancreatic duct. Secretin-enhanced MRCP allows a functional assessment of the pancreas by visualizing and quantifying duodenal fluid volume after secretin stimulation of the exocrine pancreas. Although EUS and MRCP are now commonly used for diagnosing CP, there are many unanswered questions regarding the true accuracy, interobserver variability, and cost-effectiveness of these tests.
When imaging test results are negative or equivocal, direct, hormone-stimulated pancreatic function tests are helpful to confirm or rule out the presence of mild exocrine insufficiency as a surrogate marker of early CP. Pancreatic function tests involve stimulating the pancreas with intravenous CCK or secretin and then collecting fluid from the duodenum. Pancreatic fluid is analyzed for enzyme and bicarbonate production. Several noninvasive indirect tests of pancreatic function are also available, including fecal fat analysis, fecal chymotrypsin, and fecal elastase assays. These tests are sensitive for moderate and late-stage exocrine insufficiency but lack sensitivity for detecting early disease.
Three components are essential to the optimal management of CP: control of pain, improvement of maldigestion, and management of complications.
The management of chronic pancreatic pain is challenging. The AGA has published an evidence-based technical review on the management of pain in CP. In this review, the available medical, endoscopic, surgical and other techniques for pain control are critically evaluated in the context of existing literature. The following discussion reflects these guidelines and most recent literature.
Some have advocated supportive therapy for abdominal pain on the premise that fibrosis and scarring ultimately progress to pancreatic burnout and spontaneous relief of pain. Although long-term improvement in pain has been observed in some patients with CP, a significant subset of patients experiences debilitating pain for decades. The AGA technical review has stated that "a strategy of waiting for spontaneous pain relief is not reliable and may be unreasonable advice for the patient with persistent, severe pain." Furthermore, there is growing evidence that there is no association between duration of CP and improvement in pain. Medical options for pain relief include abstinence from alcohol and smoking, analgesics, and pancreatic enzymes. Abstinence from alcohol is critical because continued use can hasten disease progression, aggravate chronic pain, and increase mortality. Non-narcotic analgesics (e.g., nonsteroidal anti-inflammatory drugs, acetaminophen, and tramadol) are the next step in managing painful CP. The role of smoking in the progression of fibrosis and functional impairment has been established in recent studies. Therefore smoking cessation should also be strongly advised. If pain persists, low doses of mild narcotics may be added. Severe or recalcitrant pain can warrant the use of stronger opiates in selected cases.
Pancreatic enzymes are presumed to improve pain by suppressing CCK release from the duodenum, leading to decreased pancreatic stimulation. The AGA review has critically appraised the literature regarding the controversial use of enzymes for pain. A meta-analysis of six randomized, placebo-controlled trials did not reveal a statistically significant benefit for supplemental pancreatic enzyme therapy for pain relief. However, there was substantial methodological heterogeneity among the included trials. Uncoated preparations might work better by enhancing delivery to the proximal small bowel. Uncoated pancreatic enzymes may be worth trying in all patients because of their safety and minimal side effects. However, the AGA technical review has cautioned that "additional studies are required to establish the effectiveness of this modality of treatment and to define whether certain subsets are more likely to benefit from enzyme therapy."
Over the years, data have emerged to show the potential role of oxidative stress in pathophysiology of CP. The term "oxidative stress" was coined to denote presence of excessive oxygen free radicals coupled with reduced antioxidant capacity leading to acinar cell injury. Several studies - including 2 well designed randomized controlled trials3-6 — tested the hypothesis that replenishing key components in the anti-oxidant pathway will lead to improvement in pain, but showed conflicting results. Therefore, there is no evidence that antioxidant therapy improves pain in CP.
Interestingly, many patients with CP have nonvisceral pain (central or somatosensory in origin). A differential nerve blockade is helpful in determining whether there is a central or somatosensory component to the pain syndrome. A differential nerve blockade is indicated for any patient with pancreatic pain that does not respond to simple medical therapeutic measures such as non-narcotic analgesics and enzymes. Antidepressants, anticonvulsants (gabapentin), topical therapy, psychiatric counseling, and opioid rehabilitation may be of use for patients with nonvisceral pain. Celiac or sphlanchnic nerve blockade may be used in select patients with visceral pancreatic pain. Limited studies have suggested that a subset of patients obtain significant short-term pain relief from CT-guided celiac plexus blockade. EUS-guided celiac plexus blockade has emerged as an effective and perhaps safer alternative to percutaneous methods. Meta-analysis that pooled data from 6 studies reported EUS guided celiac plexus blockade leads to short-term pain relief in approximately half of patients.7 However, there are no placebo-controlled randomized trials to date evaluating the efficacy of celiac plexus blockade in treatment of pain in CP.
It is generally accepted that pain in CP can result in part from obstruction of the main pancreatic duct from stones and strictures, leading to increased ductal and parenchymal pressure. Because obstruction contributes to pain, patients with an enlarged, obstructed main pancreatic duct might benefit from endoscopic therapy, lithotripsy or surgical duct decompression therapy. Endoscopic techniques include biliary or pancreatic sphincterotomy (or both), removal of pancreatic duct stones, and placement of pancreatic stents. Extracorporeal shockwave lithotripsty (ESWL) is also an effective ancillary treatment for patients with pancreaticductal stones either alone or in combination of endoscopic therapy. In experienced centers ESWL alone is equally effective compared to combined ESWL and endoscopic therapy, and maybe more cost effective.
Several surgical options exist for select patients with visceral pain resulting from CP. In patients with a dilated main pancreatic duct, a side-to-side pancreaticojejunostomy (Puestow procedure) may be performed. Most studies of surgical and endoscopic decompressive therapy in CP have revealed good short-term but poor long-term pain control. In patients with large duct obstructive disease, surgery seems to be superior to endoscopic therapy in relieving pain. A recent randomized control trial, surgery relieved pain in significantly greater proportion of patients than endoscopic treatment with comparable morbidities.8 The benefit was durable at 5 years and patients assigned to surgery required less number of procedures.9 A Cochrane systematic review also found that the surgical therapy was superior to endoscopic treatment in treating pain in this particular subset of patients (i.e. painful, large duct obstructive CP).10 The AGA technical review has stated that the surgical procedures are best performed based on "need for long-term narcotic therapy, marked diminution of the quality of life because of intractable pain, or major nutritional consequences of pain."
Pancreatic resection is reserved for patients with disease of the small duct and pain unresponsive to medical therapy. The Whipple procedure and distal pancreatectomy have been used in the past to treat patients with small-duct CP. Newer resection techniques have offered substantial relief of pain related to an inflamed and scarred gland, with preservation of surrounding structures. For example, the Beger procedure involves resection of the inflamed pancreatic head with careful sparing of the duodenum; the Frey procedure adds longitudinal duct decompression to the pancreatic head resection. The AGA review has cited several potential drawbacks of resection procedures, including paucity of randomized trials; loss of exocrine and endocrine function, including diabetes; technical expertise required for organ-sparing methods; and lack of pain relief in some patients, even after total pancreatectomy. In spite of these drawbacks, resection offers significant relief to a subset of carefully chosen patients, particularly if performed in high-volume centers. Total pancreatectomy with auto–islet cell transplantation (TP/AIT) has been performed at several centers in the United States. TP/AIT involves removing the entire pancreas. The pancreatic explant is enzymatically digested to isolate the insulin-producing islet cells. In a second surgery, the islet cells are infused through the portal vein to engraft within the liver. Several case series and systematic review of TP/AIT suggest that most patients receive substantial relief of pain, significant reduction in narcotic requirement and up to 64% achieve insulin independence at 5 years.11
Pancreatic enzymes are used for the treatment of maldigestion in CP. Exogenous pancreatic enzymes are safe, are well tolerated, and produce few side effects. Pancreatic enzyme preparations differ based on enzyme content, the use of microspheres versus microtablets, and the presence of a coating for delayed release. Lipase is the most important determinant of the effectiveness of individual preparations. A minimum of 30,000 U lipase per meal allows adequate intraluminal digestion of fat and protein in most patients. The dose might need to be titrated to as much as 80,000 U lipase per meal, because not all the lipase may reach the proximal small intestine in active form. In a recent double blind randomized controlled trial, administration of 72,000 lipase units per meal in patients with moderate to severe exocrine pancreatic insufficiency was well tolerated and significantly superior than placebo in improving fat absorption, flatulence and stool consistency.12 After open label follow-up of 6 months, treatment arm achieved an average weight gain of nearly 3kg from baseline and decreased stool frequency.13 This finding was reproduced by another study using a similar dose of a different formulation in another country.14 Dosing of enzymes is most physiologic if one half the amount is taken at the onset of the meal and the other half is taken approximately 15 minutes into the meal. Because uncoated preparations are more easily denatured by gastric acid, acid suppression with a proton pump inhibitor (e.g., omeprazole 20 mg once daily) or histamine-receptor antagonist (e.g., famotidine 20 mg twice daily) is required.
Response to enzyme therapy may be monitored through an assessment of symptoms or, more objectively, through 72-hour stool fat quantification. A poor response to pancreatic enzymes can suggest noncompliance, loss of enzyme potency, improper timing of enzymes in relation to meals, or coexisting mucosal disease. A daily proton pump inhibitor may be added for those refractory to therapy because gastric acid can denature exogenous enzymes. A general approach to the treatment of maldigestion in CP is shown in Figure 1.
Large or symptomatic pseudocysts may be drained endoscopically through transmural or transpapillary approaches. Large pseudocysts may also be drained surgically through cyst gastrostomy. Biliary and gastric outlet obstructions are best managed through surgical decompression. The initial management of the complications of pancreatic duct disruption or fistulas (pancreatic ascites or pleural effusions) includes prolonged pancreatic rest (parenteral nutrition), octreotide, and endoscopic placement of pancreatic duct stents. In some cases, surgical resection may be necessary.
Because of the significant challenges inherent in the management of this disease, we have developed a multidisciplinary approach similar to the AGA algorithm. Patients first undergo a diagnostic and staging evaluation (Figure 2). Most patients undergo CT as an initial diagnostic test. Secretin-stimulated endoscopic pancreatic function tests and EUS are used as second-line tests to diagnose early CP in patients in whom initial imaging is negative.
Once the diagnosis of CP has been established, patients with severe pancreatic pain refractory to initial conservative management may be referred for a differential nerve blockade to clarify the origin of their pain syndrome (Figure 3). Patients with nonvisceral pain are referred for psychotherapy and treatment of chemical dependency. Patients with visceral pain are first given a trial of conservative medical management. If pain persists, patients with large-duct disease or pseudocysts are referred for surgical management or endoscopic therapy. Patients with small-duct disease (or those with large-duct disease failing duct-decompression measures) are referred for an EUS-guided celiac plexus blockade. Other options for these patients include sphlanchnic radio-frequency ablation, spinal cord stimulation, and total pancreatectomy with auto–islet cell transplantation.