Acute and Chronic Pancreatitis
Pancreatitis is more prevalent in children than previously believed. The diagnosis of pancreatitis has increased over the last decade, and large children’s hospitals treat 100 to 150 children a year for pancreatitis. Most patients have acute pancreatitis and 10% or less have acute recurrent or chronic pancreatitis. Greater physician awareness likely accounts for the increase in the diagnosis of acute pancreatitis in childhood. The etiologies of acute pancreatitis differ between children and adults where gallstone and alcohol-associated pancreatitis predominate. In children, systemic illness, biliary disease, medications, and trauma encompass the majority of children with an identifiable etiology. Most patients have no identifiable cause. Similarly, many patients have no identifiable cause for their chronic pancreatitis. Despite advances in understanding the pathophysiology of pancreatitis, care remains supportive. Fortunately, most children have a benign clinical course. Some develop complications, with fluid collections being the most common problem. A small group will have recurrent episodes of acute pancreatitis.
Definition and Classification
Over the years, the classification of inflammatory disorders of the pancreas has changed, but two broad categories have remained constant: acute and chronic. Acute pancreatitis is a reversible process with no lasting effects on the pancreatic parenchyma or function. Chronic pancreatitis is an irreversible process leading to changes in the parenchyma and function of the pancreas. Clinical criteria define both diagnoses. The diagnosis of acute pancreatitis requires two of three criteria: (1) appropriate clinical symptoms, (2) elevations of serum amylase or lipase greater than three times the upper reference limit (URL), and (3) radiographic evidence of pancreatitis, generally by transabdominal ultrasound or by contrast-enhanced computed tomography (CECT). Acute pancreatitis can be further separated into two types: interstitial and necrotizing pancreatitis. Interstitial pancreatitis is most common and generally follows a benign clinical course, whereas necrotizing pancreatitis often heralds a severe clinical course with complications.
The diagnosis of chronic pancreatitis requires the demonstration of typical histologic and morphologic changes in the pancreas or evidence of decreased digestive function. Radiologic methods and rarely biopsy can demonstrate changes in the structure of the pancreas. Typically, most radiologic methods detect advanced disease. Detection of early changes has proven difficult and makes the identification of patients in the early stages of chronic pancreatitis almost impossible. Both direct and indirect measures of pancreatic function can help establish the presence of pancreatic insufficiency, but most cannot detect losses in function that do not result in maldigestion.
Acute Pancreatitis
Pathophysiology
The development of pancreatitis can be modeled into three phases. First, an event triggers the process ( Figure 82-1 ). Obstruction of the pancreatic duct from gallstones and ethanol ingestion commonly incites pancreatitis in adults. Common triggers in children include systemic illness, medications, trauma, and bile or pancreatic duct disease secondary to gallstones or congenital abnormalities. The triggering event stresses the pancreatic acinar cells through mechanisms that remain speculative.
This stress, likely metabolic or oxidative, initiates the second phase of acute pancreatitis: acinar cell injury. Ever since Chiari proposed that pancreatitis results from autodigestion by digestive enzymes, the premature activation of digestive enzymes, in particular trypsinogen, has been central to theories on the mechanism of acinar cell injury. Under this scheme, trypsinogen is prematurely activated to trypsin inside the acinar cell. To this day, the mechanism of activation remains obscure. Because an early event in experimental pancreatitis is the co-localization of digestive enzymes and lysosomal enzymes in vacuoles, many theories of pathogenesis speculate that activation of trypsinogen occurs in the vacuoles through the action of lysosomal enzymes, in particular cathepsin B. Although some studies have implicated cathepsin B in the intracellular activation of trypsinogen, other studies do not support this theory and the relevance to human disease has not been established.
Premature activation of digestive enzymes may not be the dominant event in the pathogenesis of pancreatitis. Other mechanisms may contribute to acinar cell injury. For instance, some have proposed that reactive oxygen species may produce injury by increasing lipid peroxidation, or alterations in the microcirculation of the pancreas may produce areas of hypoperfusion and subsequent hypoxic damage to acinar cells. The observed changes in cell membrane permeability may lead to pathologic increases in cellular calcium, another early event in experimental pancreatitis, or the leak of proteins from the cells. Impairments in autophagy, a cytoprotective mechanism for maintaining homeostasis of cellular functions such as organelle and protein turnover, likely contributes to the pathogenesis of cell injury. The formation of autophagosomes is an early event in experimental pancreatitis. Some have proposed that autophagy contributes to trypsinogen activation within the acinar cells. Recently, the role of the endoplasmic reticulum (ER) stress response in acute pancreatitis has been recently appreciated. One of the major triggers of the ER stress response is an imbalance in the ER protein folding capacity. Increases in the concentration of unfolded proteins in the ER sets off a series of pathways known as the unfolded protein response. Prolonged activation of the unfolded protein response leads to programmed cell death. Of interest, misfolding of human chymotrypsin C, carboxypeptidase A1, and trypsinogen mutants, which are associated with human disease, activated the ER stress response in transfected tissue culture cells, thereby implicating the ER stress response in the disease mechanism of pancreatitis.
Whatever the mechanism, acinar cell injury is followed by a local inflammatory process that sets off the final stage of pancreatitis: the inflammatory response. The transcription factor nuclear factor (NF)-κβ is central to this response. NF-κβ regulates genes encoding cytokines, chemokines, adhesion molecules, and proteins in cell death pathways among other gene products. In experimental models of pancreatitis, NF-κβ is activated in the early stages. Moreover, direct activation of NF-κβ results in pancreatic inflammation and a systemic inflammatory response. The subsequent release of cytokines and chemokines mediates a systemic inflammatory response. The clinical severity of pancreatitis depends largely on the vigor of the inflammatory response. A brisk response leads to pancreatic necrosis and inflammation of adjacent and distant organs. Clearly, the systemic complications of acute pancreatitis result from the activated immune response in distant organs and not from the circulating pancreatic digestive enzymes.
Pathology
The histology of interstitial pancreatitis shows relatively minimal changes. The prominent feature is interstitial edema. The inflammatory infiltrate is minimal and includes neutrophils, monocytes, macrophages, and lymphocytes. The leukocyte migration is triggered by chemokines released from resident macrophages, acinar cells, and, likely, stellate cells. The inflammatory response results in little cell death, mostly through apoptosis, although small areas of acinar cell necrosis may be present. In addition, focal areas of fat necrosis may be present within the gland or adjacent to the gland. The involvement of fat tissue may contribute to the local and distant complications of acute pancreatitis. In the more severe necrotizing pancreatitis, the areas of necrosis are extensive and include acinar, ductal, and islet cells. Erosion into blood vessels produces hemorrhage into the parenchyma. Areas of fat necrosis adjacent to the pancreas can also be extensive. Gross examination shows a pancreas containing areas of red–black hemorrhagic necrosis and foci of yellow–white fat necrosis.
Etiology
A wide array of factors can trigger acute pancreatitis in children ( Box 82-1 ). The prevalence of any specific etiology is not well established, since there is wide variation in reported prevalence among published studies. The variation stems from the retrospective nature of the studies, the bias or experience of the clinicians at a particular institute, the lack of thorough investigations for the etiology in many patients, and the recognition of new etiologies in the time between reports. All reports include a sizable number of patients with no identifiable etiology. Of those with an identifiable etiology, several broad categories—systemic illness, biliary disease, trauma, and medications—stand out as causing the majority.
Biliary
Cholelithiasis
Choledochal cyst
Biliary sludge
Anatomic
Pancreas divisum
Anomalous junction of the biliary and pancreatic ducts
Annular pancreas
Ampullary obstruction
Crohn’s disease
Diverticulum
Cyst
Ulcer
Tumor
Drugs
L-Asparaginase
Valproate
Metronidazole
Mercaptopurine
Azathioprine
Tetracycline
Pentamidine
Didanosine
Genetic (mutations)
PRSS1
CFTR
SPINK1
Systemic Disease
Sepsis
Hemolytic uremic syndrome
Diabetic ketoacidosis
Collagen vascular disease
Kawasaki disease
Organ transplantation
Sickle cell disease
Anorexia nervosa
Shock
Inflammatory bowel disease
Trauma
Blunt injury
Duodenal hematoma
Child abuse
Post-ERCP
Metabolic
Hyperlipidemia
Hypercalcemia
Glycogen storage disease
Organic acidemias
Malnutrition (re-feeding)
Postoperative
Spinal surgery
Cardiothoracic surgery
Autoimmune pancreatitis
Indeterminate
The reported proportion of patients with acute pancreatitis in the setting of a systemic illness has varied considerably, 3.5% to 48%, with more recent reports finding the higher rates. Although acute pancreatitis can occur with a number of different systemic illnesses, hemolytic uremic syndrome has been reported most often in association with acute pancreatitis. The mechanism of pancreatitis in this illness is unknown and probably multifactorial. Both uremia and hemolysis are risk factors for acute pancreatitis, and acute pancreatitis should be considered in every child with hemolytic uremic syndrome.
Biliary disease includes both gallstones and congenital anomalies of the biliary tree. Gallstone pancreatitis is likely more common in pediatric patients than previously believed. Solid data on the incidence are lacking, but this condition clearly occurs at all ages. Almost 30% of 56 patients in a Korean study had gallstone pancreatitis. In a study of infants and toddlers, nearly 10% had gallstone pancreatitis. Congenital anomalies include pancreas divisum, choledochal cysts, and choledochocele. Pancreas divisum is the most common anomaly, occurring in up to 10% of the population. Its role in acute pancreatitis remains controversial, and most people with pancreas divisum do not develop pancreatitis. The subgroup that develops pancreatitis may have another risk factor such as dysfunction of the cystic fibrosis transmembrane regulator (CFTR) protein or serine protease inhibitor Kazal type 1 (SPINK1).
A variety of medications is associated with acute pancreatitis. Valproic acid has the highest rate reported in a series of children with acute pancreatitis. Because the incidence of pancreatitis in children taking valproic acid is low, another factor may be involved. Consequently, the possibility of an underlying inborn error of metabolism, particularly of fatty acid metabolism, should be considered. The next most commonly associated medication is L-asparaginase.
Trauma remains a common cause of acute pancreatitis in children. Usually the trauma is blunt and accidental, but child abuse should be considered. Cutaneous findings of trauma may not be present. The injury can vary from a small hematoma to rupture of the main pancreatic duct. In the latter case, endoscopic or surgical intervention may be required.
Despite increased awareness of pancreatitis and potential etiologies, no cause is found in about one-fifth of patients. As genetic testing and better radiologic testing becomes more widely available and new etiologies are recognized, the proportion of children with indeterminate pancreatitis will decrease.
Clinical Presentation
The symptoms and signs of acute pancreatitis are nonspecific and vary with age ( Table 82-1 ). Upper abdominal pain and vomiting are the most common symptoms. Abdominal pain is present in 68% to 95% of patients in various series. Some of the variation may reflect the age distribution of patients in the studies. Only 29% of patients younger than 3 years of age had abdominal pain, and even if irritability was considered a surrogate for pain, the percentage of patients was still only 46%. Vomiting is present in 45% to 85% of reported cases. In younger patients, vomiting is the most common symptom. Less common symptoms and signs include abdominal tenderness or distension, tachycardia, fever, hypotension, jaundice, and back pain. Ecchymoses of the flank (Turner’s sign) or blue discoloration of the umbilicus (Cullen’s sign) is unusual in children. Acute pancreatitis should be considered when a hospitalized patient has worsening of clinical status or feeding intolerance.
| Common | Uncommon | |
|---|---|---|
| Symptoms |
|
|
| Signs |
|
|
Diagnosis
Because the signs and symptoms of acute pancreatitis are nonspecific, the health care providers must consider acute pancreatitis in any child with gastrointestinal complaints. The accepted clinical definition of acute pancreatitis in adults requires two of the following: (1) abdominal pain consistent with pancreatitis; (2) serum amylase or lipase levels at least three times the URL; and (3) radiologic evidence of acute pancreatitis. Although most pediatric patients with acute pancreatitis will fulfill these criteria, a number of children may not complain of abdominal pain.
Laboratory Studies
There is no gold standard for the diagnosis of acute pancreatitis; however, in practice, serum amylase and lipase levels are measured to screen for pancreatitis. The reliability of these measurements to diagnose acute pancreatitis is uncertain. Several studies have attempted to define the sensitivity and specificity of serum amylase and lipase for acute pancreatitis, but all lack a method to separately and definitively diagnose acute pancreatitis. Furthermore, both can have normal levels when there is clear clinical and radiologic evidence of acute pancreatitis, or both can have elevated levels in other conditions where there is no other evidence for acute pancreatitis. Finally, a significant fraction of patients will have a selective elevation of either amylase or lipase level at presentation. Consequently, both amylase and lipase levels should be measured in patients suspected of having acute pancreatitis.
Infants present a theoretical problem with the use of amylase and lipase level to diagnose acute pancreatitis. Expression of both pancreatic amylase and lipase are low at birth. Amylase may not reach adult levels until adolescence and lipase may not reach adult levels until 1 year of age. The clinical significance of these developmental patterns of expression is unknown, but should be noted when considering acute pancreatitis in infants.
Other conditions such as renal failure rarely increase serum amylase or lipase levels more than three times the URL. Two exceptions can increase serum enzyme levels considerably more than three times the URL. Both are benign and their recognition prevents unnecessary testing and treatment. In these patients, the elevated amylase or lipase is discovered by routine laboratory testing for non-specific symptoms such as generalized abdominal pain and the patient is mistakenly given a diagnosis of pancreatitis. See Box 82-1 . In each case, the serum enzyme elevations persist even in the absence of symptoms. First, either lipase or amylase can complex with a larger molecule, often an immunoglobulin, to form macroenzymes called macroamylase or macrolipase. Because the complex does not clear the serum as rapidly as unbound enzyme, the enzyme accumulates to higher than normal levels in the bloodstream. At present, there is a commercially available test for macroamylasemia but not for macrolipasemia. Second, some patients have elevations of pancreatic enzymes in the serum without any clinical or radiographic evidence of pancreatitis or of macroenzymes. Typically, lipase or amylase is detected but all pancreatic enzymes are elevated in most patients. The levels fluctuate from normal to many times the URL. Other family members may have benign elevations of pancreatic enzymes. The mechanism is unclear and may involve alterations in the normal secretory pathways, with shunting of more enzymes to basolateral secretion as happens in acute pancreatitis.
Imaging
No studies have compared the utility of different imaging modalities in children with suspected acute pancreatitis. In patients with abdominal pain and elevated amylase or lipase, imaging may not be necessary unless there are systemic signs of severe disease or suspicion of biliary pancreatitis. If there are clinical symptoms of acute pancreatitis and the serum enzyme levels are not greater than three times the URL, imaging may help make the diagnosis. Transabdominal ultrasonography represents a good initial choice for the radiographic evaluation of children. Ultrasonography is widely available and less expensive than other options ( Figure 82-2A ). Typical findings are pancreatic hypertrophy, decreased echogenicity, dilated pancreatic duct, and peripancreatic fluid. In addition, ultrasound can detect gallstones. Obesity or the presence of bowel gas may limit the ability of ultrasound to provide useful information.
Contrast-enhanced computed tomography (CECT) is another option. CECT is probably the best imaging modality for the overall assessment of the pancreas and complications ( Figure 82-2B ). Information about the presence of pancreatic inflammation, the presence of gallstones, possible etiologies, and complications such as pancreatic necrosis or fluid collections can be obtained. Animal studies raised concern that intravenous contrast may increase the severity of pancreatitis, but this has not held for humans.
The prevalence of gallstone pancreatitis in childhood, particularly adolescence, provides the best argument for imaging studies early in the course of acute pancreatitis. Clearly, elevations of serum transaminase or bilirubin levels should prompt investigations for gallstone pancreatitis. Although ultrasound and CECT can detect gallstones, studies in adults suggest that endoscopic ultrasound (EUS) or magnetic resonance cholangiopancreatography (MRCP) may be better options ( Figure 82-3 ). No similar studies have been done in pediatrics.
In addition to gallstones, MRCP can reveal other pancreatobiliary disorders. Choledochal cysts, pancreas divisum, and anomalous junction between the pancreatic and biliary ducts can be identified. MRCP performs well in children of all ages and does not require ionizing radiation. The major limitation of MRCP is the need for sedation or general anesthesia in many pediatric patients. In most centers, MRCP has replaced endoscopic retrograde cholangiopancreatography (ERCP) in the diagnosis of pancreatobiliary disease. ERCP is now used primarily for therapy of gallstones or duct anomalies.
In adults, EUS increases the diagnostic yield for gallstones and microlithiasis. The size of EUS endoscopes and the reluctance of trained EUS endoscopists, primarily adult gastroenterologists, to perform procedures on young patients have limited EUS in pediatric patients. Still, experience with EUS in children is accumulating. As happened with ERCP, smaller endoscopes will become available and the comfort of the endoscopists with pediatric patients will increase. Eventually the role for EUS in children with pancreatitis will be defined.
Investigations of Etiology
Clinical judgment should determine which testing is necessary to identify the etiology for a patient’s first episode of acute pancreatitis. Systematic consideration of probable, possible, and rare causes based on history and physical examination findings should limit unnecessary testing. For instance, patients who develop acute pancreatitis associated with a systemic illness may not require additional studies. History can identify drugs that may cause pancreatitis or suggest trauma as the cause. Contrast-enhanced computed tomography (CECT), endoscopic retrograde cholangiopancreatography (ERCP), or magnetic resonance cholangio-pancreatography (MRCP) can provide evidence of duct rupture ( Figure 82-4 ). A family history of pancreatitis should prompt genetic testing. Jaundice should prompt an investigation into biliary causes of pancreatitis. Serum calcium and triglycerides should be measured in all patients. Although hypercalcemia and hypertriglyceridemia are uncommon in pediatrics, they can be treated, and future episodes of acute pancreatitis can be prevented.
In the 10% of patients who have a second episode of acute pancreatitis, a more thorough investigation for etiology is necessary. A complete evaluation for structural anomalies of the pancreatobiliary tree and upper gastrointestinal tract should be pursued. MRCP is the choice for ductal abnormalities in children of all ages. CECT or an upper gastrointestinal series can identify duplication cysts of the stomach or duodenum and congenital cysts in the pancreas. Duodenal ulcers, duodenal Crohn’s disease, and tumors of the papilla can be identified with use of esophagogastroduodenoscopy. Genetic testing for genes associated with recurrent pancreatitis should be considered (see Chronic Pancreatitis-Gene Testing below).
Management
Treatment, as it always has, relies on supportive care to provide adequate intravenous hydration, to control pain, and to monitor for complications. Pancreatic rest or nothing by mouth is still practiced but should be limited.
Fluid Management
In recent years, the importance of fluid resuscitation at presentation of acute pancreatitis has undergone much discussion. Animal models and retrospective studies in humans indicate that early fluid resuscitation may improve outcome and prevent severe disease, and aggressive intravenous fluid replacement was recommended in the American College of Gastroenterology practice guidelines for acute pancreatitis without defining what aggressive means. Gradually replacing fluid deficits (5 to 10 mL/kg per hour) may give better outcomes than more rapid fluid infusions (10 to 15 mL/kg per hour). The use of lactated Ringer’s solution may improve outcome compared to other solutions in adults. It has been suggested that monitoring the blood urea nitrogen (BUN) during fluid resuscitation improves outcomes in adults. There are no studies of pediatric patients to guide fluid management. Without clear guidelines, it seems reasonable to provide intravenous fluids at a rate that exceeds basal requirements and to give additional fluids determined by the patients’ hemodynamic status.
Pain Control
Patients with pain should be treated with parenteral narcotics. Some centers avoid morphine in the belief that morphine may increase sphincter of Oddi pressure and worsen the course of pancreatitis. Frequently, meperidine is given. In truth, all opiates, including meperidine, can increase sphincter pressure, and no clinical evidence indicates poorer outcome when morphine is used in acute pancreatitis. Morphine has the advantage over meperidine in that it has a longer half-life and fewer side effects. Regardless of the choice for analgesia, adequate pain relief should be provided. Dosing of pain medications should be monitored regularly and the daily dose adjusted as required.
Nutritional Therapy
Pancreatic rest, starvation, has long been a mainstay of care in patients with acute pancreatitis. Now, experimental and clinical evidence strongly supports early feeding of patients with pancreatitis, and nutritional support is considered an active and beneficial intervention. Enteral nutrition should start within 24 hours of admission after fluid resuscitation and pain control are implemented.
Patients with mild acute pancreatitis can be safely allowed to eat and drink as tolerated without a period of fasting. In this study, patients who were fed advanced to solid foods faster and had a shorter hospital stay than fasted, control patients. In another study, patients with mild acute pancreatitis were randomized to a low-fat diet or a clear liquid diet. The reoccurrence of pain and the need to stop oral intake did not differ between the groups. Because patients with mild acute pancreatitis generally tolerate oral intake early in their course, fasting until symptoms resolve or until serum amylase or lipase return to normal is not necessary.
Patients with severe pancreatitis can also be fed shortly after they have been stabilized. In fact, early enteral feeds may provide more benefit and cause fewer complications than parenteral nutrition. Various studies support the safety and tolerance of both nasogastric and nasojejunal feeds. Delivery of nutrients beyond the ligament of Treitz has theoretical advantages over gastric feeds. It avoids the stomach in a group of patients with a high prevalence of abnormal gastric emptying, and modern jejunal feeding tubes allow simultaneous decompression of the stomach. In addition, jejunal feeding causes less stimulation of the exocrine pancreas. Presently, the choice of route is institution specific.
The final consideration is what type of diet to provide. Standard practice in most centers is to start with a low-fat diet or formula although there is no evidence that a low-fat diet offers any advantage over a regular diet. Immunomodulating diets supplemented with glutamine, arginine, omega-3 fatty acids, or antioxidants are controversial in critically ill patients. Their use in severe acute pancreatitis is not indicated currently.
Outcome
In most cases, children and adolescents have a mild course of acute pancreatitis, and their symptoms resolve without complications. However, the average length of hospital stay can exceed 2 weeks. The mean length of hospital stay before 1999 was 13.2 ± 2.4 days. Recently, two studies reported 24 and 25.7 days as the mean length of stay. In one, the median hospital stay was 8 days, and the authors suggested that the presence of comorbid conditions strongly influenced the length of stay. Another study focusing on infants and toddlers supports that contention. The median length of stay for all subjects was 19.5 days, whereas patients admitted with only acute pancreatitis had a median length of stay of 8.5 days. Overall, about one-fifth of children will have prolonged courses, with persistent symptoms or complications. Many of these patients may have another illness when they develop acute pancreatitis, and it is likely that these conditions rather than pancreatitis cause the more significant complications like shock or renal failure.
Children have the same local and systemic complications of acute pancreatitis that occur in adults ( Table 82-2 ). Peripancreatic fluid collections and pseudocysts are reported the most often and can be found in about 15% of patients ( Figure 82-5 ). Age may influence the rate of fluid collection, since only one of 79 patients younger than age 3 developed a fluid collection. Limited data suggest that most fluid collections resolve without therapy, although some cause persistent symptoms and require intervention. Surgical, endoscopic, and interventional radiologic approaches have all been employed in children. Some children develop pancreatic necrosis. Reliable data about the rate are not available, but pancreatic necrosis appears to be uncommon, with only one case of 380 subjects from seven centers. The diagnosis is made when CECT shows a segment of the pancreas without perfusion ( Figure 82-2B ). Infected necrosis, a rare event in children, should be suspected if abdominal pain worsens, leukocytosis develops, or a fever appears. These patients should have antibiotics added to their care. Typically, the necrosis will liquefy and wall off with conservative therapy and surgical debridement is rarely indicated. Deaths do occur in children and rates from 2% to 11% have been reported. Two series found that all deaths occurred in patients with complications of systemic illness and not from complications clearly related to acute pancreatitis. Most likely, death from complications of acute pancreatitis is uncommon in children, but better data are required to definitively determine the death rate.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
