Laboratory Diagnosis of Gastrointestinal and Pancreatic Disorders

Cystic Fibrosis (CF)

CF is the most common genetic disorder in white North Americans and is often fatal in childhood. Some Native American tribes (Pueblo) have a similar incidence. It is also frequent in Hispanics but uncommon in Asians and African Americans. More than 25,000 Americans have CF at this writing, with almost 1000 new cases expected to be diagnosed every year. The incidence is 1 in 1600 white births and 1 in 17,000 African American births in the United States. With the advent of more widespread prenatal screening, the incidence is declining. While much less frequent, CF still is the most common cause of steatorrhea in African American children.

CF of the pancreas is an autosomal-recessive disease of chloride ion transport. It is caused by mutations in the CF transmembrane conductance regulator (CFTR) gene on chromosome 7, which encodes for an epithelial chloride channel protein. More than 2000 CFTR gene mutations with different associations with the disease have been identified; 15% of these are not associated with CF. The available genetic testing can identify 70 mutations that account for >90% of the CF cases. When genetic screening fails to identify a mutation in affected patients, whole exome sequencing can be performed ( ).

Newborn screening for CF is now routinely performed in many states. The initial test, which has high sensitivity and poor specificity, determines immunoreactive trypsinogen; gene sequencing is then performed for confirmation.

CF is characterized by abnormally viscous mucous secretions from the various exocrine glands of the body, including the pancreas and the salivary, peritracheal, peribronchial, and sweat glands. Involvement of the intestinal glands may result in the presence of meconium ileus at birth. Two-thirds are diagnosed before 1 year of age. Chronic lung disease and malabsorption resulting from pancreatic insufficiency are the major clinical problems of those who survive beyond infancy, but intelligence and cognitive functions are unaffected and are normal ( ; ).

The degree of the defect depends on the nature of the mutation, and well-defined genotype-phenotype associations are now known. There are several characterized mutations that lead to a milder form of the disease. The classic ΔF508 (or Phe508del) mutation leads to cystic fibrosis when two copies of the gene are inherited. Persons heterozygous for the R117H (or Arg117His) mutation may develop pancreatic insufficiency due to plugging of ducts, causing idiopathic chronic pancreatitis ( ; ).

Heterozygotes have no recognizable clinical symptoms. Most homozygotes fully express the syndrome of recurrent pulmonary infections, pancreatic insufficiency, steatorrhea, and malnutrition. These manifestations are caused by abnormally dehydrated tenacious secretions of all exocrine glands. The viscid inspissated mucus plugs ducts and causes chronic inflammation with atrophy of acini, fibrosis, dilatation, and cystic duct changes.

Pancreatic abnormalities occur in >80%. Islets of Langerhans are initially spared, but diabetes often develops during adolescence. The pancreatic lipase deficiency causes maldigestion of fat and steatorrhea. Thick intestinal mucus may cause intestinal obstruction in the neonate due to meconium ileus. Most men with CF are infertile, with azoospermia due to duct obstruction. Pulmonary changes are the most serious in CF. No cure is available. Median survival has increased in the past 25 years from 18 to 36 years of age due to advances in treatment. A total of 90% die from pulmonary complications.

In a child with suspected CF, the demonstration of increased chloride in the sweat is a useful initial test in the workup. More than 99% of children with CF have concentrations of sweat chloride greater than 60 mmol/L. The sweat chloride may not be as dramatically increased in adolescent or adult patients. The test needs be performed with care ( ).

Pilocarpine is introduced into the skin by iontophoresis to stimulate locally increased sweat gland secretion. The resulting sweat is absorbed by filter paper or gauze, weighed, diluted with water, and analyzed for sodium and chloride concentrations. High rates of incorrect results have been attributed to problems associated with sweat specimen sample collection and test analysis ( ; ). In children, chloride concentrations of over 60 mmol/L in sweat on at least two occasions are diagnostic, with a sensitivity of 90% to 99%. Levels of between 50 and 60 mmol/L are suggestive of CF in the absence of adrenal insufficiency.

Patients with indeterminate sweat chloride results in whom CF is suspected may undergo confirmatory testing following administration of a mineralocorticoid such as fludrocortisone. In those patients with CF, the electrolyte values would remain unchanged, whereas normal controls would show a decrease in sweat electrolytes. Sodium concentrations in sweat tend to be slightly lower than those of chloride in patients with CF, but the reverse is true in normal subjects. Alternatively, genetic mutation analysis of the CFTR gene can be performed.

False-positive elevations of sweat chloride concentrations of >60 mmol/L may be seen in malnutrition, hyperhidrotic ectodermal dysplasia, nephrogenic diabetes insipidus, renal insufficiency, glucose-6-phosphatase deficiency, hypothyroidism, mucopolysaccharidosis, and fucosidosis. These disorders are usually easily differentiated from CF by their clinical symptoms. False-negative sweat test results have been seen in patients with CF in the presence of hypoproteinemic edema.

Sweat electrolytes in about half of a group of premenopausal adult women were shown to undergo cyclic fluctuation, reaching a peak chloride concentration most commonly 5 to 10 days prior to the onset of menses. Peak values were slightly under 65 mmol/L. Men showed random fluctuations up to 70 mEq/L. For this reason, interpretation of sweat electrolyte values in adults must be approached with caution ( ; ).

Hemochromatosis

Excessive body iron accumulation from any source is directly toxic to cells and causes fibrosis. The symptoms include the triad of bronze coloration of the skin, cirrhosis, and diabetes, among others. Humans have no major iron excretory pathway. The screening test consists of transferrin saturation (TS) = serum iron/total iron binding capacity × 100. Interpretation: Abnormal, >60% in women and >50% in men. Confirm with fasting TS and ferritin levels. Genetic testing and liver biopsy with assay for iron is used to confirm and assess extent of tissue iron load ( ; ). Recently, MRI has been utilized to quantify liver iron overload and to monitor response to chelation therapy ( ).

Primary or hereditary hemochromatosis (HH) is a human leukocyte antigen (HLA)-linked autosomal recessive defect in duodenal iron absorption regulation. Mutations (C282Y, H63D) in the HFE gene, on the short arm of chromosome 6, are responsible for >80% of HH cases (see Chapter 22, Chapter 72 ). The disease is most common in white males; the homozygosity frequency is 1 in 220. When HH is diagnosed, other family members should be screened; one quarter of siblings will be positive ( ; ).

Secondary hemochromatosis is typically seen in congenital or hemolytic anemias that require multiple blood transfusions, or when oral iron intake is increased, resulting in excess iron storage.

Early diagnosis and chelation therapy and/or phlebotomies are effective in preventing tissue damage. In disease, the pancreas is slightly enlarged and deep brown due to accumulated hemosiderin, the iron containing pigment. When untreated, progressive fibrosis of the pancreas with atrophy occurs. The iron is deposited in the acinar and duct cells and in the β-cells of the islets. Other cells of the islets appear spared. Similar pigments are in the skin. The β-cell loss results in bronze diabetes. Hypogonadism with pituitary dysfunction is present in half of cases and cardiomegaly and osteoarthritis is present in most cases. Cirrhosis is seen in 70% of the cases. Hepatocellular carcinoma occurs in 30% of cases; this tumor has become a chief cause of death in HH ( ; ).

The American Association for the Study of Liver Diseases has recommended liver biopsy and a quantitative serum iron assay for (1) all homozygotes with clinical evidence of liver disease; (2) all homozygotes with serum ferritin greater than 1000 ng/ml; (3) all homozygotes older than 40 years with other risk factors for liver disease; and (4) compound or C282Y heterozygotes with elevated transferrin saturation, particularly those who have had abnormal liver enzyme levels or clinical evidence of liver disease ( ).

Acute Pancreatitis

Acute reversible inflammation is due to necrosis of the pancreatic tissue by proteolytic digestive enzymes that are inappropriately activated. It can occur at any age, most commonly between 30 to 70 years, but is also noted in children. Its diagnosis is based on compatible clinical features, including abdominal pain, nausea, and vomiting. The clinical suspicion is supported by finding elevations of the serum amylase and/or lipase ( Table 23.1 ). The pancreas contributes 40% of total serum amylase; the rest is mostly from the salivary glands ( ).

In 30% of patients, the diagnosis of acute pancreatitis was not suspected and was only made at autopsy ( ). There are many causes. Gallstones continue to be the leading cause (30%–60%). Alcohol is the second most common cause, responsible for 15% to 30%. Other causes include duct obstruction due to tumors, duct anomalies such as pancreas divisum, infections (mumps, coxsackievirus A, parasites), blunt trauma or post–endoscopic retrograde cholangiopancreatography (ERCP), many drugs (diuretics, sulfonamides), organophosphates, methyl alcohol, nitrosamines, hypertriglyceridemia, and hypercalcemia ( ; ).

Hemorrhagic pancreatitis, a severe form of acute pancreatitis, results from necrosis within and around the pancreas with hemorrhage, which may cause shock and death. Initially, necrosis is coagulative, but the necrotic cells rapidly undergo liquefaction. Biliary tract disease with gallstones or inflammation of the gallbladder or bile ducts, or alcoholism, are present in about 80% of the patients. The male:female ratio is 1:3 in acute pancreatitis associated with biliary tract disease and 6:1 in alcoholism. Pancreatic microlithiasis may be responsible for many of the other cases.

The sequence of changes following release of activated intrapancreatic enzymes in acute pancreatitis are microvascular leakage causing edema, necrosis of fats, and acute inflammatory reaction. Proteolytic destruction of pancreatic tissue and blood vessels causes edema and focal dilatation of acini with variable amounts of hemorrhage. In fat necrosis, neutral fats are broken down, glycerol is reabsorbed, and the fatty acids combine with calcium salts to form soaps (saponification) with a zone of acute inflammation around the foci of necrosis. After a few days, secondary infection with suppuration and abscesses may occur.

In 15% to 30% of those with pancreatic necrosis, poorly defined areas of acute fluid collection occur, with fibrosis. The liquefied areas are walled off and pseudocysts form. A pseudocyst contains pancreatic fluid enclosed in fibrous tissue with no epithelial lining; they often communicate with a pancreatic duct and continue to increase in mass.

Computed tomography (CT) is the most useful test to establish diagnosis, with characteristic radiologic findings of enlarged edematous and inflamed pancreas with or without surrounding fluid collection, with or without necrosis. An ultrasonogram may be useful in showing a diffusely enlarged, hypoechoic pancreas, and may show presence of gallstones in the gallbladder, indicating a possible etiology. A CT severity score (the Balthazar score) is based on the degree of necrosis, inflammation, and the fluid collections. There is a 23% mortality rate associated with any degree of pancreatic necrosis plus a strong association between necrosis and morbidity and mortality. After initial assessment, a CT scan need not be repeated unless one suspects development of a complication such as pancreatic necrosis. Magnetic resonance imaging (MRI) is being increasingly used to detect pancreatitis and to characterize the “pancreatic necrosis” seen on CT into peripancreatic necrotic fluid collection, necrotic pancreatic parenchyma, and hemorrhagic foci. MRI can also detect pancreatic duct disruption, seen early in the course of acute pancreatitis. Endoscopic ultrasound is now recognized as an invaluable tool to diagnose and drain pseudocysts.

Predictors of severe acute pancreatitis are a hematocrit >44% with a failure to decrease at 24 hours, which is indicative of pancreatic necrosis and organ failure, and a C-reactive protein (CRP) >150 mg/L. Serum creatinine >2.0 mg/dL or marked hyperglycemia >150 mg/dL are predictive of mortality ( ). There is a strong association between the extent of urea nitrogen increase and mortality at 24 hours. Each increase in urea nitrogen of 5 mg/dL was associated with a corresponding increase in mortality, and a decrease was associated with significantly improved survival ( ; ) ( Table 23.2 ).

Chronic Pancreatitis

It is the irreversible damage, and often progressive inflammation with irregular fibrosis, duct dilatation and loss of pancreatic parenchyma, that characterize chronic pancreatitis. It occurs after repeated bouts of acute pancreatitis, obstruction of the pancreatic duct by mechanical blockage or congenital defect, by neoplasm, gallstone duct obstruction, or alcoholism. Early in the course, the pancreas is enlarged. Subsequently, due to scarring, the gland usually shrinks, with loss of acini and still later a loss of ductules. Preserved or even increased islets are seen in the fibrous scar. Patients seek medical attention because of abdominal pain or maldigestion.

Maldigestion/malabsorption and steatorrhea are due to pancreatic insufficiency with loss of enzymes, glucose intolerance or diabetes, and islet damage. Low fecal elastase levels correlate well with disease. Clinically, there is recurrent or chronic pain. Since only a small number of patients with acute pancreatitis go on to develop chronic pancreatitis, the incidence of the latter is lower than that of acute pancreatitis but is increasing. Incidence is greater in males than in females, and the average age of onset is 40 years. It is more prevalent in tropical countries; the main form is chronic calcifying pancreatitis with duct calcifications, which can often be seen on plain films of the abdomen. In temperate areas, chronic alcoholism is found in more than half of the cases. Etiology is not apparent in 40% of the cases.

The central enzyme involved in activation of all the digestive proenzymes is trypsin. Trypsin is synthesized and maintained as inactive trypsinogen in secretory granules in the pancreatic acinar cell. After release through the pancreatic duct into the intestinal lumen, trypsinogen is cleaved by enterokinase on the brush border of the duodenum to yield active trypsin. Trypsin is stabilized in the pancreatic acini by a serine protease inhibitor, SPINK1. Mutations in SPINK1 increase the risk of chronic pancreatitis almost 12-fold by impairing the ability of acinar cells to counteract and inhibit the damaging effects of intracellular trypsin ( ; ). Cationic trypsinogen (PRSS1) mutations involving codon 29 and 122 cause autosomal-dominant forms of hereditary pancreatitis ( ; ).

Pepsin and Pepsinogen

Pepsinogens are the biologically inactive proenzymes of pepsins that are produced by the chief cells and other cells in the gastric mucosa. They are found in two distinct types, pepsinogen I (PGI), also known as pepsinogen A , and pepsinogen II (PGII), also known as pepsinogen C . Pepsinogen secretion is stimulated by the vagus nerve, gastrin, secretin, and CCK. It is inhibited by gastric inhibitory peptide (GIP), anticholinergics, H ₂ -receptor antagonists, and vagotomy ( ). PGI is produced in the chief cells and mucous cells of oxyntic glands; PGII is produced in mucous cells in oxyntic and pyloric regions and the duodenum. The ratio of concentration of PGI to PGII in serum or plasma of healthy individuals is approximately 4:1 ( ). Pepsinogen is converted to the active form, pepsin, by gastric acid that can activate additional pepsinogen autocatalytically. Both groups of pepsinogen are activated at an acid pH below 5 and destroyed by alkaline pH. Both types can be detected in blood. Only type I pepsinogens are present in the urine. Pepsins are responsible for the hydrolysis of proteins to polypeptides. The pepsinogen released from the gastric mucosa constitutes a major component of gastric fluid. Only approximately 1% gets into the peripheral blood. Active pepsin is rapidly inactivated in the bloodstream, whereas pepsinogen is stable in the blood. Pepsinogen is then filtered by the kidneys and is excreted in the urine, where the slightly acidic pH converts the pepsinogen, now called uropepsinogen to uropepsin ( ). Immunoassay is the method used to detect serum pepsinogen. However, the PGI isoform is commonly analyzed in the clinical laboratory since it is the isoform commonly associated with disease.

Serum levels of PGI are an accurate estimate of parietal cell mass and correlate with acid-secretory capacity of the stomach. Increased pepsinogen levels and associated activity is observed in patients with disease states that lead to increased gastric output or with increased parietal cell mass, namely, gastrinomas, ZE syndrome, duodenal ulcer disease, and acute and chronic gastritis. Decreased levels of pepsinogen are associated with decreased parietal cell mass, atrophic gastritis and gastric carcinoma, as well as in patients with myxedema, Addison disease and hypopituitarism ( ). The PGI/PGII ratio decreases linearly with worsening atrophic gastritis. Absence of pepsinogen is noted in patients with achlorhydria. PGI levels measured by immunoassay usually range from 20 to 107 μg/L and PGII levels usually range from 3 to 19 μg/L.

Pepsinogen assays are being explored for their utility in the noninvasive identification of patients with chronic atrophic gastritis and to provide an estimate of the extent of atrophic gastritis, a known precursor of gastric carcinoma. Severe atrophic body gastritis causes a four- to fivefold increase in the risk of gastric carcinoma compared with healthy individuals ( ). This will hopefully identify a subgroup of individuals with chronic atrophic gastritis that would benefit from endoscopic evaluations for detection of early-stage gastric tumors. These assays are currently utilized in Japan, an area marked by high prevalence of gastric cancer, as a potential method for widespread screening of high-risk individuals ( ). Study authors recommended that criteria for diagnosing chronic atrophic gastritis be persons with PGI <70 μg/L and PGI/PGII ratio <3.0. In Japan, the pepsinogen serum screening test has been demonstrated to detect a higher percentage of early cancers compared with conventional methods; a considerable number of patients have subsequently been candidates for treatments with endoscopic surgery ( ). The most sensitive test for fundic atrophic gastritis is considered to be the PGI/II serum ratio, with 99% sensitivity and 94% specificity ( ). Furthermore, PGII levels may be a useful marker of prognosis, serving as an independent predictor of tumor biology and survival in patients with gastric carcinoma. The absence of PGII production has been associated with aggressive tumor behavior and shorter overall survival in gastric cancer patients ( ). Pepsinogen assays may therefore prove to be a useful serum screening method for detection of gastric carcinoma among high-risk individuals.

Diarrhea

About 8 to 10 L of fluid enters the duodenum every 24 hours. Much of this fluid is absorbed in the small intestine and about 1.5 L enters the large intestine. However only 100 to 150 mL is voided in stools. Increased fluid secretion or decreased fluid absorption in either the small or large intestine, or increased intestinal motility, results in diarrhea. One of the most important parameters defining diarrhea in an individual patient is a change in the usual bowel habit to more frequent looser stools. Diarrhea is the passage of three or more loose or liquid stools per day or more frequently than is normal for the individual ( ). Decrease in fecal consistency (i.e., increased fluidity) is difficult to measure; thus, increased stool weight, frequency, and duration are used in defining diarrhea.

The diagnosis of diarrhea starts with a thorough history to characterize the condition. Is the diarrhea bland or bloody (dysentery)? Are there constitutional symptoms? What is the duration of the illness? Important issues to determine are travel history, the use of antibiotics and other medications, ingestion of uncooked fish or meats, exposure to animals, and previous history of diarrhea and family history of gastrointestinal illness. Self-limited, acute diarrhea (<2 weeks in duration) without bleeding or constitutional symptoms rarely requires diagnostic testing. Chronic diarrhea, the passage of blood, and constitutional symptoms all suggest the need for a specific diagnosis. History is the key to developing the differential diagnosis and guiding the laboratory evaluation ( Table 23.4 ). The physical examination, although usually less helpful than the history, must still be comprehensive.

Acute diarrheas, defined as <4 weeks’ duration, generally have an infectious etiology. Those with a course of >4 weeks are considered to be chronic diarrheas ( ) and are categorized as osmotic, secretory, or inflammatory. Hypermotility or shortened gut may reduce the transit time and absorptive surface, resulting in diarrhea and/or some degree of malabsorption.

Osmotic diarrhea results from unabsorbable or poorly absorbed solutes. These include polyethylene glycol in colon-cleansing solutions, magnesium salts (magnesium citrate in cathartics, magnesium hydroxide in some antacids), sorbitol in chewing gum, lactulose in the treatment of hepatic encephalopathy, and lactose in lactase-deficient individuals. These osmotically active substances in the lumen alter the osmotic gradient that normally favors Na+ absorption, drawing fluid into the lumen.

Fasting or cessation of consumption of the suspected solute for 24 hours stops osmotic diarrhea. Microbial fermentation of malabsorbed carbohydrates yields short-chain fatty acids. The increased osmolarity draws additional water into the colon and lowers the stool pH to values of <5.6 ( ). Sodium and potassium concentrations in stool water are measured to calculate the fecal osmotic gap, which estimates the contribution of electrolytes and nonelectrolytes to water retention in the gut lumen. The osmotic gap is useful in differentiating osmotic from secretory diarrhea; it is best calculated as: 290 − 2 × (fecal Na ⁺ + fecal K ⁺ ), where 290 represents the stool osmolality that approximates plasma osmolality, and a factor of 2 is used to account for associated anions. Osmotic gaps of >125 mOsm/kg characterize osmotic diarrhea and those of <50 mOsm/kg are seen in secretory diarrhea ( ).

Stool osmolality must be measured on a freshly collected specimen. Because of continued degradation of stool carbohydrates, osmolality of the specimen increases over hours. Continued diarrhea during a 48-hour fast is suggestive of a secretory process, although fecal weight may decrease because of increasing dehydration ( ; ).